DC Battery amps vs. Motor Amps

General topics and discussions about the VESC and its development.
maxkinz
Posts: 29
Joined: 30 Jun 2017, 14:41

Re: DC Battery amps vs. Motor Amps

Postby maxkinz » 07 Jul 2017, 08:46

Hey devin,

Thanks for the suggestion.
I am busy with other stuff today but I will go make some more experiments especially measure the voltage rails whenever I get to it. KB control does not influence the results. As the screenshot of the realtime graph shows the current is supposed to be 3.5 motor amps - translating to roughly 0.5 Battery amps. That is at least what should be happening which is why I was at a loss to explain the measurements in the current plot. I will retry using a battery later as well.
Thank you guys for all the help.

devin
Posts: 255
Joined: 08 May 2017, 01:55
Location: San Francisco, California, US

Re: DC Battery amps vs. Motor Amps

Postby devin » 07 Jul 2017, 09:40

maxkiz wrote:Hey devin,

Thanks for the suggestion.
I am busy with other stuff today but I will go make some more experiments especially measure the voltage rails whenever I get to it. KB control does not influence the results. As the screenshot of the realtime graph shows the current is supposed to be 3.5 motor amps - translating to roughly 0.5 Battery amps. That is at least what should be happening which is why I was at a loss to explain the measurements in the current plot. I will retry using a battery later as well.
Thank you guys for all the help.


This is just a theory:

Maybe the "current plot" graph shows a graph of what the VESC was "intending to happen" in advance of the fact based on the best information available in advance at the time-- 10v voltage measurements and current measurements etc, but it doesn't show what really happened after the fact if the lab supply was dipping the voltage wildly to stay under 3a at all times while the VESC was attempting to draw 8.88A, 33.78% of the time -- 3 battery amps average.

Image

By extension of logic perhaps the "Real Time Data" graph shows what really happened after the fact-- less battery amps and less motor amps than were planned in advance, since possibly the lab supply voltage was dipping wildly....

Image

Simply maybe the VESC was "intending" to draw 3 battery amps in advance (first graph), but actually drew 0.4 battery amps after the fact (second graph)... and maybe the vesc was intending to supply 14 motor amps (first graph), but it actually supplied about 3.4 motor amps after the fact (second graph)... due to a wildly dipping supply voltage... perhaps.

maxkinz
Posts: 29
Joined: 30 Jun 2017, 14:41

Re: DC Battery amps vs. Motor Amps

Postby maxkinz » 10 Jul 2017, 10:24

Hey Devin,
Thanks for the work.
I will wait for my other VESC to arrive so I can do head to head comparisons - no value in guessing for me at this point - if It turns out the behaviour is specific to this VESC (which I think it will be) I know for sure that something is broken and will try and repair (probably replace DRV). If not I will investigate further.

Thank you again for all the effort !

rew
Posts: 943
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.

Re: DC Battery amps vs. Motor Amps

Postby rew » 10 Jul 2017, 13:23

Just for your information, Devin seems to live in a universe where the rules of physics are slightly different.

You should not take for a fact when he says the VESC will plot the current it intended to push through the motor. What you see in the graph is what was actually measured. Sometimes when e.g. a sudden load change happens, you can see the "WHOA need to adjust things, the current is much more than requested". THis is then a transient and it happens pretty quickly, but if you see a steady "3A" line, that is what the VESC measured.

Of course you can fool it. If you mount say 2mOhm resistors instead of 1mOhm resistors, then the 3mV of measured voltage does not correspond to 3A, but only to 1.5A. So in that case, unless you confess to the VESC that you modfiied the resistors (= recompile with the right settings), the indicated current will be wrong.

devin
Posts: 255
Joined: 08 May 2017, 01:55
Location: San Francisco, California, US

Re: DC Battery amps vs. Motor Amps

Postby devin » 10 Jul 2017, 13:49

rew wrote:Just for your information, Devin seems to live in a universe where the rules of physics are slightly different.

You should not take for a fact when he says the VESC will plot the current it intended to push through the motor. What you see in the graph is what was actually measured. Sometimes when e.g. a sudden load change happens, you can see the "WHOA need to adjust things, the current is much more than requested". THis is then a transient and it happens pretty quickly, but if you see a steady "3A" line, that is what the VESC measured.

Of course you can fool it. If you mount say 2mOhm resistors instead of 1mOhm resistors, then the 3mV of measured voltage does not correspond to 3A, but only to 1.5A. So in that case, unless you confess to the VESC that you modfiied the resistors (= recompile with the right settings), the indicated current will be wrong.


@rew I'm living in a universe where the VESC "Current Plot" graph shows amps through 3 phases when there's only 2 wires leading to his DC motor... & the Realtime Plot shows completely different numbers from the Current Plot, even though he tested this very VESC as still being capable of normally driving a separate BLDC motor after his "incident" which shouldnt be possible if the VESC's current sensor was destroyed.

Image

Image

rew
Posts: 943
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.

Re: DC Battery amps vs. Motor Amps

Postby rew » 10 Jul 2017, 17:25

The VESC (4.x hardware) has only 2 current sense channels. For BLCD and FOC you need to know all three currents. However there is a law in electrical engineering: all currents in a node add up to zero. And this holds even when you expand "node" to black box. So if we make the motor a black box, and we see 1A going in on one wire, 2A going in on the second wire, then we know there must be -3A going into the third wire (a negative amount going in equals current coming out)....

In DC motor mode, the two sensors are mounted on the two wires going into the DC motor. So in theory, if you measure 2.3A going into lead A, you should measure -2.3A going into lead C (i.e. that 2.3A coming out).

Imperfections make that you don't always exactly measure perfectly. This means that the third phase (Ib) that is calculated from Ia + Ib + Ic = 0 does not always end up being exactly zero. If you're seeing something "constant and not close to zero", then the calibration is off by a bit. (I wrote a terminal command dc_cal to re-calibrate it)

rew
Posts: 943
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.

Re: DC Battery amps vs. Motor Amps

Postby rew » 10 Jul 2017, 18:02

I have patched my screwdriver. I now have both motor leads and the original drive lead outside of the screwdriver. I measured the current, and it went above 20A that my multimeter can handle. So it should be reasonably safe to tell the VESC that the motor can handle 20A.

There is quite a lot of friction in the gearbox: The motor requires 5A just to run at top speed.

P.S. The charger is not very good. Leave your battery on the charger for a month and it (the battery) will be dead. Try to remove the battery from the charger as soon as the light turns green.

maxkinz
Posts: 29
Joined: 30 Jun 2017, 14:41

Re: DC Battery amps vs. Motor Amps

Postby maxkinz » 11 Jul 2017, 08:23

Hey rew,
Thanks for the update ! I know about the friction in the planetary. My drill requires about the same to get to top speed. I was going to subtract whatever current I need to get to a certain speed before calculating the torque from the "residual" current. I know about the charger and I didn't use it at all - got an iMax B6 here to fullfill most of my charging needs. There seems to be some electronics built into the battery (didn't open it) as when running off of that battery the VESC looses power after a while - do you observe this, too ? it repowers after reconnecting the battery. This does not happen on the lab supply.

devin
Posts: 255
Joined: 08 May 2017, 01:55
Location: San Francisco, California, US

Re: DC Battery amps vs. Motor Amps

Postby devin » 11 Jul 2017, 14:14

maxkinz wrote:Hey rew,
Thanks for the update ! I know about the friction in the planetary. My drill requires about the same to get to top speed. I was going to subtract whatever current I need to get to a certain speed before calculating the torque from the "residual" current. I know about the charger and I didn't use it at all - got an iMax B6 here to fullfill most of my charging needs. There seems to be some electronics built into the battery (didn't open it) as when running off of that battery the VESC looses power after a while - do you observe this, too ? it repowers after reconnecting the battery. This does not happen on the lab supply.


@maxkinz -- are you saying the VESC now appears to be functioning normally? if so do you have any more current plots (showing all "3 phases" while in dc mode) vs realtime data plots of the vesc operating in dc mode while your motor is stalled on the torque sensor? (in particular with no 3a current limit on your 10v lab supply - or using a 12v battery with no battery management system) and can we see a screenshot of your batt/motor/absolute amp limit settings?

rew
Posts: 943
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.

Re: DC Battery amps vs. Motor Amps

Postby rew » 12 Jul 2017, 06:00

maxkinz wrote: There seems to be some electronics built into the battery (didn't open it) as when running off of that battery the VESC looses power after a while - do you observe this, too ?
I did not run the VESC from the battery. I've modified my drill to have three wires coming out. Motor black, motor-red and original-controller-red. So I connect both red wires to use the original controller, or connect both motor wires to a VESC.

A battery module like that needs to have an "Over discharge" protection. Hmm. If it has a balancing circuit too, then the residual current at 12.6V may be "eaten" by the balancing circuit.


Return to “General”

Who is online

Users browsing this forum: Bing [Bot] and 5 guests