General Resistance DAS56A

[r6502]

Hello all,

I also own one DAS 56A. Mine has now 2 problems:

• 1st the reference broke down, so I build a new one, based on LM399
• 2nd after I moved, the CSA was not working any more - it output is always on the positive rail (about 13,8V)

I checked the power supply - it is fine at +/- 15V DC, I checked the reference voltage - all fine at 10V (+/- some 10µV) I checked the input filter for the CSA, Caps are fine, isolation resistance is ok, and capacity is also fine as well as the 3 resistors.

I also removed the CSA from it's socket and placed it back - same result.

So my thought is, the the CSA has a defect. Does somebody have a schematic of this PCB - in the manual you find a black box for the CSA ...

I do not find a schematic for this, so I'm thinking to do some reverse engineering, if nobody has a schematic for this, but this will take some time. On this board there are 2 good old boys: LM308, sitting in the metal can  - I love it to work again ...

Kind regards

Guido

[chuckb]

To block the high frequency noise on interconnect lines I use the Fair-Rite 0475164281 clamp-on ferrite bead. It uses type 75 Material with peak impedance around 1 MHz (100 ohms). This model has a 0.25" ID. The 0475164181 Model has a 0.5" ID. For broad band performance use two different type of beads on the interconnect lines.

I use these on the power line and input cables of Nano-voltmeters to reduce EMI from a local 1 MHz broadcast station. I must have 30 of these around the lab for the sensitive instruments.

[r6502]

Hello All,
 I just figured out the layout of the CSA inside the DAS 56A.

red lines: tracks top layer
green lines: tracks bottom layer
yellow dots: pads

interesting thing: in the schematics of the DAS they do not show the GND pin for the CSA...

The schematics will follow ...

Guido

edit: updated graphic with white background, spelling

[r6502]

Hello All,

I just finished the schematic for the CSA inside the DAS devices.

This represents the schematic of my board. Any comments are welcome.

There are questions regarding the parts:

T6: This MOSFET is marked as "MFE300" on its housing, but I did not find any data on this MOSFET type.  It is also a N channel type MOSFET.

D6, D7: Marking on  is not readable.

Guido

edit: infoe on T6, D6 and D7

[kleiner Rainer]

Guido,

I checked my old ECA Data Books from the eighties, but there was a distinct "hole" between MFxxx and MFE3002, so no MFE300 to be found.

The Motorola Databook on bitsavers.org also only gave data for the MFE3002. Did you check the lettering with a magnifier? Sometimes the lettering rubbed off or was not fully printed to begin with.

Here the link to the Motorola data book:

http://bitsavers.org/components/motorola/_dataBooks/1974_Motorola_Discrete_Semiconductor_Series_A_Vol3_Non-Registered_Type_Numbers.pdf

Databook page 310, PDF page 327 has the datasheet for the MFE3002. Maybe the MFE300 is a selected version?

Greetings,

Rainer

[r6502]

Hello Rainer,

that's great - Thanks.

meanwhile I found another post, also on EEV blog, where the user wrote, that there are two MOSFETts of the type MFE3002 assembled:
https://www.eevblog.com/forum/repair/sub-for-mfe-3002-mfe3002/

Guido

[r6502]

Hello all:

Update schematic CSA:

I mixed up gate signals of T5 and T6. It is corrected in rev. 01 of the schematic.

Kind regards

Guido

[bastl_r]

Hello

I has a GR Voltage Standard too. Its Name ETS15.
There is also the same CSA build in. But the output voltage is maximum 1V.
My CSA is not stable. The voltage is drifting a litlle up and down.
Once i built another CSA with an TLC2652 but its not powerfull enough.
Now with the schematic i can look for the bug again. Thank you for sharing.

bastl_r

[r6502]

@bastl_r:

thanks for the reply.

Did you include a outplut transistor stage in your design, or did you just use the output of the TLC2652 ? The TLC2652 has only +/- 5V (abs. Max: +/-8V) supply, so, it is fine for a max output of 1V, but not for the units with 10V output.

I'm not getting the defect mosfet's for the chopper amplifier, and the russian replacememts did not work, so I think, that I will use a LT2057 followed by an output stage simmilat to the original design.

Guido

[bastl_r]

Hello

The ETS15 has only +/-6V on the CSA and max +/-100mV at the output.
But it ist not working perfect with the big cap on the output.
Now i will try to change the Transistorchopper with the TLC2652 or a new LT2057 and use the Transistorstage to power the output up.
There are only small differences of resistorvalues between my CSA and the CSA from your DAS.
I'm at work now and will post more details later.
regards

[bastl_r]

When the device is at rest it starts to drift away. Over 8 hours it is about 5 - 6 µV.
If the polarity selector is set from + to -, the continuously measured reference voltage before the KVD is reduced from 99.915mV to 99.897mV. This can be repeated at any time with different output voltages. I have no explanation for myself where the extra load on the reference voltage should come from.  I have attached a circuit diagram.It is on the last page of the attached pdf.

The CSA is exactly the same as in the DAS56 CSA with small value changes to resistors.
For R10 a bridge is inserted.
R24 and R25 have only 33Mohm instead of 100M.
R27 is a 9K1 instead of 3K9
T3 is a 2N4250

I will now try to place a LTC1052 on the CSA board and connect it as sketched in the schematic. Or is there something against it?

Regards bastl_r

[r6502]

HJello  bastl_r,

I'm planning something similar, but I'd like to try the LTC2057 for this. But did not try it yet ...

Guido

[r6502]

Hello All,

It took a while, but today I was able to update my DSA 56A and perform a first test.

In my unit, one of the reference diodes are defect on the original 5001-0048b reference board. I decided to install a new reference board, based on the LM399.  The new reference board fits in the same small box like the old one, and uses the same connections. To be able to use the original reverence supply I did adjust the output voltage to 18V, but this was not enough, as the old reference board was consuming less power then the new one especial at startup when the LM399 heater powers up. To be able, to get enough power at startup I modified the power regulator for the reference by decreasing the resistor responsible for current limitation. Original 16R was installed, now replaced by 3.3R.

On my Unit the CSA is also defect, and so I generated also a replacement for this board, based on the LTC2057 and the original output configuration with 2N2219 and 2N2905. Both schematic's are attached to this posing. Here I think is not mutch to say, have a look at the schematic - it is relay simple. as connection pins I used sub-d type male turned contacts, that have the same diameter than the contacts from the original CSA.

I did also an update on the power input port. I removed the direct connected line cord and made a cut-out for for an IEC input with integrated fuse holder. The internal line wiring, the transformer connection and the line switch have been insulated now. The line filter I left untouched for the moment.

First test after assembly looks promising. Next step will be a real performance check.

Please have also a view on the attached  PDF documentation and photos.

Guido

Edit: spelling