Replacing Zener with vreg?

[ogden]

Well I reckon the TL431B (0.5% version) sounds the way to go. 

I'll just have to figure out the resistor values now and see how it goes.

Any progress?

[TERRA Operative]

Soon, was waiting on the other parts to arrive and I've just laid out the PCB.
I'll just have to have a second look to tweak the PCB layout then I can see if it works..

I'll have to figure out how to determine the resistors needed, and dial them in in-circuit to nail the values.

[ArthurDent]

As imo and Andreas have said, that original zener was chosen for a very good reason. A couple of precision calibration D.C. supplies I have use a 1N829A or a 1N3499 as their reference and are still working within spec after about 40 years. To maintain the accuracy of the voltage of these temperature compensated diodes, the supply voltage to the zener is supplied by a constant current source so they were immune from power line variations. Dave has a video where he got one of these standards on eBay and checked a few voltages and found it was still within the original specs. Your circuit uses a resistor from a regulated 12VDC supply which pretty much is a current regulator although I believe your reference doesn't require the ppm accuracy of the voltage standards. The actual voltage of the reference zener isn’t that critical and is generally compensated for later in the supply circuitry. Below is a description of how one of my references works.

To maintain the original accuracy and stability of your meter I would not try to modify the circuit, especially where the original part is still available from some suppliers. I’d be careful of eBay sellers in China and try to make sure you get a genuine part.

[TERRA Operative]

Can you point me to any suppliers of the original part? (HP part number 1902-0692 which cross references to the JEDEC part number DT790107A)?

I'm having trouble finding it anywhere....

[TERRA Operative]

So, in my desperate digging for diodes, I'm starting to think imo was on the right track.

It seems that the part I am looking for was a 'HP selected part' and the voltage tolerances of 1N828/829 diodes cover 6.3v.
I'm thinking HP were characterising batches of diodes and were selecting those that conformed to the 6.3V 1% requirements.

Problem is, I'm not quite rich enough to go buy a batch of those diodes at $10-18 a pop just to find one that is the correct value and not use the rest....


I need to make a friend at Mouser/Digikey who can go test a few diodes and just send me the good ones. 

[ogden]

I need to make a friend at Mouser/Digikey who can go test a few diodes and just send me the good ones. 

You don't need to - because R7 trimpot is already your friend

[TERRA Operative]

Alrighty, a bit of an update on the schematic.

I updated it with a few corrections, which weren't completely clear from a cursory glance at the schematic. Most notably the + and - 6.3v rails from the zener at the top right corner and a few labels here and there.

Might help make things clearer maybe.



Also, re. the zener and R7, the different voltage (if it isn't 6.3V) isn't going to cause problems with the 'vref' input in U3 (AD7530LN)?

[ArthurDent]

The actual voltage of the reference zener isn’t that critical and is generally compensated for later in the supply circuitry.

As I said in my previous post, the actual voltage isn't that critical. The LM399 and LTZ1000 references HP has used in their most accurate and stable multimeters have a range of voltages and that isn't a concern because these are compensated for in the circuitry. As pointed out in your schematic, R7 is a gain adjustment to correct for the minor differences in reference voltages. You will gain nothing by trying to select this zener.

[TERRA Operative]

Cool. I'll get myself a 1N828/829 and stick it in.

Sorry if I sound stubborn or something, it's not that (well I am stubborn, but not right now. ), just that I am still learning this stuff so I'm still figuring out where I can and can't tweak numbers and values.

[ArthurDent]

Here is a little more info on the circuit you want to recreate as I understand it. First, there is an error on the first 001 option schematic you show. The 100K pot marked R1 should be marked R8.  Second, this 001 option only allows you to use an internal bias supply instead of using an external precision supply. This eliminates the need to use a separate supply to bias the DUT with up to +/-40VDC.  There are two ranges on the internal bias supply of 0 to +/-10VDC and 0 to +/-40 VDC selected by the U6.2/U6.3 switch and U6.4/U6.5 selects polarity. I also assume R1-1 through R1-6 is a low temperature coefficient precision resistance module. Here is the adjustment procedure that indicates that the actual Zener voltage isn’t too critical because there is an adjustment on the supply to get the exact voltage HP wanted.