[joeavaerage]

Hi peteeng,
Fu******king Oath its a problem!!! How do you suppose a coil in a car works. A current is feed through a coil and then when the points open, ie the current is suddenly open circuited
the voltage spike is sufficient to throw a multi-kilovolt spark at the spark plug when its under compression. That is how Kettering ignition works.

When the coil current of the electromagnetic brake is suddenly open circuited the voltage spike could literally be kilovolts and there will be an arc between the opening contacts.
Lenz's Law:
V=L.dI/dt

So effectively an open circuit causes a very large if not infinite rate of change of current, and therefore even with a low inductance the voltage will be high....kilovolt high!

Craig

[peteeng]

Hi Craig - Thats with a primary and secondary coil so no inductance? This is only one coil. Peter

[joeavaerage]

Hi peeteng,
the coil of a car is an autotransformer. The primary is inductive. When current flows through that coil it generates a magnetic field, and that field has energy. When the current is
stopped suddenly that energy has to go somewhere. The primary has a voltage spike of somewhere between 500v and 1500V, limited only by the condensor and the arc-over voltage of the points.
That voltage spike is transformed up to 15kV to 25kV by the autotransformer

A coil has inductance, and therefore it will store energy when a current flows through it. If that current is interrupted the 'voltage excursion of the inductor will oppose the change in current'. This principle
was discovered by Faraday in the early 1800's. Maxwell took Faradays experimental observations and generated a universal theory of electromagnetism in the 1840's. These remain the basis of electromagnetic theory
to this day.

Craig

[GeckoSub]

Hi,



Cant imagine what's going on. My brake locks on within a few milliseconds, I've never had the servo rotate at all between when the servo Enable signal goes inactive and the brake locking on.

Craig

I will have to read the rest a few times to make heads or tails of it. But the reason I have a delay is that I am running the 24V totally dumb. There's no relay, no connection to an "enable" or anything like that. It's just hooked up directly to mains like the servo drive. The delay comes from the ~ 2 secs it takes for the servo drive to power on and hold the servo whilst the 24V PSU turns on way faster after they both get fed mains power.

As for the rest of the posts on this, I gather you think that's an unhealthy idea?

[joeavaerage]

Hi,

But the reason I have a delay is that I am running the 24V totally dumb. There's no relay, no connection to an "enable" or anything like that.

Well that explains it. there is s delay between when the mains supply is applied to the power supply module and when its output is useful. In that period of time the
brake status and the servo status are not in sync.

You need a relay or switch so the the brake is released at the same time as the servo is enabled, and going the other way the brake is de-energised, ie locked on when the servo is disabled.
Use a relay or a switch. The millisecond delays have not proven problematic for me, but I can well imagine half second and longer delays that you are experiencing would screw my machine as well.

If you do use a relay, and want it to last, then fit a catch diode. The relay will work OK without it but the contacts will suffer arc damage and likely at some point in
the future weld themselves together.

Craig

[GeckoSub]

Heya guys,
This build took another turn a while back. Apologies for not updating
It's no longer a build - I ended up buying a used, smallish machine to get started on.
Same reason as last time I was trying to buy one: Realizing if/when I design my own, it would be smart to actually have some real experience plus, I really need to start making parts.

I lost out on a very good Taiwanese machine frame at a steal of a price, which really hurt and probably prompted this buy. Maybe a bit hurried, but the machine was cheap enough that it's OK and it will be an OK learner. The Taiwanese one was pro level and I would never have grown out of it. This one is possibly more of a stepping stone. But it comes more or less ready to run with a controller and all. Stepper, rolled ballscrews and no idea of the accuracy or stiffness of it, yet. But it will cut alu and get me started. Then in half a year or so I will know if I need to sell this one and upgrade.

The machine just arrived on my island but is still in storage at the freight people's warehouse until I settle a deal on a small workshop space.

Anyways, here are some snaps from the seller and it being sent out:


So, while this has been making its way here from China, I have been busy shopping for vises, clamps, (a lot of) parts for an MQL system, a bit of precut 6061 alu. A lot of stuff to sink my teeth into.
I think I will swap out the controller at some point failry soon as it's a bit limited in features but if the frame is not too terrible, I might also put the servos on it. It's +220kg so I hope it's not as flimsy as it might look.

Not sure if it makes sense to keep posting here. I do think the MQL system I have in mind will be pretty interesting. Ideally, it will be an improved version of a Fogbuster as I will drop the needle valves completely and use a peristaltic pump for dosing and I have some really nice low pressure regulators, too. Also, I will test some Luer Lock dispensing needles as nozzles.

I might post about that here or in a standalone build thread.

Again, thanks so much for the help gents!

[G-Spot]

Was tempted to do the same myself, but a manual machine that could do steel.
If finances are tight and you just need to machine parts and not join a CNC cult then this is the smart thing to do.
You should have waited until you had it all clean and polished and really make us jealous.

[GeckoSub]

Was tempted to do the same myself, but a manual machine that could do steel.
If finances are tight and you just need to machine parts and not join a CNC cult then this is the smart thing to do.
You should have waited until you had it all clean and polished and really make us jealous.

Not sure I can make anyone jealous here. There are some kick ass builds out there. But yeah, I can't wait to get to learn how to run this thing. Honestly, I haven't even slept well for the past week because my mind keeps racing with little projects I wanna make on this thing

But I think you are spot on, this machine had an attractive price and I really, really had to get going on this. Also, that realization that I didn't have any experience to base my design objectives on in terms of a DIY build.

Will post a bit more soon and get some pics uploaded, but yesterday I got it into the workshop. Yes, I found a space. But... it's sitting on the floor. 5 guys were not enough to get it lifted onto the base, haha. But that's alright, I'll spend some time cleaning it up and will try to find an engine hoist to rent for a day. Worst case, I'll get two long pieces of pipe and 8 guys to come lift it.

First impressions is that it's heavy for it's size. But whether it's accurate, we have yet to see. Also, while it's dirty, I cleaned it off in a few places and it cleans up pretty nice. A bit of spot rust but nothing on the ball screws and rails. Strangely enough, I have yet to find a metal chip anywhere. Mouse or rat **** aplenty, though.

Either it was cleaned up really nicely - or it was never used much - before being left for the rodents to use as a toilet

Whether this machine will be a starter, stepping stone or a keeper, I am far from knowing. Only time will tell.

[G-Spot]

But... it's sitting on the floor. 5 guys were not enough to get it lifted onto the base, haha. But that's alright, I'll spend some time cleaning it up and will try to find an engine hoist to rent for a day. Worst case, I'll get two long pieces of pipe and 8 guys to come lift it.

Be careful if you get an engine hoist and have not used one before, they are stable when the load is over the long feet that stick out, but if you try to use them to put something on a bench then those long feet wont fit underneath. You will think "Simple" and position the load sideways to the bench then just swing the load over the bench, if you do so the hoist will become, in the blink of an eye, unstable, the base because it is on wheels that rotate, will fly off away from the bench and the load will come crashing down. Don't swing the load sideways and if you have to then you will need to put chocks under the wheels to stop them from moving. Just blocking the wheels with your feet will not work, the weight of the load is now working like a trebuchet.

[GeckoSub]

Great advice/reminder!
I will go measure the gap under the base right now and then check whether I can fit the hoist feet under it.
If not, I will block it up a little bit higher as the engine hoist really is my best bet here.

[G-Spot]

You will be fine, you will be able to extend the boom fully because you are lifting less than 500kg. This means you wont have to get the feet far under the bench if at all.
What I should have said last time is the danger with these hoists is their wheels auto steer(I said rotate), they can be pointing in one direction when you start to lift then midlift they turn in another direction.
Makes them great for getting an engine or gearbox perfectly aligned because wherever you push the swinging load the base will follow but for lifting other loads in other situations they can be really dangerous.
I have my own, I modified the wheels so I get to do the steering and not the load.

[GeckoSub]

You will be fine, you will be able to extend the boom fully because you are lifting less than 500kg. This means you wont have to get the feet far under the bench if at all.
What I should have said last time is the danger with these hoists is their wheels auto steer(I said rotate), they can be pointing in one direction when you start to lift then midlift they turn in another direction.
Makes them great for getting an engine or gearbox perfectly aligned because wherever you push the swinging load the base will follow but for lifting other loads in other situations they can be really dangerous.
I have my own, I modified the wheels so I get to do the steering and not the load.

Yeah, I was supposed to get the it done today but the car I was gonna borrow to go pick the crane up is broken. It's OK, I put some more time into cleaning the machine up instead.

It's a mystery machine of sorts.
It has some really nice parts on it and some weird ones, haha. E.g. it has Japanese NSK linear guides and blocks and I think it may have ground ballscrews, too. On the other hand, the limit/home switches are USD 1 inductive switches...
Also, on the floating end of the ballscrews they didn't bother to turn down the ends to get rid of the grooves. The ends are just mounted straight into a ball bearing, grooves and all. Maybe they used second hand motion parts and just lopped the ends off.

It has cheap non-closed loop steppers but those are easily upgraded to closed loop or with new motor brackets to the servos I already bought when I thought I was gonna build my own machine.

Speaking of brackets, besides some stripped screws on the sheet metal the only thing that was broken on this machine was the motor bracket for the Y. Either it took a hit in transport or this happened long time ago and was why hte machine was left aside for the rats to **** in
Now, sadly the has some non standard motor brackets that I haven't seen to be able to find anywhere. They have a center height 5mm higher than normal. So, I may just print a 5mm spacer plate and order the standard bracket and then machine my own riser plate as the very first thing.

The sheet metal is fairly thick and the cast iron frame seems very solid, so if they mounting flanges were machined precisely and the machine put together properly trammed and square and all that, I think this will serve me well. I will order a granite triangle soon to check this at some point.

[GeckoSub]

Buy the ready made cables!

Usually:
Step pulse, ideally differential, therefore 2 wires
Direction pulse, ideally differential, therefore 2 wires
24v or 0v (depending on preference/wiring) common for inputs and outputs
Estop signal into an input - estop should ideally physically cut motor power input, but drive can actually actively stop motor (when told to do so) faster than cutting the drive motor power input will (large capacitors in the drive)
You can also feed in limit switches

So, even though, for now, I abandoned the idea of building myself (I bought a used cast iron 3040 router) I'd like to reply to a few things I left hanging here.

The cheap ready made IO cables work. Granted, I've only hooked up 4 pins for testing but they do work. (I set the B2 drive to enable the servo always - not sure if that's wise but for testing it works)
The cables are shielded and the shield on the pin end terminates in a piece of cable, too.

I don't seem to have a picture of the IO cable on its own, but here are a few from when I ran a quick test the first time:


Also, the knock off programming cable for the ASDA-B2 servo drives works:


It took a while to get my head around setting up the gear ratio for the servo drive as well as settling on how many pulses. I just went with 3200PPR as the machine - for a little while longer - will run a mix of steppers and servos and the steppers are running 1600PPR. Also, my Chinese standalone controller is supposedly not super fast.

But yeah, ASDA Soft or whatever it it called works pretty well:


For a while, I thought I had fried one of my servos though (and yes, that's a 3D printed riser plate under the motor bracket...):


But it turns out that I just lost the auto-enable setting in the drive when I did a factory reset. From the vendor, they came with enable on as soon as they see power.

As for the printed riser, the machine uses ballscrew mounts with a bit of a non-standard center height. It came with open loop steppers but since I have the servos, I am going to change them out. And the Y needed it right away as the original stepper bracket was broken. But for the life of me, I couldn't find a new motor bracket for the servo with the same height as the old one. Standard for this size ball screws seems to be 25mm and the old one was 30mm. So, hence the 5mm riser.
The idea is that the very first part the machine will make is its own new riser plate in alu...

Also, I settled on what will be replacing this controller. I will go down the LinuxCNC rabbit hole. The people on the LinuxCNC forum has been super helpful so far and my fears have been alleviated. I am now sure I can make it work

[GeckoSub]

Gap Filling (and maybe a wrap up) - Part 1
I'll add some pics and try to fill in some more gaps
The transport part of all this worked pretty well. The seller had it picked up by a local courier in China who built a simple pallet and box for it and then sent it to my freight forwarder who put it on a cargo ship.

It looked like this when it arrived at the local warehouse here:


We had to take the top of the box off, otherwise it wouldn't fit in the little K-van I could borrow:




Actually, the lift didn't work for this, so six local guys got it into the van:


I also got a workshop space, not too close to the ocean and not so far from home that I can't scoot over there whenever I want. But here we could only get it out of the car, onto the floor and slide it inside. No chance of lifting it up on the table:


Work got in the way and the machine sat on the floor for weeks. I would work on stripping it down and cleaning it from time to time. It was pretty dirty, but no oil nor chips, just dirt and mice crap. Here's the top of the X cover:


The controller box, like the rest, cleaned up pretty nicely:


To be continued...

[GeckoSub]

Gap Filling (and maybe a wrap up) - Part 2
The table showed quite a bit of surface rust - luckily zero rust on the rails nor ballscrews - and I was kinda worried. Started looking into all sorts of rust converters but then decided just to give it a thorough clean and scrub followed by a sand with #800 wet and dry on a tooling plate piece of alu plate. I am pretty sure I will be making a full size fixture plate so I'm not super worried about the table anylonger:


Here's a pic as the strip down began:


And further into the process:


And one of the back of the Z assembly. Chunky enough for the type of work I have in mind, I think. Oh, also about 170mm of travel which is a fair bit more than most of these (fixed) gantry style routers and one of the reasons I jumped on it. Many of them only have 80-120mm.


As mentioned many times by now I have found a ton of mouse poop but not. a. single. chip. Crazy...
I don't think this machine has actually ever been run. My theory is that this was made by someone who wanted to sell these as it has a brand name and contact number on it (dead end). He likely bought the castings and then added used proper Japanese motion hardware to making a working machine he could show to customers. It has NSK C5 ballscrews and linear rails (so, maybe like Chinese "C3") with medium/high preload. But somehow the business endeavor never came to fruition and the machine was stored somewhere for the mice to sh*t in.

Also, at some point in its life, perhaps very early on its first transpo, the machine was bumped pretty badly - the rear took a hit and bent the drain spout of the coolant bassin and broke the Y stepper motor bracket:


Perhaps that's when it was set aside.

Anyhow, I finally managed to get it on it's stand. I had found a car mechanic I could borrow an engine hoist from but the day I went to pick it up we gave it a quick test and it spurted out hydraulic oil... So, I went with the 'Island Lift'. Which is rounding up your friends and neighbors and paying them in Red Horse which is the local strong beer. Sadly, no pictures.
But I had a long piece of galvanized pipe that could accomodate two guys on each side of the machine. And then I had two guys grab the front so six in all. They lifted it fairly comfortably while I rolled the stand/table under it - I put castors on it a while back. Also, it was a bit more stripped down for the lifting. No Z assembly, no covers, no motors, no spindle. I had put the table back on though because there's no way I could lift that up later on by myself.

So, here it is, finally on its stand:


Now, it's all about putting the parts back on and start running it and then upgrading the X and Z to servos (the Z is already swapped for a servo) and switch out the controller for LinuxCNC with a Mesa card and a touch screen. Also, at some point I will be putting on the better spindle I have but for now, I will run and hopefully get the rookie crashes done using the older spindle the machine came with, haha.

Anyways, the bones look strong, they sure are heavy for this little router machine. I will be getting a granite square to tram it fully at some point but time will tell how well it is "in tune".

[GeckoSub]

A Few More Bits & Bobs

I lost a few balls when I pulled the table off, so I 3D printed some dummy rails to help keep them in place. I have just ordered a used rail block which should have the same preload/same size balls and my micrometer is incoming so I can double check:


Oh, I am not always that stupid. Or rather when I am, I sometimes manage to mitigate my foolishness, haha. E.g. I replaced the missing balls on the table with some from the Z-assembly as I can take the Z on and off myself but can't take the table off alone. So, the new balls will go into the Z when I get them and it will be easier this way.

Also, this little baby arrived:


The tool setter was like most other parts here second hand and again at a price I couldn't resist. Let's hope it's not banged up inside. Looks good from the outside.

Also, I mentioned earlier that the machine had rolled ballscrews. I made that statement because on the sellers pictures it looked like that. I think the tell tale is that extra groove deep in the middle of the groove for the balls (I could be wrong). But that was just a line of fat, dark grease and I'm now pretty sure they are ground. Also, they say C5 on them and are proper Japanese made NSKs.

[joeavaerage]

Hi,
good to see you making progress.

The bottom line is try to NOT EVER to remove the cars, and MOST ABSOLUTELY NEVER THE BALLNUTS.

I have made a few improvements to my machine in recent times.

Just before Christmas I bought a new spindle (3.5kW(S1), 3.4Nm(S1), 10,000rpm rated and 40,000rpm max). Between the seller and I we ran foul of New Zealand Customs,
and damn near ended up forfeiting the lot! Lesson learned: NEVER TRY TO DIDDLE CUSTOMS, they have the might and that makes them right! I managed to talk them out of seizing it, but it took a few months.

Anyway, once I got it I then found a second hand 11kW Delta (Highspeed model, ie 0-1500Hz) VFD in the US. Then I had to assemble and make the control gear. I was extremely lucky to find a perfect coolant
circulation pump in NZ cheaply. The rest of the controls are the air solenoids for the tool change, labyrinth seal air purge and taper blow. I also made a control and interlock PCB. It monitors the spindle temp, coolant flow,
drawbar position switches (ToolIn, ToolOut, NoTool) and also the VFD to determine when the spindle is stopped.

At this stage I have just a 10l reservoir for spindle coolant but after three hours continuous use the coolant is heating up to maybe 50C. For prolonged use I am going to have to fit a radiator and cooling fan.

The tool change feature is a real boon to my CNCing. I do not have a tool carousel yet, but even just using the pneumatic tool clamp/unclamp is a revelation and game changer in how you do things.

To date I have ten HSK32/ER20 tool holders that I got with the spindle. I have since added one HSK32/ER25 tool holder for larger (16mm) tools. I also found some Hiamer shrink fit tools. Normally these would sell for over $300USD,
but as second hand I got them for $65USD each. I have yet to make the induction heater for changing tools for them yet....but its underway.

I have underway from the US a HSK32 to 3/4" shell mill arbor and a 1.5" face mill. Should be here in a fortnight or so.

Yesterday I just bought and paid for another ten of the HSK32/ER20 tool holders from the same Vietnamese seller whom provided the last lot at $55USD each (including shipping). They are identical brand to those that I already have
and am very VERY satisfied with them.

Still have many things that I want and need to do. Making a new headstock is near the top of the list. I've pretty much finished the pattern to have it cast in grey iron...just need to find another $1000NZD to pull the trigger on it.
Really need to make a much better flood coolant pump and coolant tank. This new spindle means that my existing flood coolant is not really up to it.

Craig

[GeckoSub]

Hi Craig,
Great to hear from you. I know customs fees is part of how the system works and in a place with zero corruption and benefits, that's normally alright. But yeah, sometimes it's just nuts. And If I lived in Oz or NZ, I wouldn't be able to afford a lot of the tinkering I can here.

Cool that you managed to find that high powered, high frequenzy VFD. I recall you saying you were on the lookout for a while.

I am trying hard not to think about an ATC but time will tell if I will get one. I think, as you say, that even without a tool mag, they make so much sense. Setting tool offsets just once and every tool change being so fast and probably even satisfactory working in tandem with the robot. Clack, hizz, tool dropping into your hand, hizz, clack, new tool in! Or something like that, but it sure looks nice on the the videos, I have seen

As for the linear rail carriages/blocks, the table is put together in such a way I can't unscrew it from the blocks unless I take the whole thing off incl. the blocks. With the Z assembly, I possibly could have. Will keep this in mind though!

On a another note, as I was scrambling around on the internet trying to get the gear ratio correct for the B2 drive, the main lead was an old post from you, I think, on a Mach3 forum. So, thanks for that

[GeckoSub]

Hello again Craig,
Perhaps you can help with this one or perhaps I don't yet have the needed info to make the decision. I surely don't have the know how, yet. Anyways, here goes. I need to settle on the Mesa boards for the Linux conversion and for the digital outputs I need to decide if they should be NPN or PNP as that decides the model of board. And I honestly don't know what to choose.

I guess I will be sending some DOs to the servo drive. Reset alarms, enable, maybe a signal to home if I want to have the drive handle that. And then also I plan on running some air/mist coolant so maybe some signals to run relays/solenoids. Not sure what else needs DOs. Is this enough info to help you guide me to whether I should be looking at board with sinking or sourcing DOs?

[GeckoSub]

Hello again Craig,
Perhaps you can help with this one or perhaps I don't yet have the needed info to make the decision. I surely don't have the know how, yet. Anyways, here goes. I need to settle on the Mesa boards for the Linux conversion and for the digital outputs I need to decide if they should be NPN or PNP as that decides the model of board. And I honestly don't know what to choose.

I guess I will be sending some DOs to the servo drive. Reset alarms, enable, maybe a signal to home if I want to have the drive handle that. And then also I plan on running some air/mist coolant so maybe some signals to run relays/solenoids. Not sure what else needs DOs. Is this enough info to help you guide me to whether I should be looking at board with sinking or sourcing DOs?

I had a closer look at the wiring diagram and also asked on the Linux forum. The Delta drives can be wired either way. I was hoping they would sway the decision so I didn't have to, haha.