DC motor speed control - anyone build one?

Hi Folks, I know a lot of you are across a few more subjects than metalworking, so..
Looking for a practical "I can build it" speed control for a 180V DC
treadmill motor I got. The original controller is stuffed, traced the fault to (surprise) the MOSFET driver cunningly encapsulated in epoxy, so no hope of seeing what it is. Plan to keep the DC supply (which is 375V DC after full wave rectification from our 240V AC mains supply.) and the existing MOSFET, back EMF diode, and what LOOKS like a zener on the gate of the MOSFET (drive limiting?)
I have a vague knowledge of how the things work, but not at the voltages this will use - how do I limit input motor volts to 180v ?
Ideally, would like dynamic braking, constant speed (there is an optocoupler on the back for a feedback circuit) forward/reverse etc etc...But would settle for a plain vanilla stop-slow-fast-flat out controller.
Anyone got a circuit I could copy? - its no use sending me links to US suppliers, the logistics of getting it here are hopeless.
Besides, I wanna build it meslef (and that should resonate with you guys)
Regards,
Andrew VK3BFA.
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@hotmail.com wrote:

Look up IR's line of FET drivers. I use the IR2113 in servo amps, that has both a high and low-side driver in one package. You could just use the low-side, assuming the rest of the driver is set up for that. Otherwise, there is a floating high-side driver on the chip, too. It needs some extra parts to "bootstrap" the floating bias supply. Go to www.digikey.com, look up the IR2113 and click on the part # in the list. It will allow you to see a PDF file of the data sheet that will explain just about everything, and may also point to other parts in the same product family.
Jon
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Jon Elson wrote:

Also: You can plug in the part number into the search window on the top right corner of: http://www.irf.com/indexsw.html
or this might work: https://ec.irf.com/v6/en/US/adirect/ir?cmd ΚtSearchFrame&domSendTo=byID&domProductQueryName=IR2113
--Winston
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Hi Jon and Winston. Thanks for the reference, it looks like a interesting device, does everything!.Excuse me for being thick, but have been caught before trying to use the "typical applications" circuits in datasheets - they tend to leave a lot of things out which an experienced engineer/ designer would know about.
I am but a lowly technician, I can sus out something dead on the bench in front of me, but to design something from scratch is too much like Rocket Science...
Any further references you can point me to?
Andrew VK3BFA.
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@hotmail.com wrote: (...)

Not off the top of my head. I will let you know if I stumble across something that you might find useful.
Until your design is complete, you might consider: http://www.surpluscenter.com/item.asp?item-2269
--Winston
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

Jon - do you have any particular specs you wish of this controller? if not, a variac is the simplest approach. If you want some feedback, acel control and braking, you will find used controls on ebay and in surplus stores that will do the trick for $10 to $35
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
<snip>
The original controller is stuffed, traced the

This implies that you might have used the existing controller if it hadn't failed ("is stuffed"). If so, maybe you can more easily find another complete unit that has a good controller. Not as elegant, but pretty simple. Some friends of mine are currently using the "2 hp" type treadmill motors to build 2" X 72" belt grinders for knife making.
A friend who lives several hundred miles from those guys decided to build one too. He went to local guy who advedrtises in the local newspaper that he will come to your house and pick up your metal scrap. This guy says he gets a LOT of treadmills and he asked my friend how many of that brand he wanted. I think he paid about $10 each. This was in Rochester, MN, USA.
Pete Stanaitis
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

    Hmm ... I didn't look at where the original poster was posting from, but given the voltages mentioned (elsewhere in this thread), I suspect that the original motor and controller were designed to be powered from a 120 VAC line (typical US power outlets), and may have been connected here to a 240 VAC line (typical UK power). The voltage from the rectifier is just *wrong* for the motor. This could explain why the controller died.
    Enjoy,         DoN.
--
Email: < snipped-for-privacy@d-and-d.com> | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
  Click to see the full signature.
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Mon, 13 Oct 2008 00:33:12 +0000, DoN. Nichols wrote:

Perhaps, although I doubt that much 120V stuff gets imported into Australia. OP = "Andrew VK3BFA"; VK3 prefix is Victoria (the State where Melbourne is the capitol).
--
jiw

Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
DoN. Nichols wrote:

(context restored by Tim:) Looking for a practical "I can build it" speed control for a 180V DC treadmill motor I got. The original controller is stuffed, traced the fault to (surprise) the MOSFET driver cunningly encapsulated in epoxy, so no hope of seeing what it is. Plan to keep the DC supply (which is 375V DC after full wave rectification from our 240V AC mains supply.) and the existing MOSFET, back EMF diode, and what LOOKS like a zener on the gate of the MOSFET (drive limiting?) (end restored context)

A DC rail that's twice the rated motor voltage is a bit odd, but not to be ruled out if the motor and drive are designed for it. Basically you design the system so that the drive pulses get averaged out by the motor inductance; the motor effectively "sees" no more than (rail V) * (duty cycle), so if you restrict the duty cycle you're fine.
It's one way that I'd consider designing the circuit if I were trying to maintain the most common parts for all possible power line voltages in the world. If you select or specify the right motor (and that's a big "if"), then this scheme should work just fine, possibly with just one controller board for the whole world.
I have no clue how the treadmill was supposed to work, but I could easily see the driver being a one-quadrant device, i.e. no braking, no motor reversal, etc. If you want four-quadrant operation (drive in both directions, brake in both directions) then you need a different controller.
--

Tim Wescott
Wescott Design Services
  Click to see the full signature.
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
BTW, a friend who sell and maintains commercial health club gear has an issue with the controllers smoking.
He found the real issue is the bearings drag, and the motor has to work far harder, and you can guess the rest.
Now his repairman look at the bearing drag every inspection.
--
A host is a host from coast to snipped-for-privacy@nrk.com
& no one will talk to a host that's close........[v].(301) 56-LINUX
  Click to see the full signature.
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Sun, 12 Oct 2008 00:13:15 -0700 (PDT), snipped-for-privacy@hotmail.com wrote:

I have. It certainly isn't rocket science, and I get my MOSFET's free. That said, there are so many good commercial products available on EBay etc (or from Noble) that it really isn't worth the bother -- particularly since you want "features".
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

A variable autotransformer and bridge rectifier are suitable.
Electronic speed controllers are relatively cheap and standard, not worth making unless you after a hobby.
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Hi Richard, here in OZ, such things are expensive - a really quick ring around , about the $400 mark for a controller. And we are not at your level of affluence, where treadmills get junked and are virtually free for the taking. (Violin music in background)
I don't want to manufacture the things, I just want to build one. Sorta like cutting threads on a length of rod instead of going to the store and buying a bolt - it interests me, stuff like this has all my life - curious how things work, and like to try it myself. Will even fix $10 toasters for the hell of it.
Yes, the variac and the bridge rectifier "would" work - but would rather go the PWM route as I believe you can get better torque from the motor at low speeds, and thats where I want to run it so I can do thread cutting on my 9 by 20 lathe . Like I said, have the basic theory (Google is your Friend) but theres a dearth of actual schematics (and PCB layouts, but thats a luxury anyway) on the net. And the ones I do find are for your mains of 120V. This one runs at 180V. from our 240V AC mains supply.
Correct me if I am wrong, but if used a PWM modulator (pair of 555's) and made the mark-space ratio 1:1 would it effectively limit the applied volts to half VCC? - or is it more complex than that? (almost certainly) The IR links from Jon and Winston were great, didn't even know such specialized drivers existed - and probably not available here in OZ - and they got lots of nice protection things in them as well - but coupling to the single ended FET in the original board...might be a bit spectacular if I get it wrong, as a tech. have a healthy wariness of high voltage at high current...(add a motor to the mix, well, even more interesting)
The mechanical conversion is going to stretch my neophyte knowledge to the limit, so want it up and running on the bench before starting on the lathe conversion...and its school tomorrow (Tuesday) so going to learn how to do Morse Tapers..(rough it out on the lathe, then use the cylindrical grinder to get to tolerance and finish)
Andrew VK3BFA.
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@hotmail.com wrote:

You can easily make a PWM driver using that or a 556 (dual 555) and a mosfet. The tricky parts will be timing for the slowest speed and finding what chop rate is ok for your use. It really might just come down to what "sound" you like if you use audible frequencies.
To really see what's going on a scope is handy. The oldest crappiest tube heating monster will do just fine. You won't be getting pretty square waves like you expect. Also, cheapo digital volts meters will give strange readings on a PWM driver- low voltages may show as high, high as low etc.
It might be best to start with something low voltage (24 volts etc) and once you can't blow that up any more, work on the line powered version.
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Mon, 13 Oct 2008 02:01:15 -0700 (PDT), snipped-for-privacy@hotmail.com wrote:

Yes. You do need a snubbing diode or switch to short the motor during the "space" periods.
If IR isn't available in VK land, ST Semi certainly is. They make a lot of power stuff. Sales offices in Melbourne and Sydney.
Spehro devised a clever PWM cct based on a 555 and a CD4053 analog switch. Email me if you'd like a schem. I don't think Spehro would mind. I've used it on a couple of my little projects.
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
<> Hi Richard,

I would be willing to bet that there are several of us who would be happy to ship stuff to our southern neighbors (so to speak), so with a little pre-arrangement you could have something shipped to a US address and then forwarded to you. You might even find some of us who would trade for fine Australian products shipped back by return mail. An ebay controller plus $20 for shipping will get you a fine quality item. Oh, and there are junk shops at least around Adelaide, which have real junk that you can buy kinda cheaply - maybe twice US prices but still pretty cheap.
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Essentially you are trying to design a well-regulated, high-current, DC power supply with adjustable output voltage. You can cheaply get either DC voltage regulation or unregulated high DC current; getting both is not cheap and why commercial speed controllers are expensive.
My theoretical understanding would be that the Variac would be better at supplying arbitrary torque (DC current) to maintain a set speed (DC voltage). Variacs are "gutsy" (very VERY low impedance, a benefit of the autotransformer design, maintaining steady output voltage irrespective of current draw, that is, nonlinear and self-regulating). On the other hand, you get some voltage droop with an unregulated bridge rectifier with capacitor. But a PWM MOSFET or SCR arrangement is even less self- regulating, depending on the impedance of the supply backing it. But then SCRs directly on the line are the ultimate low-impedance supply.
It's possible to make a cheap speed controller from a few parts, it just won't have the torque-independent speed regulation of the expensive commerical item.
Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Hi Andrew, I've rebuilt/fixed many treadmills and I've found the following. On our (USA)120V mains a 90V motor is used with an SCR bridge controller. A PWM controller running off of about 135VDC is used with a 130V motor. Double the voltages an you get what you have. It's possible somebody may have mated the wrong controller to your motor. If this is the case then it could be what caused your controller failure. I can post a SCR type controller in the dropbox if you want it. It's output is a form of PWM on full wave rectified DC. Go here http://en.wikipedia.org/wiki/Thyristor and scan down about 1/2 way to "Applications" and see the graphic. Art

Add pictures here
βœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.