Voltage regulator no juice

knurf

100 W
Joined
Aug 28, 2016
Messages
129
I have a pretty powerful mid drive ebike in the 2-4kw range. I just tried to attach this step down regulator thing https://www.aliexpress.com/item/Hig...630.html?spm=a2g0s.9042311.0.0.5e504c4dly9FG1 in parallel to drive some LED lights in the 6-8w range but the results are not what I hoped for.. Under medium acceleration the lights fade out to nothing. I know the battery can supply way more current than this (14s lipo / 16ah) I only drew a max of 25 amps while testing. Why can't I get the juice to go around?
 
If it didn't affect the accleration, then the DC-DC probably either can't handle the voltage sag on the input (only works with a steady voltage at the input), or it isn't adjusted or set for the correct input voltage / voltage range.
 
Hm yeah that sounds reasonable. There was no difference in acceleration, it just choked out the low power stuff I had connected. Once going, and keeping the power low enough it worked.
I'll try to dig out some other 12v step down thing, never used the module for this application before so it's a likely suspect.
 
knurf said:
Hm yeah that sounds reasonable. There was no difference in acceleration, it just choked out the low power stuff I had connected. Once going, and keeping the power low enough it worked.
I'll try to dig out some other 12v step down thing, never used the module for this application before so it's a likely suspect.

I'm running this one on a 36 volt battery to run my 12 volt stuff. I pull about 40 watts at night and it never fluctuates even when I'm pulling about 1000 watts from the battery. It is also weather proof (well - maybe not a lightning strike :^))

https://www.amazon.com/gp/product/B0151BF6DI/ref=oh_aui_detailpage_o07_s01?ie=UTF8&psc=1
 
I just tried plugging in something very similar to that, but rated 12v 20a. Didn't help much at all. I plugged it in before the original regulator, so that it would receive 12v from the new one (maybe this is a bad idea), this was to avoid having to remove wiring and other inconveniences.

Could it be bad/wrong wiring? I just used some scrap I had laying around, now I'm staring to think it was too thin or something. But it only needs to handle 5-10 watts or so. And there's not a huge amount of voltage sag when I draw 1-2kw from the battery maybe 3 volts, thought any regulator worth its name would handle that..

It's more or less exactly the same thing that happens when starting an older car - stereo and head lights die while the starter cranks. Maybe my battery isn't as strong as I thought.
 
knurf said:
It's more or less exactly the same thing that happens when starting an older car - stereo and head lights die while the starter cranks. Maybe my battery isn't as strong as I thought.

Simpler is better, wire it direct. Also, put a voltmeter on your batteries to see how much they are sagging under load.
 
knurf said:
I plugged it in before the original regulator, so that it would receive 12v from the new one (maybe this is a bad idea), this was to avoid having to remove wiring and other inconveniences.
If you mean that it is connected to the one that has a problem, then the results are unpredictable, and causes of whatever behavior you get are unknown.

To troubleshoot things, you ahve to remove each potential cause of a problem, to ensure it's eliminated.

So I'd remove the problematic one, and use *only* the new one, and see what happens.

(in theory, if the new one regulates the power sufficiently that there's no sag on it's output, then the problematic one shouldn't hvave a problem anymore, if it's issue was caused by voltage sag, so your idea *should* have worked...but since we don't relaly know why it didn't work to start with, we also don't know why it did'n't work this way).



Could it be bad/wrong wiring? I just used some scrap I had laying around, now I'm staring to think it was too thin or something. But it only needs to handle 5-10 watts or so.

How thin is is it? Remember--wiring isn't about watts, its' about amps. Since the input of the converter is at a high voltage, the current needed to make those watts is significantly lower than it is at the output, so it shouldn't take much of a wire to handle it. :)

And there's not a huge amount of voltage sag when I draw 1-2kw from the battery maybe 3 volts, thought any regulator worth its name would handle that..
It should...but they're not all designed the same, and cheap converters may simply pass any differences in voltage thru to the output. So a 3v sag on the input could get translated directly to a similar sag on the output, and that's a major loss at only 12v starting voltage.


Another possibility is a bad ground, and when more battery power is used, the ground is "saturated" (not a proper term, but...) and can't handle the current from the DC-DC at the same time as the increased current under motor load. This would mean no matter which DC-DC you use, you'd get the same result. What happens in this case is that the ground voltage rises with current thru it, relative to the output voltage, and so the output voltage "appears" to drop (causing dimming, etc). Effectively it's teh same thing as having voltage sag, but it's a different cause, so a different fix.



Check what the voltage ouptut of the DC-DC is under motor load and no motor load conditions, measuring across the DC-DC output wires.

Then check votlage from the DC-DC output *ground* wire to the battery negative is under the same conditions.
 
Interesting points, thanks! All forms of measuring are made a bit complicated, by the fact that i need a real load on the bike for this to happen. I do have a BMS that monitors max/min voltage at least but the regulated voltage would be very interesting to get a hint of.

But that ground problem.. I drive a LED (with its own regulator), a simple ohm meter (for monitoring a thermistor) and a atmega8 which is connected back to the ESC via the throttle signal and ground. I've always thought that stuff should share a common ground, even/especially when powered from different sources (I powered the atmega thing with a separate battery before). I'm gonna try take the ESC thing out of the equation first, see if that connection interferes. Also easy to try not connecting the regulated ground to the ESC, I guess.
 
knurf said:
But that ground problem.. I drive a LED (with its own regulator), a simple ohm meter (for monitoring a thermistor) and a atmega8 which is connected back to the ESC via the throttle signal and ground. I've always thought that stuff should share a common ground, even/especially when powered from different sources (I powered the atmega thing with a separate battery before). I'm gonna try take the ESC thing out of the equation first, see if that connection interferes. Also easy to try not connecting the regulated ground to the ESC, I guess.

I keep the grounds (negative leads) separate for each voltage I'm running. So the 48v, 36v, 12v and 5v circuits are all independent once the voltage conversion has happened.
 
I keep the grounds (negative leads) separate for each voltage I'm running. So the 48v, 36v, 12v and 5v circuits are all independent once the voltage conversion has happened.
Yes, that appears to have been my problem! I disconnected the regulated ground from the ESC and now the lights don't even flinch at max load or uphill starts, and that is with the original cheapo regulator. Lesson learned, keep regulated grounds separated I guess. Thanks all for the input/help!
 
knurf said:
I keep the grounds (negative leads) separate for each voltage I'm running. So the 48v, 36v, 12v and 5v circuits are all independent once the voltage conversion has happened.
Yes, that appears to have been my problem! I disconnected the regulated ground from the ESC and now the lights don't even flinch at max load or uphill starts, and that is with the original cheapo regulator. Lesson learned, keep regulated grounds separated I guess. Thanks all for the input/help!

Cool. Looks like Amberwolf got it pretty much right.
 
Back
Top