Author Topic: Ovenized MAX6226 reference  (Read 1311 times)

0 Members and 1 Guest are viewing this topic.

Offline golf32Topic starter

  • Regular Contributor
  • *
  • Posts: 54
  • Country: us
Ovenized MAX6226 reference
« on: October 22, 2021, 03:39:54 am »
Disclaimer: I don't know what I'm doing and I'm doing this for fun. Please keep in mind this is my first metrology project.

I've been working towards building an adjustable DC reference (my goal is "6.5 digit performance": lowest range 1-0v in 1uV resolution) and am in need of a voltage reference board for that purpose. My first thoughts are either LTZ1000, LM399, or MAX6226. LTZ1000 is off the table because as of yet I have not contracted volt-nutieutus. I rejected the LM399 because of the high noise compared to the MAX6226. The issue with the MAX6226 is the high temperature coefficient. I thought I'd rectify this by ovenizing it myself. What could go wrong? Side note, thank god for free samples. They make projects with fun expensive chips much more affordable.



Here's the schematic I came up with, pretty simple. Note that the heater setpoint is derived from the reference voltage with a trimmer. This is bad for two reasons: If the reference had a positive temperature coefficient (it doesn't luckily) a thermal runaway could occur. Second, the trimmer is poor in terms of stability.

I thermally bonded the reference, heater, and temperature sensor together by stacking them on top of eachother. This was very hard, and in the future I will not do this unless it works very well.



I 3D printed a cover to thermally insulate the reference and to squish the reference sandwich together. A piece of foam is inserted into the cover to aid in squishing.  No thermal compound was applied because I didn't have any on hand.



With lots of wrangling involving total re-assembly I got it to work. (edit: note that the flux was cleaned off later)

I built a long term test setup to monitor the voltage. Thanks to my university's lab manager for allowing me to steal this multimeter and corner of the lab for this.



I have a raspberry Pi logging voltage data coming from the multimeter and temperature data coming from the oven and ambient temperatures. All of the data is uploaded to the cloud so I can monitor it remotely. I plan to leave it going for at least 1000 hours to stabilize.

This was a great learning experience in soldering, experiment design, and datalogging. Any suggestions in how to improve my absolutely terrible test setup would be great.
« Last Edit: October 22, 2021, 07:07:24 am by golf32 »
 

Offline bob91343

  • Super Contributor
  • ***
  • Posts: 2675
  • Country: us
Re: Ovenized MAX6226 reference
« Reply #1 on: October 22, 2021, 05:33:47 am »
How do you get free samples?  I have found them impossible to get, probably because I don't have an engineering job.
 

Offline golf32Topic starter

  • Regular Contributor
  • *
  • Posts: 54
  • Country: us
Re: Ovenized MAX6226 reference
« Reply #2 on: October 22, 2021, 06:18:06 am »
They technically weren't free. They just cost college tuition...
 

Online Andreas

  • Super Contributor
  • ***
  • Posts: 3222
  • Country: de
Re: Ovenized MAX6226 reference
« Reply #3 on: October 22, 2021, 06:49:41 am »
Hello,

just some thougths:

Noise is usually not a large concern in 6.5 digit DMM applications (which integrate away the noise partially).
It may be different if your reference influences a RF oscillator and phase noise is a concern.

On the pictures the board is not cleaned after soldering. This is a no-no.
(after cleaning: do not touch with the fingers).

I am missing some kind of I-controller on the heater.
You have a relative large amplification on the heater error amplifier.
I fear this might give oscillations on the heater output.
Especially if there is some time delay from heater to temperature sensor.

The poor heater transistor may get up to 0.5W power dissipation with a 75 Ohms heater resistor.
This is too much for a SOT-23 package.

For precision trimming you usually only influence a small range by the trimmer.
Typically you use 2 fixed resistors as voltage divider for coarse voltage and influence the divided voltage by a 3rd resistor from the wiper output of the trimmer.

Layout of ground connections is critical. (I see no measures in schematics to avoid this).
I fear that the heater current creates several 100 uV voltage difference on output voltage (due to common usage of ground wires) depending on environment temperature (different heater current).

with best regards

Andreas

 

Offline golf32Topic starter

  • Regular Contributor
  • *
  • Posts: 54
  • Country: us
Re: Ovenized MAX6226 reference
« Reply #4 on: October 22, 2021, 07:06:50 am »
I had to do much troubleshooting after taking that photo, I cleaned the flux off afterwards.

Stability of the thermal control wasn't something I was thinking about. I initially had a lowpass filter in hopes that it would dampen some oscillations, but I removed that. Initial testing seems to indicate that there aren't any oscillations. One of my professors gave me a Jim Williams article about thermal control loops like this, and I'm looking at that for improvements. Will definitely take another look at that.

I may turn down the voltage on the heater resistor to reduce power dissipation. It seems to have no problems staying warm set at 12v.

That's a good point about the grounding. I should be able to cut the trace on the ground leg of the transistor and give it it's own path back to the power supply

Thanks for your suggestions.
 

Online magic

  • Super Contributor
  • ***
  • Posts: 6733
  • Country: pl
Re: Ovenized MAX6226 reference
« Reply #5 on: October 22, 2021, 07:16:45 am »
What could go wrong?
Since you ask ;)

The first problem I see is that you are regulating the temperature of the resistor, not the reference IC. Depending on how much heat flows from the resistor through the IC to the board, the temperature of the IC will vary.

Another potential issue is that if the IC package isn't perfectly thermally symmetric, different solder joints may be at different temperatures, and the temperature differences may vary with PCB temperature and heat flow, resulting in thermocouple errors.

I wonder if it wouldn't be better to solder the heater to the PCB bottom side centrally under the chip and definitely glue the temperature sensor directly to the chip. Assembly would be easier, too.

You could experiment with using the reference itself to sense its own temperature. One probably viable method is to measure its current consumption (it ought to increase proportionally with absolute temperature), another possibility is that perhaps some of the DNC pins output a voltage which varies with die temperature (but that would take testing and/or reverse engineering the IC to find out).

edit
Maybe a single DPAK/D2PAK transistor would suffice as a heater?

And another remark: if you ever tear down a TO92 package, you will find that the die is glued or soldered to a copper plate which is one piece with one of the leads. Obviously the lead is long and narrow and the plastic case is thin and wide, but copper is a better thermal conductor too. Therefore you are measuring some sort of weighted average of resistor temperature and PCB temperature and the resistor likely dominates, but dunno what the ratios are and it's not clear if PCB influence is completely negligible.
« Last Edit: October 22, 2021, 05:08:37 pm by magic »
 

Offline golf32Topic starter

  • Regular Contributor
  • *
  • Posts: 54
  • Country: us
Re: Ovenized MAX6226 reference
« Reply #6 on: October 22, 2021, 05:38:46 pm »
I actually left a resistor footprint on the bottom of the reference. My original plan was to put the resistor there, but I ended up making the sandwich because I wasn't sure about the thermal transfer properties of FR4. Maybe I should add another resistor on the bottom side for more even heating. I should also probably print a bottom cover for the board too.

Rev 2 (if there ever will be one) will definitely use a different temperature monitor. One with less thermal mass and better thermal resistance.

Thanks for the ideas.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf