Author Topic: Voltage reference not working properly #SOLVED#  (Read 7331 times)

0 Members and 1 Guest are viewing this topic.

Offline iMo

  • Super Contributor
  • ***
  • Posts: 5570
  • Country: va
Re: Voltage reference not working properly
« Reply #25 on: March 02, 2019, 05:14:27 pm »
Above my hint on an external -5V (hard one, ie an external floating 5V source), and you may pass 20mA via the 431 (125ohm) during the troubleshooting..
The LM431 voltage must be 2.5V +/-0.05V always.
« Last Edit: March 02, 2019, 05:17:00 pm by imo »
Readers discretion is advised..
 

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: Voltage reference not working properly
« Reply #26 on: March 02, 2019, 05:16:06 pm »
I still have a 2.2k resistor on the 431, If I change it the output voltage changes as well, I'll try it though and update you on the results.
 

Offline iMo

  • Super Contributor
  • ***
  • Posts: 5570
  • Country: va
Re: Voltage reference not working properly
« Reply #27 on: March 02, 2019, 05:18:09 pm »
And doublecheck (measure!) all 40 resistor's values on your pcb..

BTW, what is the value of R18? I would remove it during the troubleshooting.

Also doublecheck the polarity of the capacitors in your LT1046 -5V generator. I would also try to load it with a resistor to get an idea how much current it gives you actually while having -4.88V on it.

« Last Edit: March 02, 2019, 05:40:08 pm by imo »
Readers discretion is advised..
 

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2850
  • Country: se
  • Country: Broken Britain
Re: Voltage reference not working properly
« Reply #28 on: March 02, 2019, 05:39:03 pm »
"I still have a 2.2k resistor on the 431"

I've had to change my previous calculation for R1, I was adding the OPA277's -5V current into the -2.5V.
https://www.eevblog.com/forum/beginners/voltage-reference-not-working-properly/msg2235837/#msg2235837

Now (4.8-2.55)/1.66 = 1.4k or slightly lower for R1.
« Last Edit: March 02, 2019, 05:42:22 pm by StillTrying »
.  That took much longer than I thought it would.
 

Offline iMo

  • Super Contributor
  • ***
  • Posts: 5570
  • Country: va
Re: Voltage reference not working properly
« Reply #29 on: March 02, 2019, 05:44:55 pm »
..
3. it could be at a specific R29 position the whole stuff starts to oscillate wildly (look at it with o'scope). ..
Did you check this?

Read the previous posts and do make a checklist :)
« Last Edit: March 02, 2019, 05:46:59 pm by imo »
Readers discretion is advised..
 

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: Voltage reference not working properly
« Reply #30 on: March 02, 2019, 06:02:59 pm »
@StillTrying So I found out I had the LM431 wired wrong  :palm: on the breadboard and after I fixed it and added the 1.4k resistor to the cathode the output is still -2.5v when it isn't connected to the load. But now when I do connect it to the -2.5v rail on the PCB the output drops to -0.72v and the output on the REF5040 drops to -0.7v as well. The LM431 also gets warm to the touch and I am now unable to set the current null calibration. Previously I would be able to set it to 0mA which is what I need to do but now it can only go down to 13mA. BTW, current null just means the leakage current when the load is turned off.

@IMO I did and there is nothing out of the ordinary, I am gonna write a quick checklist and go through this entire post thread to look for anything I have missed
« Last Edit: March 02, 2019, 06:05:05 pm by FotatoPotato »
 

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2850
  • Country: se
  • Country: Broken Britain
Re: Voltage reference not working properly
« Reply #31 on: March 02, 2019, 06:11:37 pm »
"But now when I do connect it to the -2.5v rail on the PCB the output drops to -0.72v"

You must have the 431's anode and cathode reversed, have a look a its internal diagram, it's a 2.5V zener the right way, and just a normal diode the wrong way.
.  That took much longer than I thought it would.
 

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: Voltage reference not working properly
« Reply #32 on: March 02, 2019, 06:39:29 pm »
@IMO The checklist was a fantastic Idea!, I was stupid not to do it earlier. I have fixed some of the problems now. First The 431 was wired completely wrong  :palm: I wired it back to what the original was in my schematic and It output -2.506v with a 1.3k resistor for R1. The LTC1046 now outputs a steady -4.96v no matter what. Also, the REF5040 outputs a stable 4.0962v perfectly. The only problem that still persists is that when I try to adjust the current null, while the 4.096v stays stable, the -2.5v goes up to 0.9v. This is an improvement to the plethora of issues that I had earlier but its still not perfect. Also, the current readout on the Load is terribly wrong. I'll send some pics of the discrepancies between the current readout on my load and the actual current it draws.

If you look at the pic of the LCD the set current is 0.1A, it is reporting to draw 0.132A and the actual current being drawn as shown by the 34401A is 0.93A.
 

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2850
  • Country: se
  • Country: Broken Britain
Re: Voltage reference not working properly
« Reply #33 on: March 02, 2019, 07:24:12 pm »
I'm still suspicious of the OPA277, are you sure it's powered from -5v and not the -2.5V. :)

"the -2.5v goes up to 0.9v."

Are you sure it's +0.9V, the OPA277 could possibly pull the -5V and -2.5V up to about -0.9V.
+0.9V on the -2.5V ref. should be impossible, because of the LM341's internal reverse diode.
.  That took much longer than I thought it would.
 

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: Voltage reference not working properly
« Reply #34 on: March 02, 2019, 07:31:28 pm »
I just checked and the OPA277 is getting all of the correct voltages. And yes, I just checked again and the output of the 431 is 0.9v

I think something is overloading the 431 and pulling its output High but I may be wrong, I just checked the current draw on the -2.5v rail and it is 4mA, so I don't think it is being overloaded but hey, I have been wrong plenty of times before.
 

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2850
  • Country: se
  • Country: Broken Britain
Re: Voltage reference not working properly
« Reply #35 on: March 02, 2019, 07:36:55 pm »
"I just checked the current draw on the -2.5v rail and it is 4mA"

What does the -2.5V supply, as far I know it's just VR29 10k = 0.66mA.

Are you sure the pin out of VR29 is correct on the PCB, the 4.096V 10mA ref. could easily pull the -2.5V ref. up to ~ +0.8V.

I'm running out of ideas on what's wrong with the -2.5V.
« Last Edit: March 02, 2019, 07:50:32 pm by StillTrying »
.  That took much longer than I thought it would.
 

Offline iMo

  • Super Contributor
  • ***
  • Posts: 5570
  • Country: va
Re: Voltage reference not working properly
« Reply #36 on: March 02, 2019, 07:37:56 pm »
You must limit the output current of the OP277 with a resistor as we told you (10k-100k for example). You should not wire an output of an opamp powered -V..+V with an input of an opamp powered 0..+V. Did you insert the resistor there? That would be an important point on my checklist.
« Last Edit: March 02, 2019, 07:40:55 pm by imo »
Readers discretion is advised..
 

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: Voltage reference not working properly
« Reply #37 on: March 02, 2019, 07:44:14 pm »
I have not done that, I'll do it now and see what happens.
 

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2850
  • Country: se
  • Country: Broken Britain
Re: Voltage reference not working properly
« Reply #38 on: March 02, 2019, 07:49:20 pm »
You must limit the output current of the OP277 with a resistor as we told you (10k-100k for example). You should not wire an output of an opamp powered -V..+V with an input of an opamp powered 0..+V.

I'd agree, but the -5V is now always good at -4.96V. Still a good idea to protect the next op amp's input from -0.3V or more with a series 1k-10k.
.  That took much longer than I thought it would.
 

Offline iMo

  • Super Contributor
  • ***
  • Posts: 5570
  • Country: va
Re: Voltage reference not working properly
« Reply #39 on: March 02, 2019, 07:59:39 pm »
Put there 100k for a test as it may show something. When all works fine you may start to lower the resistor when necessary.

Hint: It is not a good methodology to wire entire complex system together, power it on and watch various meters and displays  :bullshit:. Especially when you mess with a prototype.
Always wire it part by part, module by module while you have to always test the parts or modules individually. Take notes on voltages and currents.
Readers discretion is advised..
 
The following users thanked this post: FotatoPotato

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: Voltage reference not working properly
« Reply #40 on: March 02, 2019, 08:06:42 pm »
Ok, so I added a 100K and 10K on the output of the OPA277 and for both the LTC1046 outputs exactly -5.006v, so I guess that helped. The output of the 431 is 0.99v now....  :scared:
 

Offline iMo

  • Super Contributor
  • ***
  • Posts: 5570
  • Country: va
Re: Voltage reference not working properly
« Reply #41 on: March 02, 2019, 08:18:59 pm »
Have you wired the resistor exactly as I depicted in the schematics above?
Readers discretion is advised..
 

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: Voltage reference not working properly
« Reply #42 on: March 02, 2019, 08:21:16 pm »
Yes, from the output of the OPA to the input of the AD8630, Like this right?
 

Offline iMo

  • Super Contributor
  • ***
  • Posts: 5570
  • Country: va
Re: Voltage reference not working properly
« Reply #43 on: March 02, 2019, 08:24:24 pm »
ok, that is right.
Readers discretion is advised..
 

Offline iMo

  • Super Contributor
  • ***
  • Posts: 5570
  • Country: va
Re: Voltage reference not working properly
« Reply #44 on: March 02, 2019, 08:27:34 pm »
When you turn/set with the R29 - does the -2.5V voltage at 431 change "linearly" or "abruptly"?
« Last Edit: March 02, 2019, 08:33:55 pm by imo »
Readers discretion is advised..
 

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: Voltage reference not working properly
« Reply #45 on: March 02, 2019, 08:29:33 pm »
So when I disconnect the 431 from the -2.5v the 431 actually does output -2.5v and when I probe the -2.5v rail with the 431 disconnected there is 4.096v on that rail... And once I re-connect the 431 the -2.5v rail goes to 0.99v

It doesn't change rapidly, just linearly as I change the pot.
 

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2850
  • Country: se
  • Country: Broken Britain
Re: Voltage reference not working properly
« Reply #46 on: March 02, 2019, 08:31:57 pm »
"The output of the 431 is 0.99v now"

LOL.
There should be only 3 things connected to the -2.5V line, the LM341, the 1.3k supply resistor, and one end of VR29 10k. So, either VR29's wiper is wrong, or there's something else connected to the -2.5V.
.  That took much longer than I thought it would.
 

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: Voltage reference not working properly
« Reply #47 on: March 02, 2019, 08:38:40 pm »
Well, it must be R29, cause that's the only thing connected to -2.5v

I'll replace it and see what happens
 

Offline iMo

  • Super Contributor
  • ***
  • Posts: 5570
  • Country: va
Re: Voltage reference not working properly
« Reply #48 on: March 02, 2019, 08:40:51 pm »
There is something wrongly connected which is able to pull -2.5V up to +0.99V, something like 800ohm - 1kohm from +5V.

EDIT: +4.6mA pulls it up to 0.99V
« Last Edit: March 02, 2019, 08:58:51 pm by imo »
Readers discretion is advised..
 

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: Voltage reference not working properly
« Reply #49 on: March 02, 2019, 08:52:59 pm »
So I just replaced the old pot with a brand new one and now the voltage on the -2.5v rail is 0.77v, so something still isn't right. The resistance between pin 1 and 3 on R29 is 10.3k and here is a screenshot of the -2.5v trace on the board. This is the only spot where -2.5v is present. If you want the Altium files I'd be happy to send them over

I have a wire soldered to the center pin of U4 and that goes to a breadboard with the LM431 and R1
« Last Edit: March 02, 2019, 09:02:58 pm by FotatoPotato »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf