Author Topic: Load Resistor Causing Unwanted Voltage Shift on Comparator  (Read 1385 times)

0 Members and 1 Guest are viewing this topic.

Offline PentoadTopic starter

  • Regular Contributor
  • *
  • Posts: 56
  • Country: gb
Load Resistor Causing Unwanted Voltage Shift on Comparator
« on: November 14, 2019, 10:45:39 pm »
So I've been working on my cell balancer design and there's always been a strange problem with it where RL is acting like some kind of weird hysteresis. LED 2 is used to indicate whether the mosfet is conducting or not.  If I disconnect the load resistor and adjust the trimmer until the LED comes on and then connect the load resistor, the LED goes off and doesn't come back on unless I increase the supply voltage or adjust the trimmer. K1 4 and B+ are all connected together and are at the same voltage. I replaced the hysteresis resistor with a capacitor and this prevents the comparator switching at very high frequency causing the mosfet to overheat. It now switches at around 20hz

So I did some tests by only adjusting the supply voltage measured at B+ / K1 4.
With RL disconnected, the LED starts flickering at 3.878v and is fully on at 3.882v
With RL connected, the LED wouldn't start flickering until the supply voltage went to 3.974v

I don't understand how the presence of this load resistor on the mosfet is affecting the voltage at which the comparators are triggering because it's not directly connected to their inputs. I even measured the inputs and the resistor isn't affecting their voltage. The whole thing is set up on a breadboard currently



« Last Edit: November 14, 2019, 10:50:52 pm by Pentoad »
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Load Resistor Causing Unwanted Voltage Shift on Comparator
« Reply #1 on: November 14, 2019, 11:14:50 pm »
Try measuring the switching voltage using the B- leg of R22 as the negative reference, not the B- connection on the battery. See if it now flickers at 3.88V with the load resistor. You may need to take the B- lead from R22 back to the battery as a separate lead, if other circuit resistances are not low enough in your circuit hardware.
 

Offline PentoadTopic starter

  • Regular Contributor
  • *
  • Posts: 56
  • Country: gb
Re: Load Resistor Causing Unwanted Voltage Shift on Comparator
« Reply #2 on: November 15, 2019, 04:35:54 pm »
With the resistor disconnected I get a solid 1.2755v at the output of the potentiometer. When I connect the load resistor, the LED goes out and the voltage at the potentiometer ripples around randomly within 100mv. It doesn't affect the voltage at B- at all.
 

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15797
  • Country: fr
Re: Load Resistor Causing Unwanted Voltage Shift on Comparator
« Reply #3 on: November 15, 2019, 04:54:50 pm »
Is K1 4 connected to B+ directly? Did you look at the difference B+ - B- ? (Sorry if I state the obvious, but if the reference voltage for your potentiometer-based adjustment is the battery voltage and the battery voltage drops, will your circuit behave as expected?)
 

Offline PentoadTopic starter

  • Regular Contributor
  • *
  • Posts: 56
  • Country: gb
Re: Load Resistor Causing Unwanted Voltage Shift on Comparator
« Reply #4 on: November 15, 2019, 10:55:44 pm »
The reference is generated by a LM4041AILT-1.2. In my test setup K1 4 is directly connected to B+. Connecting or disconnecting the resistor has no effect on the reference voltage.

 

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2850
  • Country: se
  • Country: Broken Britain
Re: Load Resistor Causing Unwanted Voltage Shift on Comparator
« Reply #5 on: November 15, 2019, 10:58:27 pm »
I think R27 should be pulled higher than Vref.
Are you trying to make an oscillator. :)

If you're balancing over 1A (4.2V/3.9R), that amount of current through the GNDs could be effecting the mV hysteresis levels, you have to try and prevent the 1A flowing through the Vref and Vbatt GNDs.
« Last Edit: November 15, 2019, 11:10:55 pm by StillTrying »
.  That took much longer than I thought it would.
 

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15797
  • Country: fr
Re: Load Resistor Causing Unwanted Voltage Shift on Comparator
« Reply #6 on: November 15, 2019, 10:59:55 pm »
The reference is generated by a LM4041AILT-1.2. In my test setup K1 4 is directly connected to B+. Connecting or disconnecting the resistor has no effect on the reference voltage.

Alright for VREF, but what about the voltage at the pot's wiper?
 

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2850
  • Country: se
  • Country: Broken Britain
Re: Load Resistor Causing Unwanted Voltage Shift on Comparator
« Reply #7 on: November 15, 2019, 11:32:34 pm »
Alright for VREF, but what about the voltage at the pot's wiper?

K1 4  and B+ are the same connection, so when the battery voltage and the pot's wiper are too high the mosfet turns on to discharge it - maybe. :)
.  That took much longer than I thought it would.
 

Offline PentoadTopic starter

  • Regular Contributor
  • *
  • Posts: 56
  • Country: gb
Re: Load Resistor Causing Unwanted Voltage Shift on Comparator
« Reply #8 on: November 16, 2019, 09:49:36 am »
"K1 4  and B+ are the same connection, so when the battery voltage and the pot's wiper are too high the mosfet turns on to discharge it - maybe. "

That's the intention. I have made a video which explains the problem better.

https://youtu.be/LDTFSMxpQZ4

And here's the entire thing just to make it a bit clearer
 

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2850
  • Country: se
  • Country: Broken Britain
Re: Load Resistor Causing Unwanted Voltage Shift on Comparator
« Reply #9 on: November 16, 2019, 12:32:33 pm »
If I'm looking at the right data sheets, the shunt current through the LM4041AILT is much too high at ~30mA when Vbatt is ~4.2V.

edit:
I still can't see how the second comparator is supposed to work with R27 connected to Vref.
« Last Edit: November 16, 2019, 01:13:44 pm by StillTrying »
.  That took much longer than I thought it would.
 

Offline station240

  • Supporter
  • ****
  • Posts: 967
  • Country: au
Re: Load Resistor Causing Unwanted Voltage Shift on Comparator
« Reply #10 on: November 16, 2019, 12:54:25 pm »
Couple of things I noticed.

1. There should be an input resistor for U1D pin 10, in the same way U1A has R12.
2. I'd try disconnecting R20 from pin 13, and tie it to K14. Figure out if it's the mosfet and it's drive circuit, or the comparators causing the issues.

I looked up the datasheet for the mosfet SI3493DDV-T1-GE3 , VGS (in VDS graph) looks OK. Maybe someone else could have a look also.
 
The following users thanked this post: Pentoad

Offline ormaaj

  • Contributor
  • Posts: 20
  • Country: us
Re: Load Resistor Causing Unwanted Voltage Shift on Comparator
« Reply #11 on: November 18, 2019, 02:24:45 am »
A small cap in parallel with the zener is usually a good idea. You should be able to use a much larger resistor, or current source as I usually do. A cheap tl431 would do the job too.

edit: oh lm4041is a reference. Just a bad symbol.
« Last Edit: November 18, 2019, 03:43:05 am by ormaaj »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf