DC Motor Control from ECU ???
Turbo Mini Forums

I need to control the speed of a 12v DC motor driving my water pump.

I can program the MS3 to output a PWM signal, derived from a duty cycle map of coolant temp vs RPM, either through a spare main board injector driver or from one of the spare outputs on the MS3X. Neither are high enough rating to drive the motor directly.

The PWM signal will typically be 13Hz, although this can be set to a lower frequency (11.1 to 78Hz) if required.

I was looking at using a solid state relay, but these do not work fast enough.

Looks like a FET will do the job but I can't find a suitable circuit.

Can someone please lead me in a suitable direction.

EDIT Wrong frequency

Edited by Paul S on 8th Nov, 2013.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

From what I gather, this is the sort of thing you need.
You apply a PWM signal, and it replicates it for the motor.
Although I think it also operates at a much higher frequency.
From what I understand, as far as inductive motors go, very high frequency is good...is 20khz or so.

http://www.ebay.co.uk/itm/High-Current-DC-...=item4abfc7ca79

http://www.ebay.co.uk/itm/Big-Current-DC-M...=item256eb716ea

9.85 @ 145mph
202mph standing mile
speed didn't kill me, but taxation probably will

Paul look at the old 2wire idle valve driver circuits in the manual. Great big TIP122? Fet can sink some decent amps and larger ones are avaliable or could be paralelled, you might need to choose a slightly less meaty motor. I would also perhaps add a smoothing cap to the output but nothing too big. Im sure thats what they used anyways, been a while:(.

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

Paul, what's your maximum current (ie if PWM reached 100%) or is that unknown because of the load on the motor being uncertain ?

Or, to play safe, what is the motor's maximum rated power (presumably marked somewhere) ?

Stevie, I may be wrong but I think the "H Bridge" controllers are specifically for bi-polar stepper motors. They may work on a normal motor as well but I know they are commonly used on stepper motors (large servo motors for the high current designs).

Depending on the current, I would try, as Denis suggests, one or more TIP122 (or 120, or 121, they are just rated for lower reverse voltages, which is only an issue if the motor windings are highly inductive) but if you need more than one it gets a bit problematical as semiconductors in DC switching applications don't share current nicely when banked together. But they are nice and cheap for a test.


EDIT - bit added.

Edited by Rod S on 9th Nov, 2013.

Schrödinger's cat - so which one am I ???

This is the starter motor from which the motor has been stripped:

http://www.powerlite-units.com/products/rac403/

It's a 1kW motor, so the maximum current is around 90 amps

However, I do not intend to load it up any where near that either by limiting the speed by duty cycle or pulleys. Let's say that I need 40 amps, although that is too much for the vehicle electrics.

I think that the H-bridge controllers would work, but I do not have the control mechanism available, just an on-off relay on the ignition and a PWM signal.

The TIP122 are rated for 5A I believe. I would need a lot of those.

Other issue is that most speed control devices using a PWM signal sit between the motor and ground. As the motor grounds through its casing then it would need to be isolated..... not impossible but not easy.

I'm currently reconsidering a solid state relay. There was originally some confusion on my part as to the frequency of the PWM signal. I thought it was 13kHz, but it is actually 13Hz, so the SSR should work. Something like this:

http://www.ebay.co.uk/itm/Solid-State-Rela...=item230ffeab18

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I'm no electronics buff, so not sure.

I just know Ive seen people use those things for motor control, fuel pump control, that sort of thing.

but I'm not 100% sure on the specifics.

9.85 @ 145mph
202mph standing mile
speed didn't kill me, but taxation probably will

Actually the starter could be the same as one of these:

http://www.powerlite-units.com/products/rac468/

1.4 kW

I have one of each on the Mini engines. The one I'm planning on using is off a Toyota but the same motor.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

13Hz sounds very low (the bigger the motor, the lower the frequency used is a general rule, but 13Hz does seem low...)

But, as 13Hz is 77mS, that SSW would work and get you to about 80% duty cycle.

Or, if you are brave, try a power MOSFET which will handle much higher frequencies.

http://uk.farnell.com/stmicroelectronics/s...-220/dp/1752167

It needs a driver circuit in addition to the MS3 output to provide it's gate current, but I've found a fairly simple one on the web if you're interested.

One other point, if it's the guts of a starter motor, it will not have a 100% duty cycle - starter motors never do, hence their power/size ratio.

But if you are expecting to run well below 1KW it should be OK.


EDIT - just noticed that particular one is out of stock at the moment but there are others similar.

Edited by Rod S on 9th Nov, 2013.

Schrödinger's cat - so which one am I ???

The MS3 can output up to 78Hz, if that helps.

I like the idea of the MOSFET driver, but I'm still unsure if I can put one on the 12v side of the motor?

I found these circuits, but they work on the ground side of the motor:

http://www.picotech.com/applications/pwm_drivers/#chap9

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

what about RC place/car drivers? they handle serious currents. Still think your a little "over motored" and a lighter motor designed for 100% duty may be a better place to start. A mobility scooter drive motor perhaps? even then i cant help but feel its a bit much but i know you will have thought long and hard about the drive power required for the pump so i should probably shoosh lol

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

The motor is bigger than I need but it is free out my bin of abandoned projects.

New smaller motors are expensive.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Edited by Rod S on 9th Nov, 2013.

Schrödinger's cat - so which one am I ???

You can pick up a recon starter including the motor for £30:

http://www.ebay.co.uk/itm/TOYOTA-COROLLA-S...=item35c8c0fe91

BUT you have to machine off the gear teeth on the end of the shaft.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

P-channel MOSFET 80A :

http://uk.rs-online.com/web/p/mosfet-transistors/4857901/

Any good?

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Automotive stuff?

http://uk.rs-online.com/web/c/semiconducto...7633D4E4F4E4526

Hope tha works, was advertised when I clicked on pauls link above.

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

Denis,

Unfortunately all the ones in your RS link are N-channel.

Totally expected as that's the way things work nowadays.

Paul, the one you found is promising.

Power dissipation low but adequate current.

Worth a try on a large heat sink and with the correct drive circuit for a P-channel (looking into that now).

Schrödinger's cat - so which one am I ???

How about this for starters:

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Forgive my ignorance but why cant you use n channel devices? The wiring difference is trivial

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

This would be how I would modify what you posted:




EDIT: You might also want to consider using a few of those FETs in parallel to lower the heat dissipation (both overall and per FET). And R2 should probably be something like 100R or maybe even lower.

Jean

Edited by jbelanger on 9th Nov, 2013.

http://www.jbperf.com/

Edited by Rod S on 10th Nov, 2013.

Schrödinger's cat - so which one am I ???

Thanks guys.

I'll update the schematic. I think I understand what's going on although electronics is not my bag.

I like the idea of two MOSFETs to help with the current and heat. Just need to work out how to mount and wire them up with 10 mm^2 cable.

Is there any thing special about the capacitors. I don't understand why I need two of them.

Any thing special about the diode? Just general purpose or something else?

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Mounting - ideally on a purpose made PCB (we can think about that later) but the main thing is to support the legs relative to the bolt/heatsink hole so, for a test, what I would suggest is a bit of veroboard with a nice slab of 1/8" (3mm) aluminium down one side - visualise the standard MS3 board and its simple heatsink but make it larger. And remove the copper from the veroboard where the bolts go through.

Then drill the adjacent holes (next to where you solder the Mosfet pins) out to a suitable size for your wires and feed them through and run them along the tracks and solder to the Mosfet pins (ie, do not rely on the copper tracks, they are far too small for the current).

But you don't need anywhere near 10mm^2 for 40A especially if shared between two Mosfets (look at the size of their pins). 6mm^2 should be fine for the short lengths involved. I would probably use 4mm^2 to the individual Mosfets and only use 6mm^2 once they are doubled up, at least for the test, maybe larger later.

The different capacitor values are for different frequencies. I assume you just took the values from the datasheet ? Bear in mind those values were for a test circuit with a specified impedance, the real motor may not be exactly the same but it's a good starting point.

The diode, I'd rather wait for Jean's reply - I assumed it was flyback protection (in which case I would put it across the motor and use a good high current general purpose one) but I think Jean has another reason for his location which I can't immediately figure out.

Schrödinger's cat - so which one am I ???

4mm would be more than ample, and with proper PWM control, current should actually be much lower as the pump is never really running at max effort

9.85 @ 145mph
202mph standing mile
speed didn't kill me, but taxation probably will

Rod, Paul,

I'm not sure what I was thinking about with the diode but you're right that it is not needed across the FET. And my copying and pasting on the original diagram made the one across the motor disappear and I did not noticed for some reason.

So the diode should be as per the original diagram. As for it's size, I'm not sure a standard 1N4001 would be sufficient. You might want to go with an FR301.

As for the base resistor, if you use one of the injectors output to drive this, you will want to omit R1 and the NPN since this will be done by the injector circuit and you can use a 10 Ohm for R2 which should be good to drive the FET. The injector driver will have no problem driving the two FETs and two 10R resistors (of a suitably high rating as mentioned). The MS3X spare outputs are also able to drive this.

Jean

http://www.jbperf.com/

working back from a 125amp draw on my starter motor at 11.5v would confirm the 1.4KW but remember that when the alternator is producing, the voltage is ~14.5v so urrent will be more like 95amps at full load. also taking into consideration the pump power absorption without the fan has been estimated at 1.5hp which is more like 75amps at 14.5v

But then you already know this :)