can anyone help me with building an electric motor over-torque protection circuit?

Hi,
I have a electric motor controlled valve in my truck's exhaust which on the press-hold of a 3-way switch either opens or closes.
to fully open, you hold the switch down for 7 seconds.
to fully close, you hold the switch up for 7 seconds.

The problem I have is that if you inadvertently hold the switch for longer, the gears grind and break.

I am trying to find an over-torque circuit that I could use to sense the "full open" or "full closed" position and kill the power.

I would have thought that this is a common requirement, but it seems not.

I was hoping a current sensor could be used to detect the over-torque?

The hostility of the environment (under a truck, a part of a very hot exhaust) and a sealed motor precludes anything fancy on the motor shaft for position sensing etc.

Any help would be gratefully appreciated!

Cheers,
QB

Sensing the current across a remote resistor is the easy
way to go.

If this is for an electric cutout, the ones I've seen add a
simple current limiting resistor so that a damaging current
/ torque cannot be applied. However this makes them
slower-acting, and eventually unable to make enough
torque to overcome gear bind, gum and friction.

The better way is to use less resistance, sense I*R and
interrupt the source. You probably want an initial blanking
period as startup torque will be high and has to be left
alone if you want to get going.

But such gizmos tend to lead short, unhappy lives and
no amount of electrical protection makes up for poor and
infrequent maintenance of the internals. Take one apart
and you may even find thermoplastic gears (in an exhaust
part - great plan).

dick_freebird said:

Sensing the current across a remote resistor is the easy
way to go.

If this is for an electric cutout, the ones I've seen add a
simple current limiting resistor so that a damaging current
/ torque cannot be applied. However this makes them
slower-acting, and eventually unable to make enough
torque to overcome gear bind, gum and friction.

The better way is to use less resistance, sense I*R and
interrupt the source. You probably want an initial blanking
period as startup torque will be high and has to be left
alone if you want to get going.

But such gizmos tend to lead short, unhappy lives and
no amount of electrical protection makes up for poor and
infrequent maintenance of the internals. Take one apart
and you may even find thermoplastic gears (in an exhaust
part - great plan).

Click to expand...

Thanks, yes, this is an electric cutout.

Regarding your recommendation on the "better way to go", can you please point me in the right direction?
Would that be a microPIC connected to something(s)?

And yes, the gearbox is not very encouraging. However, the fun.is.awesome. I live in Scotland, and to be able to (effectively) run open headers on a 1956 F100 with a 302 V8 is simply fantastic. Just can't stop smiling -and the variability enables some control.

I'm 42, and haven't had so much fun in years.

Anyway, before I break the second one, can you (someone) help with the "better way to go"?

Many many many thanks.

I'd start with making the control logic based. You have
three states (closing, opening, where-it-sits). I would
make two SRFFs controlling two DPDT relays.

"Opening" switch event should set the "opening" 'flop
and reset the "Closing" flop. And vice versa.

Overcurrent should reset both (you don't care).

On power-up, "Closing" probably wants energized and
"Opening" left alone. Just so as not to annoy the neighbors
any worse.

Would you prefer to do it with discrete transistors,
with "glue logic", with a uC? (I consider that much brain
power a waste, but it could cost no more than other
ways especially if your junk box is well stocked.

Now I've had so much good luck with cutouts, that
I've been working on a solenoid-actuated one. So far
I haven't gotten it entirely quiet at rest; gear reduction
does give you a pretty authoritative sealing, unlike a
clockspring. In this case actuation is only momentary
(a WOT switch) which is good enough for the purpose.

You might like WOT as an input to the "Opening" term.

Stall detection and motor shut-off is a standard feature of electric windows. Besides requiring a current measure feature (shunt, sensing transistor saturation voltage), controller circuit and suitable electronic switch, it depends on certain gear properties. There most be a sufficient margin between operation and breaking torque.

what about a limit switch? when valve open a limit switch is engaged and power gets cut off via small contactor?

$0.02

Mr.Cool

One way to go is to use a .1 ohm current sensing resistor, this would still give you a 12V/.1 = 120A of current available. Then an op amp to increase the voltage change a bit and more importantly to put a small delay and a voltage offset so that the amps output does not change until the current = X amps. As the amps output changes use this to bias off a series pass transistor which you have remembered to put in series with the motor's feed.
Or build two bi-stables, trigger both with an "ON" switch, one then drives a relay for up/down the other runs a 7 second timer which fires a relay to switch off the motor and resets the bistable. Pushing the "OFF", repeats the above but for the 7 second timer only. So one relay does an on/off to the motor, the other reverses its direction.
Frank

dick_freebird said:

I'd start with making the control logic based. You have
three states (closing, opening, where-it-sits). I would
make two SRFFs controlling two DPDT relays.

"Opening" switch event should set the "opening" 'flop
and reset the "Closing" flop. And vice versa.

Overcurrent should reset both (you don't care).

On power-up, "Closing" probably wants energized and
"Opening" left alone. Just so as not to annoy the neighbors
any worse.

Would you prefer to do it with discrete transistors,
with "glue logic", with a uC? (I consider that much brain
power a waste, but it could cost no more than other
ways especially if your junk box is well stocked.

Now I've had so much good luck with cutouts, that
I've been working on a solenoid-actuated one. So far
I haven't gotten it entirely quiet at rest; gear reduction
does give you a pretty authoritative sealing, unlike a
clockspring. In this case actuation is only momentary
(a WOT switch) which is good enough for the purpose.

You might like WOT as an input to the "Opening" term.

Click to expand...

I assume SRFFs are switching roller finger followers? And I do not have any clue which is better: uC or another way.
Whatever is easiest to source/make. My junk box is empty of electronics. Cost is not a problem. In fact, I'd happily pay someone to do it.
Back to the SRFFs --Am I right in guessing you suggest I build my own SRFFs ie glue a lump onto the shaft and use something like a two microswitches to detect open and closed states? Or can you direct me to a picture of what you are thinking of? Everything would need to be pretty tiny to fit.

I am guessing you are using WOT (Wide Open Throttle) switches to open the flanges only when the throttle is "wide open". In my case, I like the manual control, which enables me to "tune" the exhaust to fit the circumstances.

Again, for me, I don't mind if it is not totally sealed, as I have a very quiet exhaust at the best of times, so a little noise is exactly what I want.

again, many many thanks.

- - - Updated - - -

Mr.Cool said:

what about a limit switch? when valve open a limit switch is engaged and power gets cut off via small contactor?

$0.02

Mr.Cool

Click to expand...

I am guessing the problem here is the hostility of the environment --it is very very hot with high speed gas pushing through.
But, maybe I could weld / glue something tiny. Hmmmm.

I was hoping I could simply(!) add something in series with the Live or earth cable and hey presto!, no over torque....

- - - Updated - - -

FvM said:

Stall detection and motor shut-off is a standard feature of electric windows. Besides requiring a current measure feature (shunt, sensing transistor saturation voltage), controller circuit and suitable electronic switch, it depends on certain gear properties. There most be a sufficient margin between operation and breaking torque.

Click to expand...

So, are you suggesting I find an electric window motor and use it? I like that idea. I hope the "over torque" circuits are in the motor end. Maybe I should pop over to a local scrap yard to see if I can obtain a couple.

- - - Updated - - -

chuckey said:

One way to go is to use a .1 ohm current sensing resistor, this would still give you a 12V/.1 = 120A of current available. Then an op amp to increase the voltage change a bit and more importantly to put a small delay and a voltage offset so that the amps output does not change until the current = X amps. As the amps output changes use this to bias off a series pass transistor which you have remembered to put in series with the motor's feed.
Or build two bi-stables, trigger both with an "ON" switch, one then drives a relay for up/down the other runs a 7 second timer which fires a relay to switch off the motor and resets the bistable. Pushing the "OFF", repeats the above but for the 7 second timer only. So one relay does an on/off to the motor, the other reverses its direction.
Frank

Click to expand...

Thanks Frank. I like your first suggestion, as it sounds like the "on the wire" solution I seek. I also like the second idea, but I want to be able to open/close to any point -ie not just full open or full close, so a timer wouldn't work.

back to the first idea. I guess I can source a 0.1 ohm current sensing resistor. Is this right: https://uk.farnell.com/vishay-dale/wsl2512r1000fek/current-sense-resistor/dp/1653235

Next, I am not sure what a "op amp" is --I guess an Operational Amplifier, I guess again, that's something I can buy "off the shelf". Is this right: https://uk.rs-online.com/web/p/operational-amplifiers/6606755/

Then you suggest a series pass transistor. That is something that currently eludes me. Any hints what I should be looking for?

Are these all things that I can easily purchase and "wire up" together?

Many many thanks

- - - Updated - - -


*** What about a resettable polyfuse? would that work?

A limit switch faces the problem of very small angular difference
between bound-up / broken gears, and unacceptable rattling
of the butterfly valve.

SRFF is a set-reset flip-flop. The simplest bistable element
that can be turned off as easily as on. You can make it
with discrete BJTs, use a logic IC or whatever.