584,808 active members*
5,175 visitors online*
Register for free
Login
IndustryArena Forum > OpenSource CNC Design Center > Open Source Controller Boards > Open source low cost servomotor controller
Page 6 of 8 45678
Results 101 to 120 of 141
  1. #101
    Join Date
    May 2016
    Posts
    41

    Re: Open source low cost servomotor controller

    Quote Originally Posted by macdowswe View Post
    Yeah they are open. I was thinking of making a shroud that houses a motor/encoder assembly, and that has an air inlet with an intake filter, and a fan for forced air cooling. This can probably increase the performance at low speeds fairly significantly since the motors own air cooling doesn't work unless they are spinning at high speed.
    So about the cooling and thermal management of the motor, the above is the plan. Actually, just to enumerate some of the possible ways to prevent overheating:

    1. Use reduction gearing to reduce torque required from motor
    2. Oversize the motor
    3. Use forced air cooling
    4. Active thermal management/control, with motor temperature feedback (using a thermistor or similar)
    5. Most importantly: lots of testing.

    I think in the end there is a large range of motor sizes available, and the power to cost ratio of these motors is really good. So getting the right motor shouldn't be a big problem. After I get back to a proper workstation after the holidays, I will make a spreadsheet to help people pick motors for their application, and include some motors I recommend (and of course have it so people can add their own motors).

  2. #102
    Join Date
    May 2016
    Posts
    41

    Re: Open source low cost servomotor controller

    By the way guys, check this out! ODrive v3.1 being manufactured ;D


  3. #103
    Join Date
    Jul 2012
    Posts
    106

    Re: Open source low cost servomotor controller

    Quote Originally Posted by 109jb View Post
    So the RC motors being discussed are about 1kW, but at 7000 rpm, so at 3500 rpm they are going to be 500W. Simple math tells us this. If you want to de-rate them to 500W at 3500 rpm there is nothing to say you can't. It would be easier on the motors. They are also very compact, so I would consider them a good choice.

    What is the talk about belt reduction drives not being a good choice and causing lost precision, belt stretch, etc? Timing belt reductions have been used on these kind of applications for a long time and they have proven reliability and positioning precision is not an issue in a proper belt reduction system. With a 3600 rpm max at the motor, with 3mm pitch GT3 belts you could use a 15mm wide belt and it would handle about 600W of power. 15mm wide, not 55mm, and not HTD. At higher RPM you can still use the 15mm wide GT3, you just are a little limited on the small pulley diameter. GT3 belts handle more power and have a profile that is essentially zero backlash and much quieter than an HTD tooth profile. Nobody ever said anything about plastic timing belt pulleys either. Aluminum timing pulley are the norm and are readily available. The whole notion of belt stretch affecting precision to any degree is a bunch of hogwash.
    I think you need to re-examine your statement about stretch and precision unless you are moving a table with almost no weight or moving it so slowly that it's weight doesn't interfere.

    You can be damn sure that many DIY builders have employed glass reinforced nylon pulleys due to cost so to state emphatically that all machines have aluminum or steel pulleys and is the norm is nothing shy of unsubstantiated, advised and recommended I can believe but reality has shown me differently.

    I do diamond cutter-tip scribing at 150-200 IPM on a rack and pinion router throwing a 176lb gantry around, you want me to believe that a GT2 or GT3 belt wont snap like a rubber band, good luck with that logic.

    So you also don't think slinging a 60lb or more table at >200IPM then abruptly stopping it wont bounce/stretch a GT2 or GT3 belt?

    I've checked some of my tables, my small 16.535in x 9.055in x 1.9685in (420mm x 230mm x 50mm) mill table weighs 68lbs, my rapids are 590IPM, when the machine was not bolted to a pedestal it would slide across the floor during rapids and the machine weighs over 600lbs, using a GT2/GT3 belt would be an exercise in stupidity, but, if you think it wont stretch or bounce knock yourself out and let us know how that works for you moving something that has any weight.

    I have some larger tables upwards of 135lbs (24in x 16in x 2.5in) with 590IPM rapids and 100% table travel, there is no way I would put anything less than HTD8 as a drive belt or anything less than 1.5KW servo motor to move it.

    Now ignoring the issues with inertia alone, if you don't program your mach3 or linuxcnc to have a fairly mild acceleration/deceleration curve you're asking for trouble because if it's aggressive you can kiss that belt good-bye quickly.

    The motor in the link sees a lot more than 7,000RPM but at 6,000RPM (2:1 pulley reduction = 3000RPM) the inertia of a 60lb table would still cause belt bounce/stretch and affect precision with GT2/3 x 15mm, now you want to reduce the speed from well over 15KRPM to 3.5KRPM and expect decent bang for your buck, that would be very unlikely, it's no longer cost effective unless you have the motors in hand, access to a mill and lathe to redesign the motor body, add forced air cooling and a cheap encoder and it's still going to be expensive.

    Maybe it's just me but I expect a machine to operate just like it's bigger brothers so mechanical limitations by design need to be minimized as much as possible and not be a restriction, milling off dovetails and installing linear rails is an inexpensive hardware modification but the cost of rail can be cheap to expensive depending on your taste but they will heavily impact motion due to reduced drag so anyone thinking that a GT2/GT3 belt wont stretch or that precision wont be affected has no clue about acceleration, deceleration and inertia or only plays with toys.

    Quote Originally Posted by macdowswe View Post
    Yeah I thought about it. I haven't really seen any low cost encoders that have differential signalling (without adding a differential tx yourself ofc). If more people request it, I can put it on, but I am very conscious about keeping the part count to the minimum for cost reasons. If your application requires interfacing to differential encoder signals, you can make a small breakout for that chip and add it right before the single ended inputs to the ODrive.
    I haven't seen any low cost encoders worth a dam in a servo application where high RPM is a basic fundamental operating principal and your electrical design will still utilizes single ended encoder lines to the MCU but with buffering, protection and error detection giving greater flexibility to the type of encoder that is supported.

    The cost is minimal for adding this IC and the few components but if you're concern is chip count over functionality, please recommend that all users use the motor and encoder you specify to ensure they have no compatibility or connection issues and provide the plans to convert the motor as needed.

    Because not everyone will be using the motor and encoder you use, if you're not interested in the flexibility it offers, don't worry about it, I'll wait for you to finish the design and then I'll pay someone to generate eagle files for me and have the chip added for my own use, the thought of using expensive motors with cheap crappy encoders isn't very appealing since I can make BLDC servo motors with industry standard encoders for less or buy a small quantity of them when my friend does a run than the cost of buying those 6364 or 6374 out-runners not including the cost of converting and encoder.

    The flexibility of encoder connection just by employing the chip makes logical sense and this isn't taking into account the encoder error detection as an added bonus just by adding the chip.

  4. #104
    Join Date
    Jan 2005
    Posts
    1943

    Re: Open source low cost servomotor controller

    I think you need to re-examine your statement about stretch and precision unless you are moving a table with almost no weight or moving it so slowly that it's weight doesn't interfere.
    No I don't think I need to reconsider anything. There are many many machines built that have used belts for reduction drives on their axes, and have done so successfully with precision that is more than adequate for the intended applications. To say otherwise totally ignores those successful applications. As a matter of fact there are machines using narrow XL belt profiles on RF-45 size machines running at +200 IPM speeds and doing so successfully, but the GT2/3 is a better choice in my opinion.

    You can be damn sure that many DIY builders have employed glass reinforced nylon pulleys due to cost so to state emphatically that all machines have aluminum or steel pulleys and is the norm is nothing shy of unsubstantiated, advised and recommended I can believe but reality has shown me differently.
    You can be damn sure thatnobody stated "emphatically" that all machines use aluminum or steel pulleys? Show me where that was said. It wasn't and any claim that it was is just BS. What was said was that aluminum was the norm and I stand by that. Sure some have used non-metal sprockets. However, by and large, the build threads on this forum show that the most commonly used are the aluminum pulleys. You can argue that all you want and you would be wrong. While speaking of non-metal pulleys, you are apparently claiming that they are so rubbery or so prone to exploding that they are not appropriate for using in a precision application. This has no basis in reality. When used appropriately they can be every bit as precise, accurate and reliable as metal. To say otherwise is wrong and totally unsubstantiated.

    I do diamond cutter-tip scribing at 150-200 IPM on a rack and pinion router throwing a 176lb gantry around, you want me to believe that a GT2 or GT3 belt wont snap like a rubber band, good luck with that logic.
    You either mi-interpreted my post, or you are intentionally twisting it to fit your argument. Either way, your argument has nothing to do with what I was referring to. I was referring to a belt reduction drive, not an entire belt drive axis. This was clearly stated and you have clearly ignored it.

    So you also don't think slinging a 60lb or more table at >200IPM then abruptly stopping it wont bounce/stretch a GT2 or GT3 belt?
    No I don't. When talking about the kind of belt drive I was referring to, which is a GT2/3 driving a ball screw or any other screw drive axis it won’t stretch or bounce enough to have any appreciable effect on precision. Once again, if it does then you have mis-designed the system. I have seen belt drives such as this doing way more than that. Some of the mentioned RF-45 builds sling a head around vertically that weighs about 150 pounds at those kind of speeds while employing a belt drive on the screw. You can state otherwise but the fact that those machines exist and perform well is an irrefutable fact.

    I've checked some of my tables, my small 16.535in x 9.055in x 1.9685in (420mm x 230mm x 50mm) mill table weighs 68lbs, my rapids are 590IPM, when the machine was not bolted to a pedestal it would slide across the floor during rapids and the machine weighs over 600lbs, using a GT2/GT3 belt would be an exercise in stupidity, but, if you think it wont stretch or bounce knock yourself out and let us know how that works for you moving something that has any weight.

    I have some larger tables upwards of 135lbs (24in x 16in x 2.5in) with 590IPM rapids and 100% table travel, there is no way I would put anything less than HTD8 as a drive belt or anything less than 1.5KW servo motor to move it.

    Now ignoring the issues with inertia alone, if you don't program your mach3 or linuxcnc to have a fairly mild acceleration/deceleration curve you're asking for trouble because if it's aggressive you can kiss that belt good-bye quickly.
    So since it won't work in your specific application then it is apparently no good for anybody else's application. The way belts snap is if their capacities are exceeded. Nobody said to exceed the capabilities of the belt and design criteria was mentioned that would prevent exceeding the capabilities. If the manufacturers design criteria shows that the belt pulley combination can take the maximum torque that the motor can apply then the belt will not snap. It can't because if the maximum torque of the motor is exceeded then the motor will slip and the belt will not break.

    The motor in the link sees a lot more than 7,000RPM but at 6,000RPM (2:1 pulley reduction = 3000RPM) the inertia of a 60lb table would still cause belt bounce/stretch and affect precision with GT2/3 x 15mm, now you want to reduce the speed from well over 15KRPM to 3.5KRPM and expect decent bang for your buck, that would be very unlikely, it's no longer cost effective unless you have the motors in hand, access to a mill and lathe to redesign the motor body, add forced air cooling and a cheap encoder and it's still going to be expensive.
    I don’t know what motor you are looking at, but the motor in the link is 350Kv, so at 24V it is 8400 rpm max. Used as a hobby motor the recommended battery is 5S max or 18.5V, yielding a 6475 maximum rpm. Not sure where you are getting the 15,000 rpms from but it isn't from the motor specs for the linked motor. In any case, lowering the max rpm is a trade-off that is at the user’s discretion and the maximum speed of the motor is not what is listed on the motor specifications but is what the system limits it to. Yes, reducing the rpm will reduce the power available, which was mentioned, but this kind of trade-off is made every day in industrial applications. As for as bang for the buck, if you don't think that the motor delivers enough to satisfy you them don't use it, but don’t try to dictate how others will value the motor. I for one see it as a low cost motor that has readily available brand new replacements I can buy at a click of a button. That alone has some value to me.

    Maybe it's just me but I expect a machine to operate just like it's bigger brothers so mechanical limitations by design need to be minimized as much as possible and not be a restriction, milling off dovetails and installing linear rails is an inexpensive hardware modification but the cost of rail can be cheap to expensive depending on your taste but they will heavily impact motion due to reduced drag so anyone thinking that a GT2/GT3 belt wont stretch or that precision wont be affected has no clue about acceleration, deceleration and inertia or only plays with toys.
    Maybe it’s just me but what you want from your machine may not be what others want and you are applying your criteria to everyone. You stating the reasons why a motor or belt drive won't work for you and your specific application doesn't mean that they won't work just fine for someone else. Stating your reasons as if they are set in stone is pure crap. What you have no clue about is how others want to use their machines and applying your criteria in this kind of broad brush fashion to someone else's machine is unjustified.

  5. #105
    Join Date
    Apr 2004
    Posts
    733

    Re: Open source low cost servomotor controller

    Websrv,

    If I had the choice, I would use differential encoder outputs but those encoders do cost more. Plenty of us digital, avago, HP/Agilent single ended encoders are being used today just perfectly fine. Just as servo drivers that are only single ended encoder input such as Geckodrive and leadshine. I have about a dozen Gecko320's that's been running for years. Those companies sell thousands of them.

    You don't need expensive encoders either. amt/cui encoders work well and under $25. AMS magnetic encoders are even cheaper.

    You can always add external differential encoder line drivers if you require them.

    The idea here is to get brushless driver/motor technology on a hobbyist budget. A single new tamagawa encoder that you propose cost more than the driver itself. Doesn't matter if you can get Tamagawa encoders and brushless servo motors for cheap. If we can't easily buy them at the same price, what's the point of even mentioning them.

  6. #106
    Join Date
    Jun 2010
    Posts
    4252

    Re: Open source low cost servomotor controller

    From a technical, electronics-engineering PoV, a single-ended line driver is just as good as a differential line driver up to a point. If the distance is under a few metres, there is some screening around the cable, a 'hard' line-driver is used and the frequency is under (say) 500 kHz, then going differential brings little extra value. Many computer busses run at tens of MegaHertz like this - albeit slightly shorter. If you need to run at 100 MHz over 10 metres, then you go differential - like a twisted pair Ethernet cable.

    The commercial market is a good guide for this. Huge numbers of single-ended encoders are available, and most servo-drivers can take single-ended encoder inputs. All of which means that commercial system builders around the world are happy to run single-ended.

    I looked at converting my system to differential when I was rebuilding it. The cost would have been negligible for me, but it was not worth the bother. Instead I put in some optical isolation at the receiver end, and focused on getting that to run at MHz speed.

    As for the belts business - we have had that argument here many times before. The older X and L types of belts may have given some problems, especially if mounted wrongly, but the modern GT2/3 belts (and variants) pump kWatts of power at high speed with no visible backlash. To be sure, if you have not used them, it might seem a big claim, but again, check the commercial marketplace for acceptance. Sales are ... huge.

    Cheers
    Roger

  7. #107
    Join Date
    Jul 2012
    Posts
    106

    Re: Open source low cost servomotor controller

    Quote Originally Posted by jfong View Post
    Websrv,

    If I had the choice, I would use differential encoder outputs but those encoders do cost more. Plenty of us digital, avago, HP/Agilent single ended encoders are being used today just perfectly fine. Just as servo drivers that are only single ended encoder input such as Geckodrive and leadshine. I have about a dozen Gecko320's that's been running for years. Those companies sell thousands of them.

    You don't need expensive encoders either. amt/cui encoders work well and under $25. AMS magnetic encoders are even cheaper.

    You can always add external differential encoder line drivers if you require them.

    The idea here is to get brushless driver/motor technology on a hobbyist budget. A single new tamagawa encoder that you propose cost more than the driver itself. Doesn't matter if you can get Tamagawa encoders and brushless servo motors for cheap. If we can't easily buy them at the same price, what's the point of even mentioning them.
    A china factory is now cloning the Tamagwa 16,000RPM 2500PPR quadrature encoder for one of my works customers at a cost of $26.50/ea but come in a box of 24 so a cheap source is available if you need to make a couple dozen motors and I have the advantage of buying them in quantities of 3 or 4 on occasion so making a set of motors for a machine I'm building after the project is finished and I have the design modified or building 48 motors to sell could be something to consider if a supporting driver existed and I had the money to waste on stocking a product for sale.

    Measuring cable length on my BT20 machine, the length from the X-axis motor to X-axis driver is 12M but Y-axis cable length is only 3.5M and Z-axis cable length is 6M so using differential encoders is prefered

    Yes you can suggest that people could just build a board to support differential encoders and skip the encoder error detection to use with ODrive but to me that's a solution for an incorrectly designed product so I'll go that extra mile when it's done and pay to have it added it to get the flexibility in encoder support so I don't have to make a separate board which doesn't take advantage of the available encoder error detection.


    Quote Originally Posted by 109jb View Post
    No I don't think I need to reconsider anything. There are many many machines built that have used belts for reduction drives on their axes, and have done so successfully with precision that is more than adequate for the intended applications. To say otherwise totally ignores those successful applications. As a matter of fact there are machines using narrow XL belt profiles on RF-45 size machines running at +200 IPM speeds and doing so successfully, but the GT2/3 is a better choice in my opinion.


    You can be damn sure thatnobody stated "emphatically" that all machines use aluminum or steel pulleys? Show me where that was said. It wasn't and any claim that it was is just BS. What was said was that aluminum was the norm and I stand by that. Sure some have used non-metal sprockets. However, by and large, the build threads on this forum show that the most commonly used are the aluminum pulleys. You can argue that all you want and you would be wrong. While speaking of non-metal pulleys, you are apparently claiming that they are so rubbery or so prone to exploding that they are not appropriate for using in a precision application. This has no basis in reality. When used appropriately they can be every bit as precise, accurate and reliable as metal. To say otherwise is wrong and totally unsubstantiated.
    I have personally exploded a GT3-127T-12 GRN pulley spinning at 12,000RPM and stopping it and switched to the bullet-proof KRP pulley because I was concerned the size and weight of the aluminum pulley might distort it's shape permanently.

    I'm not denying that the GT2/GT3 is a better profile over the XL and definitely quieter at low RPM over the HTD and seem suitable for spindle driving up to 2.2KW where low mass/inertia are present but it's high mass/inertia that are belt killers.

    In post #98 you state that aluminum pulleys are the norm and now you state that the posts here commonly use aluminum pulleys and this may be so but it doens't make it the industry norm, just the norm here, if you really want to split hairs, this is 100% incorrect as the majority of the machines I see on a daily basis are employing composite (GRN - glass reinforced nylon or KRP - kevlar reinforced polypropylene) pulleys but in reality, direct drive is the norm and I have yet to see a composite pulley fail before it's aluminum twin when used with 4,000RPM MAX servo motors.

    One of the links in his (macdowswe) post to hobby-king is for an SK3 6374-149kv, with a max voltage of 70VDC and the SK3 4250-350kv you point to has a max voltage of 19VDC, of course people overdrive these motors without thought or really understanding it's intended electrical design but I wouldn't advise or recommend it, I'd find a 24V BLDC motor and operate it at 24V.

    Also, ODrive wont handle this current requirement as some other real time tests are reporting the motor is a gutless wonder if the current is restricted or limited and that it doesn't start to generate any power when the current is below 15A and this 1007W (not 1190W as advertised) is input power, since the motor is only 66% efficient output power is only 671W, 19V@53A and you think 24V@6A using ODrive to power it is going to fly?

    My china clone 6364-250kv hits 6,100RPM at 24V and draws 6A (noload) and peaks at 19A and gives 310W of output, this is almost suitable if it weren't for the peak current, the required forced-air cooling and the > $80.00 price tag.

    Out runner as a spindle motors is another example, I wont go into how wrong it is for the general hobbyist to make a AIO spindle from an out-runner motor without decent plans to follow or some serious knowledge in spindle loading forces and design criteria but they do it just the same.

    I get a kick out of those that make a body, drop in a couple of cheap multi-purpose deep groove bearings and an ER16 or ER20 collet head and think they've got a MAC-DADDY spindle, only the smart guys will use a dual row angular contact bearing in the nose of the spindle for bi-directional thrust forces because they expect the thing to stand up to the abusive forces of cutting without eating the bearings or living more than 3 months.

    I've not applied my criteria to anyone else, I simply state if you have weight to move, small skinny belts are unwise to employ and you refuse to acknowledge this making claims that the belt wont stretch or break and this is unrealistic and hilarious.

    Because some yahoo has BT2-15 belts on his RF45 doesn't mean much if his acceleration/deceleration curves are so mild that the ramp-up/ramp-down speed prevent the motion from achieving >100IPM for a movement of less than 4in or ever moving faster than 200IPM.

    The table travel is small, it's a dovetail machine and with the gibs tightened down it probably helps control runaway inertia by restricting movement and the table only weighs 47lbs but I couldn't call it a good quality machine that's prime for a CNC conversion without doing some serious reworking and dumping the dovetails for some unrestricted linear motion capabilities which very few do.

    I've never seen an RF45 or even an RF45N2F (it has a larger table) move at more than 200IPM rapids and never cutting (a serious cut like 0.250in with a 1/2in, not a .025in pass with a 1/4in end mill) above 50IPM without noticing head deflection so it's not in my opinion a reasonable example of a real machine, it's a large toy.

    Oh, again my mistake, when I say REAL machine, I am referring to a smaller version of a REAL larger machine such as a CadetMate 6030 or CadetMate 7040 where the only different is the physical size of the machine.


    So despite all your bickering, your choice of a new (click to buy) BLDC motor, aluminum to make a new body, new bearings from VXB, 4150 steel for a new shaft, a new encoder from a source with an endless supply for others to buy them, a new fan, connectors and terminals of appropriate size and current rating, hardware to assemble it will cost more than I can produce a 48V or 60V 500W servo motor with a 2500PPR quadrature encoder that wont have any heat issues or require any external cooling.

    Oh, I've got a novel idea, why don't you create a kit and sell it to the users here, I'm sure they would be happy to buy your parts and buy the hobbyking motor and assemble it themselves as a cheap solution.

    I've ordered your 4250-350kv motor from hobbyking, when it arrives I'll connect it to my dyno and can then better comment on it but I doubt it's going to provide any eye opening results.

    Now if you want to talk about spindles, your 4250-350kv wont cut it as a spindle motor if you need some serious power with RPM, with the help of a friend I've made my own 0-12,000RPM 220V 3-PH asynchronous servo spindle motor with an output rating of 3HP in a really small size (90mm x 90mm x 260mm) belt driving a BT20 ATC spindle and while cheaper than purchasing a commercial product, if you don't have access to precision lathes and mills and balancing equipment to do the work I wouldn't recommend it (you can have it balanced at a electric motor repair shop if you can do everything else) and I have yet to see a BLDC solution that can come close for the same or less money and nothing in a small size that doesn't have heat issues.

    Because someone does a CNC conversion on a machine such as an RF45, it doesn't mean a whole lot when in the conversion process they plan to create limitation in movement speed to save a buck, you only see people posting about their builds if they're either doing it right or doing it cheap.

    You can't have it both ways if you expect it to be duplicated by another person (because you have motors doesn't make it cheap for the next guy who doesn't) and people tend to follow the most successful build they can afford to follow, replacing the leadscrews and dovetails with linear rails and precision ground ballscrews on an RF45 is the correct way to do it but it sure isn't the cheap way to do it.

    Now ODrive is being created to support single ended encoder signals and there is a difference between transistor and transmitter outputs and even with transistor outputs there are two types, NPN (most common) and PNP and they respond differently, personally I prefer a quadrature differential line transmission type but as long as a dependable source for the specified encoder (new) is provided I have no issues using it but encoder flexibility is currently limited to single ended transistor output encoders so making a decent servo motor available at a cheap price with such an encoder isn't a consideration I'd be willing to make.

    Now you want to talk about flexibility and how everyone is different and can decide how they want to build things and this is false, this option is nonexistant with respect to encoders and ODrive, people are buying the board, and they can't just connect any encoder because line receivers are not employed and the device is set up and tested with the OP's encoder on hand so flexibility is limited to the type currently employed and tested so now people have no choice in the encoder they can use if they can't make or buy a board that gives them the connections for the encoder in the motors they bought or made.

    You must tell them they can use any encoder they want as long as it's a single ended transistor output type.

  8. #108
    Join Date
    Jun 2010
    Posts
    4252

    Re: Open source low cost servomotor controller

    I have personally exploded a GT3-127T-12 GRN pulley spinning at 12,000RPM and stopping it
    You wouldn't have a photo of the remians by any chance. I would LOVE to see that.

    the length from the X-axis motor to X-axis driver is 12M but Y-axis cable length is only 3.5M and Z-axis cable length is 6M so using differential encoders is prefered
    12 m? Have to agree there. Some nice RS422 driver chips such as 26LV31E maybe? That's what I have used in the past.

    Cheers
    Roger

  9. #109
    Join Date
    Jul 2012
    Posts
    106

    Re: Open source low cost servomotor controller

    Quote Originally Posted by RCaffin View Post
    I have personally exploded a GT3-127T-12 GRN pulley spinning at 12,000RPM and stopping it
    You wouldn't have a photo of the remians by any chance. I would LOVE to see that.

    the length from the X-axis motor to X-axis driver is 12M but Y-axis cable length is only 3.5M and Z-axis cable length is 6M so using differential encoders is prefered
    12 m? Have to agree there. Some nice RS422 driver chips such as 26LV31E maybe? That's what I have used in the past.

    Cheers
    Roger
    I never thought to take pictures of the exploded pulley and there didn't seem to be any point in keeping it around so it was tossed at the time of replacement.

    The 26LV31E is a line driver with four single ended inputs and four differential outputs so I think you meant to reference 26LV32D or 26LV32NS which is a line receiver with four differential inputs and four single ended outputs.

    Yes in a pinch a quad RS422 driver chip would work but again, this requires adding an external board to provide encoder support, (board also needs a +3.3V REG to drop the +5V) which in my opinion should be present and I'd rather do it on the board where it's more appropriate and I can take advantage of the available encoder error detection provided by the MAX3097ECSE/MAX3098ECSE (tri-channel encoder line receiver) which will also accept being driven from NPN and PNP transistor driven encoder output signals.

  10. #110
    Join Date
    Jan 2005
    Posts
    1943

    Re: Open source low cost servomotor controller

    I have personally exploded a GT3-127T-12 GRN pulley spinning at 12,000RPM and stopping it and switched to the bullet-proof KRP pulley because I was concerned the size and weight of the aluminum pulley might distort it's shape permanently.
    Then you designed it wrong. Plain and simple. No doubt about it. No argument can be made otherwise.

    I'm not denying that the GT2/GT3 is a better profile over the XL and definitely quieter at low RPM over the HTD and seem suitable for spindle driving up to 2.2KW where low mass/inertia are present but it's high mass/inertia that are belt killers.
    No matter how much mass or inertia there is, if the motor cannot supply the torque to exceed the belt or the pulley capabilities then they won't break.



    In post #98 you state that aluminum pulleys are the norm and now you state that the posts here commonly use aluminum pulleys and this may be so but it doens't make it the industry norm, just the norm here, if you really want to split hairs, this is 100% incorrect as the majority of the machines I see on a daily basis are employing composite (GRN - glass reinforced nylon or KRP - kevlar reinforced polypropylene) pulleys but in reality, direct drive is the norm and I have yet to see a composite pulley fail before it's aluminum twin when used with 4,000RPM MAX servo motors.
    Well lets see. We are in a thread about a low cost servo drive for hobbyists using hobby motors and being discussed by hobbyists. I think most everyone would figure that the posts were in relation to hobbyists and not in relation to industry.

    One of the links in his (macdowswe) post to hobby-king is for an SK3 6374-149kv, with a max voltage of 70VDC and the SK3 4250-350kv you point to has a max voltage of 19VDC, of course people overdrive these motors without thought or really understanding it's intended electrical design but I wouldn't advise or recommend it, I'd find a 24V BLDC motor and operate it at 24V.

    Also, ODrive wont handle this current requirement as some other real time tests are reporting the motor is a gutless wonder if the current is restricted or limited and that it doesn't start to generate any power when the current is below 15A and this 1007W (not 1190W as advertised) is input power, since the motor is only 66% efficient output power is only 671W, 19V@53A and you think 24V@6A using ODrive to power it is going to fly?
    Let me ask you what machine you think I am intending to use this on? I don't think I mentioned it, so you can't possibly look only at the motor specs and say with any intelligence whether it is or is not suitable. You can spout all the numbers you want but without knowing anything but motor parameters you are only blowing smoke.

    My china clone 6364-250kv hits 6,100RPM at 24V and draws 6A (noload) and peaks at 19A and gives 310W of output, this is almost suitable if it weren't for the peak current, the required forced-air cooling and the > $80.00 price tag.
    Don't really care what you think is or isn't a good value. If I want to spend my money on something that you don't think is a good value, that is my choice.

    Out runner as a spindle motors is another example, I wont go into how wrong it is for the general hobbyist to make a AIO spindle from an out-runner motor without decent plans to follow or some serious knowledge in spindle loading forces and design criteria but they do it just the same.

    I get a kick out of those that make a body, drop in a couple of cheap multi-purpose deep groove bearings and an ER16 or ER20 collet head and think they've got a MAC-DADDY spindle, only the smart guys will use a dual row angular contact bearing in the nose of the spindle for bi-directional thrust forces because they expect the thing to stand up to the abusive forces of cutting without eating the bearings or living more than 3 months.
    Another useless diversion from the topic of this thread. Why are you discussing spindles.

    I've not applied my criteria to anyone else, I simply state if you have weight to move, small skinny belts are unwise to employ and you refuse to acknowledge this making claims that the belt wont stretch or break and this is unrealistic and hilarious.
    Tell you what. You tell me how much a 15 mm wide GT3 belt drive will stretch and how much it will affect precision on an unknown machine with unknown mass, unknown inertia, and unknown desired performance..

    You are basically saying that belt stretch will be too much without knowing anything about the end product and applying how you want your machines to work to everyone else.

    Because some yahoo has BT2-15 belts on his RF45 doesn't mean much if his acceleration/deceleration curves are so mild that the ramp-up/ramp-down speed prevent the motion from achieving >100IPM for a movement of less than 4in or ever moving faster than 200IPM.

    The table travel is small, it's a dovetail machine and with the gibs tightened down it probably helps control runaway inertia by restricting movement and the table only weighs 47lbs but I couldn't call it a good quality machine that's prime for a CNC conversion without doing some serious reworking and dumping the dovetails for some unrestricted linear motion capabilities which very few do.

    I've never seen an RF45 or even an RF45N2F (it has a larger table) move at more than 200IPM rapids and never cutting (a serious cut like 0.250in with a 1/2in, not a .025in pass with a 1/4in end mill) above 50IPM without noticing head deflection so it's not in my opinion a reasonable example of a real machine, it's a large toy.
    Didn't you just say you just state that you don't apply your criteria to anyone else? Re-read the above snippet, particularly this part.

    but I couldn't call it a good quality machine that's prime for a CNC conversion without doing some serious reworking and dumping the dovetails for some unrestricted linear motion capabilities which very few do.
    and this one too

    so it's not in my opinion a reasonable example of a real machine, it's a large toy.
    That sure seems like applying YOUR criteria to me. Many people chose RF-45 type machines and other benchtop mills for conversion and are very happy with them FOR THEIR PURPOSES. Not for YOUR purposes.

    Oh, again my mistake, when I say REAL machine, I am referring to a smaller version of a REAL larger machine such as a CadetMate 6030 or CadetMate 7040 where the only different is the physical size of the machine.

    So despite all your bickering, your choice of a new (click to buy) BLDC motor, aluminum to make a new body, new bearings from VXB, 4150 steel for a new shaft, a new encoder from a source with an endless supply for others to buy them, a new fan, connectors and terminals of appropriate size and current rating, hardware to assemble it will cost more than I can produce a 48V or 60V 500W servo motor with a 2500PPR quadrature encoder that wont have any heat issues or require any external cooling.
    Lets see some more application of your criteria.

    First, apparently the only REAL machines are machines that match your idea of performance, that being performance equal to an industrial VMC with only smaller work envelope.

    Now about the motor in question:
    Aluminum to make a new body - Why? I see no reason to for MY potential application. A new body is YOUR criteria
    new bearings from VXB - Why? I see no reason for new bearings until the ones it comes with start to go. YOUR criteria again.
    4150 steel for a new shaft - I see no reason for a new shaft for the motor. YOUR criteria again.
    a new encoder from a source with an endless supply for others to buy them - Umm what the hell is this. Any servo is going to need an encoder - Red herring
    a new fan - I have a box of surplus fans that will work fine. A new fan is YOUR criteria again.
    connectors and terminals of appropriate size and current rating - Once again, any motor is going to need these. Red herring

    it will cost more than I can produce a 48V or 60V 500W servo motor with a 2500PPR quadrature encoder that wont have any heat issues or require any external cooling
    And you are going to sell these to everyone? Your can produce it, but others can't, so this is again YOUR criteria. BTW, this is the ODrive thread, so you might want to consider a 24V motor since the ODrive has a 24V bus. Your 48V or 60V motors might under-perform

    Oh, I've got a novel idea, why don't you create a kit and sell it to the users here, I'm sure they would be happy to buy your parts and buy the hobbyking motor and assemble it themselves as a cheap solution.
    I've got a novel idea. If you approach is so much better why don't YOU put together a kit of your motor and sell it to everyone as the cheaper solution. When you do this then the above comment might bear some weight. Until then it is a hot air.

    I've ordered your 4250-350kv motor from hobbyking, when it arrives I'll connect it to my dyno and can then better comment on it but I doubt it's going to provide any eye opening results.

    Now if you want to talk about spindles, your 4250-350kv wont cut it as a spindle motor if you need some serious power with RPM, with the help of a friend I've made my own 0-12,000RPM 220V 3-PH asynchronous servo spindle motor with an output rating of 3HP in a really small size (90mm x 90mm x 260mm) belt driving a BT20 ATC spindle and while cheaper than purchasing a commercial product, if you don't have access to precision lathes and mills and balancing equipment to do the work I wouldn't recommend it (you can have it balanced at a electric motor repair shop if you can do everything else) and I have yet to see a BLDC solution that can come close for the same or less money and nothing in a small size that doesn't have heat issues.
    I don't want to talk about spindles. I want to talk about servo drives in a servo drive thread, so I could care less.

    Because someone does a CNC conversion on a machine such as an RF45, it doesn't mean a whole lot when in the conversion process they plan to create limitation in movement speed to save a buck, you only see people posting about their builds if they're either doing it right or doing it cheap.

    You can't have it both ways if you expect it to be duplicated by another person (because you have motors doesn't make it cheap for the next guy who doesn't) and people tend to follow the most successful build they can afford to follow, replacing the leadscrews and dovetails with linear rails and precision ground ballscrews on an RF45 is the correct way to do it but it sure isn't the cheap way to do it.
    More application of YOUR criteria. What you are discussing is a trade study. If a person limits performance to save money does not mean their build is invalid. Maybe to you it is, but you aren't everyone and your criteria isn't everyones.

    replacing the leadscrews and dovetails with linear rails and precision ground ballscrews on an RF45 is the correct way to do it but it sure isn't the cheap way to do it.
    By YOUR criteria it is the right way. Someone else may say that dovetails an rolled ballscrews are good enough for what they want to do with the mill, and that decision would be the right way to do it.

    Now ODrive is being created to support single ended encoder signals and there is a difference between transistor and transmitter outputs and even with transistor outputs there are two types, NPN (most common) and PNP and they respond differently, personally I prefer a quadrature differential line transmission type but as long as a dependable source for the specified encoder (new) is provided I have no issues using it but encoder flexibility is currently limited to single ended transistor output encoders so making a decent servo motor available at a cheap price with such an encoder isn't a consideration I'd be willing to make.

    Now you want to talk about flexibility and how everyone is different and can decide how they want to build things and this is false, this option is nonexistant with respect to encoders and ODrive, people are buying the board, and they can't just connect any encoder because line receivers are not employed and the device is set up and tested with the OP's encoder on hand so flexibility is limited to the type currently employed and tested so now people have no choice in the encoder they can use if they can't make or buy a board that gives them the connections for the encoder in the motors they bought or made.

    You must tell them they can use any encoder they want as long as it's a single ended transistor output type.
    Everyone can still decide how they want to build things, If ODrive can only use one encoder type then if I want to use ODrive I can use the one type encoder specified, or I could devise a way to use another type or I could choose to not use ODrive. Still a choice. Personally since the encoder that has been tested is a $10 piece, I have made MY decision that this isn't an issue for me. The fact that you don't like this style encoder and think there are better choices is up to you but has absolutely no effect on me.

  11. #111
    Join Date
    May 2016
    Posts
    41
    Hey guys, to help picking out the right brushless motor for your application, I made this spreadsheet, feel free to make a copy and play with it: Link.

    Its intended to be a basic guide, so it has some limitations: The acceleration times do not consider friction, and the current ratings are a bit dubious on hobbyking (you could possibly push more for a short time, or it could be overrated for continuous current).

    Cheers!

  12. #112
    Join Date
    Jul 2012
    Posts
    106

    Re: Open source low cost servomotor controller

    A GT3-15 belt is a 3mm pitch timining belt, following advice from someone else just like you who emphatically claimed the belt would stand up to a 3HP, 8,000RPM motor and would handle moving a 8' x 14' router gantry with a weight of 670lbs at 1000IPM was so far off the mark and the damage that was caused when the belt snapped and the gantry slammed into the front frame support was nothing shy of life threatening.

    You want to claim that the belt is suitable for an application then provide the details of the application and stop making such false blanket statements and trying to discredit opposing statements based on the same lack of information you make your claims with.

    Because I don't call an RF45 a quality machine prime for conversion doesn't mean that someone else has the same thoughts, I'm not telling them how to perceive the machine, I only state how I perceive that particular machine.

    Now, I have no particular preference for any style of encoder and only a fool would conclude otherwise, I just know from experience that there is no such thing as a one encoder works in every application and I've seen too many issues with single ended encoders and long cables to even consider it in a design where long cables are used.

    You want to accept the lack of flexibility in encoder support knock yourself out, the majority of people who will buy this product will buy it assembled and since the project creator doesn't have an available solution for use with differential encoders when long cables (>9M) are involved, it reduces the products suitability as an inexpensive solution when no option to use it with supporting encoders and long cables are involved and no, most will not design and produce a converter board just to use a differential encoder with long cables, they'll opt for a different solution.

    When all is said and done, if a differential encoder is not supported I'll spend the time and make the eagle files and pay to have the support added for my own personal use and I have no issues publicly stating my intentions regarding this product and all credit for the design will be to macdowse since it's his baby and I acknowledge this.

    That way if I have an application that needs or uses differential encoder, it wont be an issue and what you do has absolutely no effect on me or anyone else.

    Any argumentative statement about my encoder type is nothing more than someone trying to come off as the big man on campus who is wasting their futile efforts trying to publicly discredit anything I've said which offers controversy or conjecture to statements they themselves have made.

    Building and attempting to sell a BLDC servo motor when no inexpensive supporting driver is available is an exercise in foolish spending and you don't have to be a rocket scientist to figure this out.

    If later I do an updated driver for myself, there's still no reason to make and sell a BLDC servo motor for public consumption since no inexpensive driver is available so making such a suggestion shows your reason for commenting is nothing more than a personal attack and less of a useful contribution.

    My whole thought is that if ODrive at a cost of $50.00 to $80.00 offers flexibility in deployment, it will be seen as a suitable, viable and sought after product and making available a suitable and inexpensive supported BLDC servo motor would further increase it's value and makes sense to produce a suitable motor for general consumption but I'm not going to steal his project and build it with the flexibility I believe it should contain, it's macdowse's project, he ultimately decides what options and features the public release will contain and to think otherwise would be disrespectful.

  13. #113
    Join Date
    Jun 2010
    Posts
    4252

    Re: Open source low cost servomotor controller

    so I think you meant to reference 26LV32D or 26LV32NS
    Too much shorthand. Sorry.
    I was thinking of how to make a common low-cost single-ended encoder drive 12 m of wire reliably. The use of a 26LV31E as a driver 'sort of' implies the use of a matching receiver, with the 26LV32 being the obvious choice. I carry both in stock.
    The Avago etc 500 line (2000 ppr) encoders are very common, and I have been using them for .. um ... 30 years now? (Starting with their original incarnation as a HEDS unit from HP.) I find them very accurate and reliable. Plenty of similar units suitable for the LOW-BUDGET home hobbiest.

    Extra PCB - true. But I make my own PCBs on the mill anyhow, and I don't mind knocking out a few for my own use. Could work out more expensive if you have to buy them.

    Cheers
    Roger

  14. #114
    Join Date
    Jun 2010
    Posts
    4252

    Re: Open source low cost servomotor controller

    I have personally exploded a GT3-127T-12 GRN pulley spinning at 12,000RPM
    Um. Checking the Gates Powergrip GT3 Drive Design Manual I find that the 3 HP rating is no problem at all, but they don't spec beyond 10k RPM. Even lower RPM for the bigger tooth spacings. Could be outside the official specs?

    Cheers
    Roger

  15. #115
    Join Date
    Jul 2012
    Posts
    106

    Re: Open source low cost servomotor controller

    Quote Originally Posted by RCaffin View Post
    I have personally exploded a GT3-127T-12 GRN pulley spinning at 12,000RPM
    Um. Checking the Gates Powergrip GT3 Drive Design Manual I find that the 3 HP rating is no problem at all, but they don't spec beyond 10k RPM. Even lower RPM for the bigger tooth spacings. Could be outside the official specs?

    Cheers
    Roger
    When I was told the HP was in spec and it was suggested the RPM was acceptable even though it exceeded it by 20%, stopping abruptly with this pulley was more than it could handle.
    it's since been changed from the large pulleys to a multi-stage pulley design where the combined sets of pulleys are significantly smaller yet achieve the same ratio and only marginally noisier.


    Quote Originally Posted by RCaffin View Post
    so I think you meant to reference 26LV32D or 26LV32NS
    Too much shorthand. Sorry.
    I was thinking of how to make a common low-cost single-ended encoder drive 12 m of wire reliably. The use of a 26LV31E as a driver 'sort of' implies the use of a matching receiver, with the 26LV32 being the obvious choice. I carry both in stock.
    The Avago etc 500 line (2000 ppr) encoders are very common, and I have been using them for .. um ... 30 years now? (Starting with their original incarnation as a HEDS unit from HP.) I find them very accurate and reliable. Plenty of similar units suitable for the LOW-BUDGET home hobbiest.

    Extra PCB - true. But I make my own PCBs on the mill anyhow, and I don't mind knocking out a few for my own use. Could work out more expensive if you have to buy them.

    Cheers
    Roger
    Yes admirable however, not everyone has the ability to create PCB's for a powered cable extender or a differential encoder conversion board.

    I like 10,000 pulses per rev from my encoders so 2500ppr quads are my preference but agree, 2000 pulses is more than acceptable for any application requiring some precision in it's movement.

    One of the unmentioned modifications that was discussed by my friend and I tend to agree is I/we see no reason why this driver should be restricted to 24VDC and a 144W 24V BLDC motor due to voltage/current (24VDC/6A) limitations or can't be used with a 60VDC power source to drive a 300W 60V (10A peak) BLDC motor by making a few component changes and adding a buck to drop it to 12VDC for the rest of the driver. but this is unlikely to be done to the public released version so if you want to use some power, follow the thread for reference and modify and produce the design as you need remembering to credit macdowse for his work and efforts.

    There is a large selection of SOIC-8 FETS that work under higher voltages with relatively low IRDs making heat a non issue, buck converters and switching PSU chips are inexpensive and require minimal parts to implement such as the LM2596HVS-12 (60VDC to 12VDC) so 60V/10A shouldn't be an issue to achieve in an already fantastic design.

  16. #116
    Join Date
    Jul 2012
    Posts
    106

    Re: Open source low cost servomotor controller

    Quote Originally Posted by macdowswe View Post
    Hey guys, to help picking out the right brushless motor for your application, I made this spreadsheet, feel free to make a copy and play with it: Link.

    Its intended to be a basic guide, so it has some limitations: The acceleration times do not consider friction, and the current ratings are a bit dubious on hobbyking (you could possibly push more for a short time, or it could be overrated for continuous current).

    Cheers!
    A couple of nice choices in your motor list I see and the 50-50 580kv and the 50-60 270kv are worth testing to determine suitability for operation at 24V.

  17. #117
    Join Date
    Jun 2010
    Posts
    4252

    Re: Open source low cost servomotor controller

    I like 10,000 pulses per rev from my encoders so 2500ppr quads are my preference but agree, 2000 pulses is more than acceptable for any application requiring some precision in it's movement.

    Just for reference: I use a 500 line (2,000 ppr) encoder, a 3:1 GT2 2 mm toothed belt reduction and a 5 mm ball screw. That gives me 0.8 microns resolution. which is more than enough for me. Typically I set the zeros to within 5 microns. I suspect that below that I may be running into some backlash in the ball nuts ( a few microns?) and the linearity of the ball screws would be suspect. Also, the TIR on the spindle is at least 5 microns, so ...

    Yes, modern FETs have become very good compared to the early FETs we were using in the 90s. Most of them now have integrated flyback diodes, which were NOT there at the start!

    Cheers

  18. #118
    Join Date
    Jul 2012
    Posts
    106

    Re: Open source low cost servomotor controller

    Quote Originally Posted by RCaffin View Post
    I like 10,000 pulses per rev from my encoders so 2500ppr quads are my preference but agree, 2000 pulses is more than acceptable for any application requiring some precision in it's movement.

    Just for reference: I use a 500 line (2,000 ppr) encoder, a 3:1 GT2 2 mm toothed belt reduction and a 5 mm ball screw. That gives me 0.8 microns resolution. which is more than enough for me. Typically I set the zeros to within 5 microns. I suspect that below that I may be running into some backlash in the ball nuts ( a few microns?) and the linearity of the ball screws would be suspect. Also, the TIR on the spindle is at least 5 microns, so ...

    Yes, modern FETs have become very good compared to the early FETs we were using in the 90s. Most of them now have integrated flyback diodes, which were NOT there at the start!

    Cheers
    Precision and tolerance in the hobby market is a byproduct of time, effort and finances but you can achieve more than acceptable tolerance, resolution and repeatability just by component selection without going completely broke so don't discount inexpensive components exclusively by it's cost.

    Seems showing this project to a few others from work has inspired them to jump on the band-wagon and queried regarding the existence of for example, wiring the board in a lathe configuration to be controlled by Mach3 and a quick search of the thread yielded nothing obvious so macdowse, do you have some generic application wiring examples or is this too early in the game?

    While I no longer use Mach3 or LinuxCNC in new builds, these do offer easy to use GUI type controls and a lot of people still use them so I think general or generic application wiring diagrams should be available so the general DIY hobbyist will be able to easily implrement the drive in his application.

  19. #119
    Join Date
    May 2016
    Posts
    41

    Re: Open source low cost servomotor controller

    Hey guys, the v3.1 boards have come back from the factory, check it out!


    I have some more pictures here: link

  20. #120
    Join Date
    Jul 2012
    Posts
    106

    Re: Open source low cost servomotor controller

    Quote Originally Posted by macdowswe View Post
    Hey guys, the v3.1 boards have come back from the factory, check it out!


    I have some more pictures here: link
    Being tied to an expensive manufacturing process and the lack of files to produce the board at other board shops makes this project look less like an open-source project.

    Like many others I am interested in this project but since no one can currently produce this product at a reasonable price due to the lack of eagle files or even worst case non-modifiable gerber files we have to rely on you to make the product available at a reasonable price.

    Can you please inform us when the product is available at a reasonable price and also of interest is your 3-axis and 4-axis version of this product so updates regarding their availability are of interest to many.

    Depending on the results of your finished product, I may commission someone to produce a single axis version based on your code and schematic for my particular use since there doesn't seem to be anything currently available, another project here went closed source in it's final development stages and an inquiry to you in possibly producing functioning code has gone unanswered for more than a week so I'm beginning to adopt the attitude that I should not be concerned about availability for anyone else's use and unfortunately this doesn't benefit the open-source/open-hardware community.

Page 6 of 8 45678

Similar Threads

  1. Replies: 31
    Last Post: 08-24-2015, 10:03 AM
  2. Smoothie CNC controller, Open Source.
    By Rolf_K in forum Laser Engraving / Cutting Machine General Topics
    Replies: 6
    Last Post: 01-15-2015, 07:41 PM
  3. CNC Controller with Arduino - open source
    By Christian Knüll in forum Open Source Controller Boards
    Replies: 16
    Last Post: 12-22-2012, 10:46 AM
  4. Open Source CNC Controller Specification
    By gregmary in forum DIY CNC Router Table Machines
    Replies: 28
    Last Post: 12-05-2005, 04:58 AM
  5. BEST Open source Controller ?
    By Seventeen in forum Open Source Controller Boards
    Replies: 12
    Last Post: 09-22-2005, 12:27 AM

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •