Diode ID: Simpson 2701 DMM

May 30, 2012 15 Replies

I'm working on an early digital multimeter (nixie tubes, from the late



60's). It's a Simpson 2701 which was actually made by a German company:

formatting link



It has been having trouble with unstable and wildly inaccurate readings in all ranges. I found an open diode on the amplifier board, and I'm trying to figure out if it's anything special. It is labeled "NT 9757 N".



formatting link



There is one other identical diode on the board that measures like a normal silicon diode, but it looks like it could be something special.



For now, I've replaced it with an ordinary high speed silicon diode. Things are working better, but it still has accuracy problems on some ranges, and AC volts/amps aren't working at all.



If anyone is curious, I have what Simpson calls a service manual, but it's not very detailed:



formatting link


are any of the semicondutors marked with logo or name of any sort?

That looks almost big enough to be a stacked selenium pellet diode, possibly about the right era too.

Andy Cuffe schrieb:

This is obviously a diode made by Nortron, AFAIK a German Manufacturer of semiconductor devices.

Unfortunately, I can't find it in my 1969/70 catalog. The "nearest" device as to the type number is NT 9779, a diode with logarithmic reverse characteristic.

BTW, one of these meters belongs to my collection ...

HTH

Reinhard

Thanks for the information. Knowing the manufacturer may help.

The good one has a forward voltage drop of about .65v, and reads open in the reverse direction, but I haven't put much voltage across it since I don't want to damage the only good one.

I don't suppose you have a full schematic, or any other information on it?

I already have this, but it's not much more detailed than the Simpson service manual:

formatting link

Are there any the same type in places you could substitute with a common silicon type without fear the different characteristics might be a problem - such as relay back emf catch diode?

With a very cursory glance at the schematic, I noticed a transistor with a vbe reverse clamp diode - if that diode just happend to be the one you want, you could probably slide in a Shottky-barrier type and pinch that one for the replacement you need.

Andy Cuffe schrieb: [...]

You're welcome.

You're absolutely right! In fact all my information I have got via those both links you communicated here ...

Good night (it's almost midnight)

Reinhard

Good idea, but there are only two of these diodes in it. I might try swapping it into the other location to see if the symptoms change.

One problem is that they are in the middle of three boards sandwiched together. It's a 10+ minute job to get to them, and they are completely inaccessible while it's working. I might have to try adding some extension wires.

Since you already know the bad diode is open, what is the voltage across the two leads? AC or DC? You know the voltage drop for the other good diode indicates it is not germanium and probably silicon. So substitute and old silicon diode with a voltage rating greater than what you saw across the leads. You are not going to kill the device with another diode and you will then know if the diode is something special, or a common silicon device.

Paul

I already did just that, and things improved, but were still not right. The diodes are located on the middle board of a 3 board module, so access is difficult. Tonight, I disassembled the module again and soldered some wires to the one working original diode, and my substitute so I could take some reading with the module assembled.

I discovered that both diodes are reverse biased. The good one reads

17.8v, and the bad one that was replaced with a regular diode was reading 22v, so clearly it's some kind of zener. I jumped the replacement diode with an 18v zener and things are looking much better. Without any calibration things all ranges are reading very close, and the AC volts range is working. I think it's safe to say that I have found the problem.

I would like to thank everyone who responded.

Is that nested among the nixie boards? - a zener could easily be 180V in that location!

In which case you may get random symbols lighting out of turn but not innacurate readings as such, its also more common for zeners to fail S/C.

n

h a

or

Andy Why not take the good one out and connect it up to a variable dc power supply through say a 1K current limiting resistor. Then you can increase the supply slowly while monitoring the voltage across the diode and see where it clamps. I've cataloged all my odd ball diodes I've had lying around here over the years that way. Lenny

Andy Why not take the good one out and connect it up to a variable dc power supply through say a 1K current limiting resistor. Then you can increase the supply slowly while monitoring the voltage across the diode and see where it clamps. I've cataloged all my odd ball diodes I've had lying around here over the years that way. Lenny vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv

I have a home brew breakdown (onset of leakage) tester that wouldn't pass any safety inspection - but its served me well.

Its entirely made from parts salvaged from monitor mains in/SMPSU sections.

For 240VRMS use; it has a pair of 68k 2W resistors in series (SMPSU chip startup resistors) on the AC side of a moulded assy bridge rectifier block, on the DC side is a pair of 180uF 380V caps in series with the junction between the two connected to one AC arm of the bridge to make it a doubling rectifier.

An important addition is a shorting switch to discharge the caps between test - in series with a NTC inrush suppression thermistor to stop it blowing the switch contacts away first time you use it.

Doubling the voltage gives about 600V for testing diode reverse leakage - it also reduces the available current; I've tested the PIV of 1N4148 diodes without damaging them.

"Ian Field" wrote in news:luuyr.1266346$ snipped-for-privacy@fx04.am:

you'd think some enterprising soul would come out with a budget curve tracer to work with an o'scope. It wouldn't have to have a very high V range. IIRC,Heathkit used to have one. I also recall a circuit for a simple XSTR curve tracer in an engineering magazine's annual circuit collection issue.

Damn,I wish I had kept those books...along with everything else I had ratholed while at TEK.... I coulda been somebody! ;-)

For that matter,I wonder if there's a hobbyist market for other budget instruments like a scope-cal board(XTAL-controlled,divided Sq.Wave),or a DMM cal-check reference board,stuff like that?

Jim Yanik jyanik at localnet dot com

Lenny,

I have already figured it out (see my reply on May 30).

Initially I was worried about destroying the only good once since I had no idea what it was. For all I knew it could have been some exotic diode that would be destroyed by much reverse voltage. Since the diodes are completely inaccessible, I soldered wires to both of them, and reassembled it. With the unit powered up, it only took one measurement to find out that it's an 18v zener. I replaced the bad one with an 18v zener, and everything appears to be working properly at last.

Glad that you got it fixed. I recently found an intermittent cap in my

1970 Grundig portable radio. It's nice to see an old dog bark again. Lenny

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required