PDA

View Full Version : Instant full throttle bad for controller?



teddillard
21 March 2011, 1452
So. Say I had an on-off switch, ohhhh, say it was on a lever where the clutch might normally go, and, let's say it went to the throttle. That is, it disconnects the throttle input to the controller. (mmm, as I'm typing this, I'm thinking this may just error the controller out... will it?)

So then, say you pulled the "clutch" lever (switch) and cranked the throttle to full-on. Then you dump the clutch, closing the throttle circuit.

Would this be a bad thing for the controller? Instant full-throttle?

These are the things I think about...

DaveAK
21 March 2011, 1458
Most controllers won't turn on with the throttle open, (or at least won't put out juice to the motor), but you're not asking that are you? You're saying the controller will already be on, right?

Richard230
21 March 2011, 1502
My uninformed layperson's guess is that it would be bad for the batteries - maybe the motor, too.

Nuts & Volts
21 March 2011, 1512
If the controller is on and the caps inside it are charged i think it will be OK to send it a full throttle signal. You may not get the effect you are expecting because of the controller throttle curve. Even with a WOT input to the controller it may still have a time delay to ramp up the controller current (this would help keep the caps and switching devices from blowing up). Thats my thought, but someone with more expertise could give you a better answer.

The simplest "boost" button you are describing would be to bypass the controller with a contactor from battery straight to motor, but you could damage things then. Some DC drag racer guys do this off the line from what, i've read.

podolefsky
21 March 2011, 1621
Like N&V was saying, my Alltrax has a throttle delay. It ramps up to the throttle point you set. It can be adjusted from 1-15, with 15 being fastest...but I'm not sure if 15 is instant, or just a very short delay. Right now I have it set to 12.

If I understand correctly, a DC controller basically uses pulse width modulation (PWM) to vary the power to the motor. At 100% throttle, it is at 100% duty cycle: a flat, full voltage signal to the motor. The motor will then pull as much current as it can.

So just to nail down this question more specifically, it seems like you're asking whether switching the controller to 100% duty cycle instantly is bad. I guess a follow up question would be: even if you did this, would the controller actually be asked to produce full voltage and full current at this instant? I think the answer is no - even if you place full voltage across the motor terminals, I believe the motor coils will have an inductance that will delay the current. So then the question is: is the built in delay due to the motor inductance less than the time it would take to damage the controller, assuming the controller does not have a built in safety delay?

teddillard
21 March 2011, 1633
The simplest "boost" button you are describing would be to bypass the controller with a contactor from battery straight to motor, but you could damage things then. Some DC drag racer guys do this off the line from what, i've read.

Yes, this is what Ed lovingly refers to as "electric nitrous". :O

SplinterOz
21 March 2011, 1934
My take...

Depends on the motor/controller.
I have a PMBLDC motor... that means the controller is actually an Electronic Speed Controller (ESC). What this means is the throttle input is not a request for power but a request for speed. If I turn my throttle to 50% the controller will do its best to apply all the power/torque required to get me to 50% speed.

This is why some of the controllers have a delay... to stop you pulling your arms out of your sockets

Nuts & Volts
21 March 2011, 2048
My take...

Depends on the motor/controller.
I have a PMBLDC motor... that means the controller is actually an Electronic Speed Controller (ESC). What this means is the throttle input is not a request for power but a request for speed. If I turn my throttle to 50% the controller will do its best to apply all the power/torque required to get me to 50% speed.

This is why some of the controllers have a delay... to stop you pulling your arms out of your sockets

While this is true a lot of controllers also allow torque control which would come down to a set amount of current controlled by the controller. So WOT would be say 300A through the motor at that instant. So I think in torque control a "boost" 100% throttle signal should not hurt the controller if max throttle=max current limit of controller. Now there still may be a time delay, but I think it would be less than in speed control because this relies motor on both current and voltage simultaneously. This make any sense?? I feel like im missing something here thou

podolefsky
21 March 2011, 2102
Go for it...what's the worst that could happen...

(Note that motor explodes at the *end*, which means 100% throttle at the *beginning* = perfectly safe.)


http://www.youtube.com/watch?v=3dRpAZci9m0

ZoomSmith
21 March 2011, 2109
I for one, am really looking forward to this video. You might want to wear some padding in the backside of your jumpsuit Ted.

SplinterOz
21 March 2011, 2150
N&V agreed. BTW do you know what controllers are Torque controllers... I know my Infineon based one is not. I think the Sevcon ones are (but they are REALLY expensive)

podolefsky
21 March 2011, 2223
As far as I know, most PWM controllers for DC motors are speed control because they use PWM set the motor voltage (which is proportional to RPM). Sevcon sepex and Curtis AC controllers can be set for torque or speed control. I don't know about others.

But those names are kind of misleading. If you have a PWM controller, it is setting voltage. But if you increase the voltage, this will cause more current (hence torque). So the further you twist the throttle, the more current you get. It feels effectively like torque control when accelerating, going up a hill, any time you need more power.

Nuts & Volts
21 March 2011, 2254
The Sevcon can be programmed to be either speed or torque controlled. I think Kelly can be programmed for speed, balanced, or torque control. Looks like the PG Sigmadrive are speed controlled. Alltrax is speed control according to someone on ES. Not sure I know any other common controller??

If its more twitchy it is most likely speed controlled. A smoother power increase compared to a throttle change will most likely be torque controlled.

Nuts & Volts
21 March 2011, 2322
The Sevcon can be programmed to be either speed or torque controlled. I think Kelly can be programmed for speed, balanced, or torque control. Looks like the PG Sigmadrive are speed controlled. Alltrax is speed control according to someone on ES. Not sure I know any other common controller??

If its more twitchy it is most likely speed controlled. A smoother power increase compared to a throttle change will most likely be torque controlled.

podolefsky
21 March 2011, 2328
Curtis DC controllers? All I know about them is they're expensive...

EVcycle
22 March 2011, 0219
Yes, this is what Ed lovingly refers to as "electric nitrous". :O

Yes, and it works too well.

You just need to have enough battery to keep things happy....but it does work. :O

Schematic upon request.

teddillard
22 March 2011, 0402
Cool, thanks for the info. The first point of doing this is to give you another low-voltage cutoff* where you want it, which for me is the clutch lever. Seems I'm always grabbing for it to control especially low speed stuff, and mostly only when I'm in danger of losing balance or control... spaz out, grab the clutch, kills the throttle, release and you're back in business. Also, it seems like it wouldn't hurt to have another kill switch in a different place on the circuit... Then, of course, the idea got transmorphed... just wondering if it would damage anything.

I think if the idea was to get full voltage instantly then Ed's idea would be the way to go. But if I'm doing it primarily to be safe, and can beat on it occasionally with hilarious result, without damage, then I think it's a keeper... can I call it a Suicide Clutch? :D

*bad choice of words. Let's say, a low-voltage kill switch, to avoid confusion with an LVC.

EVcycle
22 March 2011, 0517
When we used the full power setup, we did NOT start off a run with it that way (full power) .
There was too much torque causing way too many sideways "butt pucker" events.
We ran a adjustable delay that was triggered from a switch that was mounted behind the thumb throttle.
We started with a delay of a 3 seconds and ended up with a 1/2 second which was good for the drag strip.
Not sure I would ever try it on the street at any delay. :O

mpipes
22 March 2011, 0543
Suicide Clutch has a nice ring to it! :D