Datron 1061 refurbish AC board

[xantia_99]

Hi,
i have refurbish my old Datron 1061, Analog board, Digital board, power supply. Together with resistance and current board it work excellent . Put in a MAT12 and AD711 after that it worked very good. Problem with sockets, drivers, optical ..
Now it works and very stable.

Today or shall i say the last year refurbish the AC-board. instability and not possible to make a zero calibration. To much noise
Lost Hfe in 2N3905, 2N214C and so on. Many measurement traditionally  with diod tester  ok , but no gain with curve tracer. Found 6 transistor without/low gain until now from input to TL7. The differential Q21 and Q22 did not match

Now my problem:
I miss information about  Teledyne double jfet SU2656M  to make it possible  to measure with a curve tracer to see if I have to exchange them
I am on the right way.  I have spend to may hours but I have retired so ..

I am a bit tired of all this but want to finish with AC-board at the same level as new or better.

Best Regards
S

[Chris56000]

Hi!

You can buy this device from Littlediode in Malden, Surrey, United Kingdom:-

https://www.littlediode.com/components/SU2656M.html

. . .their webpage I've linked here has some basic information on the characteristics of the device - Littlediode have an eBay Portal you may be able to go thro' to get one!

A whacking great thick book that covers the SU2656M can be downloaded for free here:-

http://bitsavers.trailing-edge.com/components/national/_dataBooks/1989_National_Discrete_Semiconductor_Products.pdf

The cross-reference tables suggest the SU2656M is similar to the J401-J404 series, etc., using FET Process 98, so please refer to pages 11-195, 11-196 and 11-197 for some basic info!

Let us know how you get on with your 1061 AC Board!

Chris Williams

[xantia_99]

Hi Chris5600
Thank you for the information, I guess I have looked at the wrong place, earlier version on many books 70th and manufacturers as I looked for early  jfet.
To make it easy for me I have put every transistors and jfet on "socket". I will see if I remove them in the end. However my old refurbished Tek 468 use it as many other Tek products.

My next question I guess none could answer.

I look for information about the RMS module 260026, made by Datron
Best regards
 
S

[Testtech]

The RMS chip was not actually made by Datron. It is a Silicon General SG3083J transistor array. It may be that Datron "selected" these for some characteristic, but who knows.

[dacman]

I realize that you've stated that the Analog and Power Supply PCB's have been refurbished, but C45 through C50 on the Analog PCB can cause noise in the ACV function.  Also, C6 in the Power Supply can act as a transmitter if it is in bad shape due to the high voltage across it.

Relays can also cause noise on ACV.  The Front/Rear relays can be tested from the front panel binding post to the plugs near the relays (after pulling the plugs), and also from the plugs to the rear panel.  ACV board (input) relays can be tested with normal pin leads from the wiring to the PCB.

You did not state which ACV option it has (I would assume option 10) but option 12 has choppers that can wear out and cause ACV noise.

[xantia_99]

Hi Testtech

I can now confirm that the Datron 1061 rms chip 290026 looks like it behave as it should do as a SG3083. I have have done some measurement  on the NPN transistor array. All of work, there are five independent transistors, two of them are matched, and all of them have a  collector connection to the substrate with a diod.
However I have not measure the Hfe gain as I have not tested it in my curve tracer.
I have now refurbish everything before the rms kit, except Mat01H and two SU2656M JFET. The problem/noise is transported on the virtuel earth everywhere , connected to DC earth with a resistor and capacitance.  To measure the current in these component with a scope is quite interesting.  The problem/noise  is decreasing with temperature.   I will solve it, it just takes time.  The PCB have some drawback in the design. So I guess I find the problem with a component with bad CMRR as it pickup the noise from the virtuel earth and amplify it.    I will solve it, it just takes time.
I wish I had experience in unlabeled Mat01AHZ op amp, but it may bring new problems.
Best Regards
S

[Gyro]

Pah! I always thought those RMS chips were Datron magical custom Silicon - especially the way they soldered the pins into the socket!

I'm sure the parts must have gone through the full Datron burn-in and selection process though.

[dacman]

You should be able to replace the SU2656M with a U404 (TO-71).  A company in California called Linear Systems makes a lot of older JFETs.  The J404 would be a DIP-8 cross, but LS doesn't list it.

[xantia_99]

Hi,I used the internal schematic of the chip and got values that is most likely to confirm it. However I can not be 100% sure. I spend time to understand the logical and analog function of this chip how it works in this circumstances, and I am not convinced all the way yet.

So until i know more I use it as hypothese until i have something better, but still it behave like the chip Silicon General SG3083J transistor array. I have no indication of fault in this function.

[dacman]

SG3083J is on (some of) them, along with the SG symbol, which are covered over by the Datron part number.  If it has an aluminum strip, then the Datron number is (usually) on the strip, and if the strip is removed or falls off, the SG3083J number is (usually) exposed.