Author Topic: 2MHz True RMS Millivoltmeter Circuit – Equivalents to Russian Devices shown?  (Read 1797 times)

0 Members and 1 Guest are viewing this topic.

Offline Chris56000Topic starter

  • Frequent Contributor
  • **
  • Posts: 896
  • Country: gb
Hi!

Can anyone advise me on suitable equivalents for the Russian transistors, operational amplifiers and transistor arrays in the schematic for this Wide Band True RMS Millivoltmeter please?

http://irls.narod.ru/izm/volt/skv.htm

For example, many "equivalents" commonly quoted on websites are nonsense without a redesign, the K140UD16 has an equivalent quoted as "LM741CH", but the K140UD16 has it's Null Balance pins connected to the positive supply, indicating the pins are connected to the collectors of the input stage transistors, and some other opamps shown have frequency compensation capacitors, again, the 741 quoted as "equivalent" is nonsense, because the author points out in the circuit notes the opamps with a wide gain–bandwidth product are needed!

I'm aware that a lot of these Russian devices are "flat–pack" or metal can type and I'm happy with modifying the chap's suggested PCB layouts to suit Western equivalent devices!

Can anyone point me in the direction of a Russian Op amp databook?

Chris Williams
« Last Edit: June 04, 2023, 02:06:34 pm by Chris56000 »
It's an enigma that's what it is!! This thing's not fixed because it doesn't want to be fixed!!
 

Offline ch_scr

  • Frequent Contributor
  • **
  • Posts: 807
  • Country: de
I've been looking at / for russian datasheets as well, what I've learned so far is:
- use a russian search engine (e.g. yandex.com)
- use a browser that can translate whole websites "on the fly" (like Chrome)
- search the name with russian letters "к140уд16" instead of western "K140UD16"
K140ud16 seems to be an uncommon one, I've attached the datasheet that has turned up with the method described above.

Interesting note: In the circuit diagram above, it says "A6-A8, A9 K140UD11" but a bit below it says "A9 K140UD16".
I've also seen example circuits in datasheets where e.g. power or ground connections were missing.
When I told an old TV repair hand (from east germany), he replied along the lines of "Yeah the circuit diagrams were known to be slightly off".
Good luck!

Edit: Here is the initial text from the datasheet, raw google translate, seems the "Analogy list" is not too far off after all?
Quote
GENERAL DESCRIPTION
Integrated circuits 140UD7, 140UD16, 140UD20, KR1005UD1
are general purpose operational amplifiers with internal
frequency correction. The amplifiers are made according to a
scheme close to џА741.
The circuit diagram of the op-amp 140UD7 differs from
the џA741 circuit mainly in the device of the output stage.
In addition, there is an additional output from the
input of the amplifying stage for connecting a correction
capacitor. Connecting an external capacitor between
this pin and the inverting input increases the output
voltage slew rate to 10 V/µs when inverting.

140UD16 is an export version of 140UD7

KR1005UD1 - two op-amps 140UD7 (without correction output),
located in one housing with a single-row arrangement of pins.

140UD20, two op-amps 140UD7 (without correction output),
located in one housing with a two-row arrangement of pins.

KR140UD20AM - two op amps located on one chip.
Their circuits differ from the 140UD7 circuit in the output stage and
the presence of input stage base current compensation nodes,
ue to which the input current is an order of magnitude
less than that of the 140UD7.
« Last Edit: June 04, 2023, 03:24:59 pm by ch_scr »
 

Online dobsonr741

  • Frequent Contributor
  • **
  • Posts: 643
  • Country: us
You can be dead in the water without 19VNT1B matched transistor array. Even more, we don’t know what the Russian yield of this design was. They did electronics “at all costs”.

The claimed 2MHz 1% accuracy for sine wave is difficult to believe, for both the full wave rectifier and the log-antilog converter will have less than 10KHz -3dB bandwidth, just by the look of it. Compare this schematics with the best in class, still-in-production AD637: claims only 200KHz bandwidth for 1% error. And that is for a monolithic, laser trimmed rms converter.

If you aim to recreate an analog true rms meter, the best bet is AD8361, with a VGA servo operating it in a fixed output voltage setting. It can operate down to 20Hz, surprising from a 2.5GHz detector.
« Last Edit: June 04, 2023, 07:54:24 pm by dobsonr741 »
 

Offline Chris56000Topic starter

  • Frequent Contributor
  • **
  • Posts: 896
  • Country: gb
Member Ch–Scr, thank you for your time and tips sir!


I do use Chrome that translates "on the fly", and from what I can make out from the datasheet, which is an image and not OCR'd, the "к140уд16" has the Null Balance pins connected to the diff. stage transistor emitters, so the circuit published isn't correct, unless the original author of the project found it worked by returning the null preset slider to the positive line with higher resistor values!

Chris Williams
It's an enigma that's what it is!! This thing's not fixed because it doesn't want to be fixed!!
 

Offline ch_scr

  • Frequent Contributor
  • **
  • Posts: 807
  • Country: de
While translate.google.com doesn't like the pdf, one can make a screenshot and feed the picture into it. There even is a button to get "just the text" back.
Please report back when you find a better way  |O

The offset compensation shown in "к140уд16" datasheet is also connected to negative voltage.
Would the circuit in the original post work for the "к140уд11"?
One has to consider all this was drawn by hand, and at a time when photocopiers were but an engineers dream.
So one might speculate, that it was found the circuit worked better with A9 replaced by "к140уд16" and an additional note was made -
but a rework of the whole schematic was deemed too laborious.
Also just because the offset compensation was drawn and prepared for in the layout doesn't mean it would have to be populated -
maybe it was found to be unnecessary and the parts were just left out anyway?
 

Offline terminus

  • Contributor
  • Posts: 13
  • Country: nl
It is not  K140UD16. It is К140УД1Б (cyrillic) - K140UD1B

Here should be enough info about it. https://rudatasheet.ru/microchips/k140ud1_kr140ud1/
It suggests μА702С.
 
The following users thanked this post: ch_scr

Offline Chris56000Topic starter

  • Frequent Contributor
  • **
  • Posts: 896
  • Country: gb
Looking at the circuit diagram in  the article Part 1 revealed that A6, A8 and A9 match up with the LM318H which fulfils the author's notes that say they have to be high speed/wide bandwidth devices, and the LM318 has positive null balance pins connected to the collectors of the I/P stage, so that's three solved!

A2 and A3 appear to match up to ordinary '741s, the transistor array can probably be replaced with a CA3086 with a suitable PCB layout to connect pins 1, 2, 3, 4 and 5 as the differential pair instead of 10, 11, 12, 13 and 14, and the diagram incorrectly marks A10 (К140УД1Б) as "A9", so we just need to find a match for A10 if there is one, but a μA702 does NOT match!

Chris Williams
It's an enigma that's what it is!! This thing's not fixed because it doesn't want to be fixed!!
 

Online dobsonr741

  • Frequent Contributor
  • **
  • Posts: 643
  • Country: us
A10 can be anything, just a comparator for the overflow indicator.

Very curious of your measured bandwidth…. What do you expect?
 

Offline Vovk_Z

  • Super Contributor
  • ***
  • Posts: 1385
  • Country: ua
A9 is definitely K140UD1B - that's a slow jelly bean opamp. I'm not sure nowadays there is a sense to bring that circuit to life, from ancient times.
As it was said, now we have AD637.
But if TS wants to build it, just use almost any jelly-bean opamps. It could be easier to use precision types, like OP07, with nulling circuits.
(But I'm almost sure with precision opamps we won't need nulling circuits at all).
 

Offline Chris56000Topic starter

  • Frequent Contributor
  • **
  • Posts: 896
  • Country: gb
There's one very good reason for resurrecting old circuitry "back from the dead" with the original components, where available – they're likely to be much easier to get than many of the fancy surface mount things the semiconductor industry is trying to peddle us with on every electronics web page – they don't disappear like the "Wili–O–The–Wisp" before the ink on the datasheets has dried – are you listening Maxim?!

Also I would very much prefer things I where I can actually SEE what I'm trying to solder!

(I can see enough to handle SOIC and SOT23 but much smaller than this and 0603 for passives and I start having real trouble – I'm now 62, by the way!)

Chris Williams
« Last Edit: June 05, 2023, 11:54:39 am by Chris56000 »
It's an enigma that's what it is!! This thing's not fixed because it doesn't want to be fixed!!
 

Offline bobAk

  • Contributor
  • Posts: 39
  • Country: fr
Hello.  The main problem will be with the transistor array.  This array is made in the form of a ceramic-metal body.  Consistency and thermal coupling at a  high level.  There was a version in a plastic case, very bad.  CA3086 will not work, it is worse in this case.  The problem with op-amps is that in the Soviet Union they liked to use uncompensated amplifiers.  Achieving the necessary optimal characteristics in the circuit or obtaining "non-standard" characteristics.  But yes, lm318h is a close analogue.  Pay attention to the text in the article "коррекция на опережение". The rms converter itself is a6, a7, a8, a9.  With the proper approach and tuning, it provides the declared characteristics.
 

Online dobsonr741

  • Frequent Contributor
  • **
  • Posts: 643
  • Country: us
V20/21 should be as fast as possible - like BAS69.
As far as the transistor array, the MAT14 is available on DigiKey for $13.
Or, perhaps if nonlinearity/temp dependence is tolerable, and MC3346/CA3086 would do from Jameco for $3


« Last Edit: June 06, 2023, 03:31:42 am by dobsonr741 »
 

Offline Chris56000Topic starter

  • Frequent Contributor
  • **
  • Posts: 896
  • Country: gb
I haven't found a source for the К198НТ1Б yet that I would like to trust ordering from, so will design two PCB layouts, one for the MAT14 and one for the original Russian transistor array, then if eBay slides a notification of one onto my 'fone screen I can get one!

I'm a bit puzzled as to why, among the many dozens of Soviet–era transistors and ICs available on eBay and elsewhere, that transistor array isn't among them!

If there's any Russian or other Member knows of a reputable supplier that will post this transistor array to the UK, could they send me a link please?

Chris Williams
It's an enigma that's what it is!! This thing's not fixed because it doesn't want to be fixed!!
 

Offline ZhuraYuk

  • Regular Contributor
  • *
  • Posts: 56
  • Country: ua
I haven't found a source for the К198НТ1Б yet that I would like to trust ordering from, so will design two PCB layouts, one for the MAT14 and one for the original Russian transistor array, then if eBay slides a notification of one onto my 'fone screen I can get one!

I'm a bit puzzled as to why, among the many dozens of Soviet–era transistors and ICs available on eBay and elsewhere, that transistor array isn't among them!

If there's any Russian or other Member knows of a reputable supplier that will post this transistor array to the UK, could they send me a link please?

Chris Williams

Do you trust this seller ? https://www.amazon.com/S-U-R-Tools-Microchip-K198NT1B-analoge/dp/B07C1JRLFC

If that one does not suits you I can send you from Ukraine, price would be 2-4 usd per piece.
« Last Edit: June 07, 2023, 03:19:39 pm by ZhuraYuk »
 

Offline MegaVolt

  • Frequent Contributor
  • **
  • Posts: 908
  • Country: by
Here are some guides.
Remove the doc extension and then compile the archive.
 
The following users thanked this post: ch_scr

Offline MegaVolt

  • Frequent Contributor
  • **
  • Posts: 908
  • Country: by
...
 

Offline MegaVolt

  • Frequent Contributor
  • **
  • Posts: 908
  • Country: by
3/3
 

Offline Chris56000Topic starter

  • Frequent Contributor
  • **
  • Posts: 896
  • Country: gb
Hi!

Quote from: ZhuraYuk
]

If that one does not suits you I can send you from Ukraine, price would be 2-4 usd per piece.


Perfect Sir!

I've sent you a PM – two would suit me nicely please Sir!

Chris Williams
« Last Edit: June 30, 2023, 11:06:07 pm by Chris56000 »
It's an enigma that's what it is!! This thing's not fixed because it doesn't want to be fixed!!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf