[miljnor]

I have a 130vdc treadmill motor that I am going to make a controler for.

I was thinking H-bridge mosfet driver with following criteria: Fwrd,revers and braking. with feadback control to controler (mach2 is my current choice) and the motor I am using is a 2.5Hp continious duty 130vdc 18.5amp.

That being said, I know there are probably drivers out there for this app. but I would like to DIY for learning experience.

So for the only controlers on the market I have found in a doable price range are non feedback or have on board style feed back that is SOOO slow to react as to not be usfull, or have a price tag that is in the stratosphere.

There are serveral H bridge schematics on the net but so far the only ones that I have seen are for low power robotics, nothing on the scale I need.

Since I am relatively new to electronics design, any suggestions would be helpfull. And who knows if we all put enough suggestions (group) in we could probably put a good set of plans up for everyone to use.

My main weakness in design is not knowing the proper chips for this there are so damn many mosfets and mosfet controler chips out there it will take eons to sort through them.

thanks
Michael T.

[H500]

This is a pretty ambitious project due to the high voltage and current requirements. The mosfets that can handle this power level won't come cheap and you can easily fry 20-30 them to perfect the circuit. This is not something that can be expected to work the first time around.

Are you sure the motor is actually 2.5hp continuous? That's more current than what the typical 120v outlets can provide.

Why not start smaller, like 5 amps or so. A suitable mosfet might be the irf460. Did you come across possible bridge gate driver chips? I suspect that line powered ones are not that common.

[Fred in NC]

I might be putting my foot in my mouth, but the treadmill motors run fine from a ready-made Minarik controller, which can be bought surplus for $59 from Surplus Center.

[miljnor]

Are you sure the motor is actually 2.5hp continuous? That's more current than what the typical 120v outlets can provide.

well I could be mistaken on the 2.5hp continuous but the label on the motor say 2.5hp 3500 rpm 130vdc continuous duty PM motor PN # PWM3644-5127-2 .

As far as wall 120 wall current supplying enough amperage you might be right, because the motor is rated at 18.5amps and to my knowledge most house breakers are 10-20amps personaly I am willing to use 230vac as my source.

As far as perfecting the circuit goes, since there are alot of mosfet H-bridge schematics on the net, I don't see myself frying more than a couple of them. But hey maybe I am being unrealistic.

I might be putting my foot in my mouth, but the treadmill motors run fine from a ready-made Minarik controller, which can be bought surplus for $59 from Surplus Center.

You are quite right fred, the controler will run my motor but since the controler is a 12amp?? controler (which I am alread playing with) I wouldn't think it would take running this motor for long.

I have the controller that goes with this motor so worst case scenario is, I copy it. This controller is fed from 120v outlet, but, then the controler blows up quite a bit (which is why I have it). Maybe that is the way to go to save on cost.

This controller is very simple as things go and It probably wouldn't take me long to draw it up. First I will price all of the part out to see if its feasable.

Michael

[Fred in NC]

Michael, treadmill motors are designed for treadmills, and you plug treadmills in the wall outlet, usually 115 VAC at 15 AMP. I think the trick is in the PWM controller, it never gets to draw full amperage.

I am running a 1.5 HP 90 VDC in my wood lathe, with feedback for constant torque. It barely draws 2-3 amps! I can run the lathe, compressor and lights all in the same 15 amp circuit no problem.

[miljnor]

I am running a 1.5 HP 90 VDC in my wood lathe, with feedback for constant torque. It barely draws 2-3 amps! I can run the lathe, compressor and lights all in the same 15 amp circuit no problem.

quote from fred in NC

So basically you think that the minark 12amp controler would work fine with the motor I am using?

I have that controler and it works fine with the motor but I was worried about max amp draw.

I don't think wood lathes realy load motors up all that much or are you using it cnc style? Last I used a would lathe was in high school and with hand held stuff! Don't know that a tool I was willing to hold would realy load a motor up....or am I wrong, its been about 15-20yrs since then

Michael T.

[Fred in NC]

Michael, there is a heat sink that goes with the Minarik controller, to be used when running at full capacity. A few pieces of aluminum angle should make a good heat sink( LLLLL), or even heat sinks from computers, the kind that have a lot of spikes coming out.

[miljnor]

I take that as a yes, but with a heat sink? (got lots of those)

I will probably still go forward with a controller but base it of the one I have (thats Blown) for the motor I am using. There couldn't be more than 20-30 parts all told on the board and I've already priced about half of them <15.00us so far. All though the bourns 5k Trim pot has go (listed out for 14.00us, what is it solid gold?)

michael t.

[H500]

Maybe I misunderstood what you're looking for. Are you just looking for a speed control or a full PID type positional controller?

An h-bridge is much harder to do than a simple pwm speed controller. I believe those surplus controllers are relatively simple thyristor circuits.

Do you have a link to a high voltage h-bridge designs on the web? I haven't come across any that can handle the power level you specified.

[JRoque]

Hey Michael, the Minarik can easily handle that motor. I just cut a bunch of aluminum and had to stop for 5 mins while the motor cooled down. The controller, on the other hand, wasn't even near warm. I could not detect any temperature at all when I touched the heatsink. BTW, the Minarik I use is rated for 10A continuously; the 12A I mentioned before was peak as read with my setup. I honestly would worry more with a high output controller with a lower amperage motor than the other way around. Also, that treadmill motor is 2.5HP peak and can only be run at that rate with an external fan.

Again, if you need to get up and running the Minarik deal can't be beat. But, I know you want to brew your own so let's hear you design ideas for it. What do you want to accomplish? How much power are you looking for? What kind of motor feedback, if any? Have a circuit started already that we can see?

JR

[miljnor]

Sorry H500, Your well above my knowledge base when you start throwin them big words about! Thyristor???

As I stated in my original post the only H-bridge Designs I saw were for the low power robotics. Which is why I am posting.. to either find out that it is doable (without an engineering degree) or to find another cheaper easier way to do it. I am not set on an H-bridge design maybe a "simple thyristor circuit" is the way to build one. I was hoping all/some of the more experienced electronic people would give me a little direction.

Jroque
My design goals are: 1. Computer control
2. w/feedback
3. enough current/voltage for 2.5hp (that I aleady have, which is far in a way larger than the one you mentioned using)
4. A valuable learning tool

I will end up making a controler but the final design might be something other that my original connception..

I am inproccess now of diagraming the controller that runs this motor (the blown one). Also I may not have stated it but this motor/controller setup is from a commercial treadmill and not a Home unit. The two surpluss controlers are for home units, and home units are generally lighter duty.

[H500]

A thyristor is a power control device, like a mosfet or transistor, but with it's own quirks. You probably have a few in your house ...a lamp dimmer switch is a thyristor circuit.

I don't think a high voltage, high current h-bridge is doable unless you have considerable knowledge in power electronics. Most engineers would even find it challenging.

A more practical project might be to beef up your existing controller and add feedback to it.

[miljnor]

Didn't think the h-bridge looked that scary but it is more work than beefing up the controler I have!

When I get the schematic done for this controler Hopefully you (or anyone else with more knowledge than me) will give me some feedback on were and how the beef needs to be added if at all.

The only thing most of the controlers that I've seen don't have is a reverse, which I assum I could add with a solidstate version of a relay.

Thats acutally how I came upon the hbridge circuites, I was researching a way to reverse the motor to eventuall try a r-tapping routine.

[Al_The_Man]

When you say the H bridge controllers don't have reverse, do you mean an input circuit? as the idea of an H bridge is for a reverser, otherwise you would only need a single ended output.
The big problem with H bridge HexFet output is you either have to go with N channel & P channel fets or go with all N channel and now you have to provide some kind of high side driver for the Fets at the 'top' of the circuit, I have used IR2110 which are designed specifically for this purpose.
If you can get hold of IR app. note AN-978 it shows the set up. Also a IC that gives alot of the features you would need to drive the outputs are in IC33033, this is a DC brushless chip, but it will control Brushed DC motors just as well, has fwd/rev, current sense etc. http://www.onsemi.com/pub/Collateral/MC33033-D.PDF
Al

[miljnor]

What I ment was that the controllers you can buy from surpluss center that are fo treadmills don't have reverse. Which is the reason I was looking into the h-bridge in the first place.

Thanks for the doc info. It will take me awhile to digest it.

originaly posted by Al The Man:
I have used IR2110 which are designed specifically for this purpose.

does this mean you've made a mosfet h-bridge driver of the afformentioned Voltage/amp rating? And/Or do you think its as easy as it sounds?

My problem with the H-bridge circuitry to date (besides lack of experience) is I was having problems finding controls/mosfets that would control the gate with voltage available thru said controlers.

If my understanding of mosfets is correct you need line voltage or higher to control the gate on a p-channel mosfet (which most of the controlers that I've looked at don't supply) or you have to use n-channel as you stated (which can make the circuit overly complicated for an amature like myself)

Although skimming this document as I type looks pretty infromative and my answer some of my questions. As always Al you are the Man!

[Al_The_Man]

Manufacturers like IR and motorola etc have come up with ways to provide high side switching, as you have probabally found out, the problem is that the 'High' side fets or IGBT's move from The high voltage when they are off to something way lower when they are on, but the gate to source bias voltage has to remain constant so the bias has to float with the source pin, the IR2110 is one way, Motorola has another method which I would have to dig up (IIRC its MC1000 + optoisolator), I put together one along time ago and will have to dig out my notes on it next week.
With PWM designs it is very difficult to mock up on a breadboard as in a permanent design every common has to go to a solid ground buss or all kinds of jitters occur.
Al
Al

[H500]

Al, thanks for the link. I did not realize IRF make high voltage bridge drivers.

Michel, a pair of ir2183's couple to four irf460's might simplify the h-bridge part of the design. The chip takes care of the trickier parts of bridge control, such as proper gate charging and shoot-through prevention. This would be an interesting project to try. Servomotors, powered direct from the 120v line would be interesting to says the least.

[miljnor]

spindle cw/ccw w/feedback is pushing my electronics knowlede and abilities past the limit as is. A servo motor controler is WAY behond the fold for me! But if you were to try I would love to participate in that one.

There aren't enough open source servo controlers (ie none) out there. Seems stepper win the DIY award for most engineered.

I am still reading the link for the proper way to do things, with what Al supplied.

I noticed on the IR2183 spec sheet it listed in total 3 caps,2 resistors, 2 mosfets and an ir2183. This is half of the H-bridge. I assum the other half would be the second 2183.

That being said.
Isn't that overly simplified or can you get away with that few parts for the actual H-bridge.
And if you are using 2 ir2183's how does this prevent shoot thru? I would think you would have one 2183 control both sides of the bridge in order to prevent a shoot thru from occuring?

being a newby I am probably totaly off base. or could be I missing a controler to control the control??

[Al_The_Man]

You would need 2 IR2183's for the full bridge, you could use the MC33033 probabally to drive them ok. If you look at the info sheet on the MC330033 you will see that motorola uses a MOC8204 for the hi side driver. It might be easier with the 2183 tho.
I assume the IR2183 has replaced the IR2110.
Al

[miljnor]

That sounds like a good combo and I will have to read everything so I can pose a more informed question to connect all the pins. It has 3 inputs for rotor location. don't know that i need more than one for Our app. but it certainly has enough room for expansion.