Russian L3-3 tube tester

You can use the card for ECC82 provided the current range is set for 15mA on that tube.


Jacmusic sells these.


Judging from the curves, you can use the 250V anode specification, and about -9V grid for 15mA of anode current.


Tubes have guidelines, or you can make your own card with some free software, but risky in this kind of tester.


Cheers,
V4LVE
 
Great software--but think Ive found bug...

Its to do with G1 voltage. Card I make is for 7 pin 6AQ5, All apparently seems to work OK--Except when you come to set the neg G1 volts with internakl meter. That doesnt work.
I can set the correct G1 neg volts, -12.5, using digi meter to pins 1/7, but on pressing the 'measure' button to check with internal meter, the neg volts drops to -5v, internal meter not reading.
Ip seems correct and so does Transconductance readings--providing the correct neg 12.5 is set via digi-meter.

I have a 6P1P (6n1n) card and tube, this works out normally, so I think tester is OK

BTW,--Its a L1-3...
 
Thanks for getting back to me--However, I'm an idiot!

Human error stamps his heavy foot all over this, as I dont have working printer (expect new one on Wednesday) I copied the numbers down as I had for Octal KT88 and test EL84, but made an error on number 22 (1) I had written 23 (1) instead--so I set machine up wrong...

Correcting my mistake, It works beautifully!

Thanks again for this great software--I just need to work out how to install it so it has a shortcut now....
(Each time I close program--it vanishes from PC, each time I need it--I reinstall....)
 
That's strange, though .. 22/I and 23/I are just taps for the heating and should have nothing to do with Ug1?
It should only make you unable to adjust the heater voltage correctly.

Or do you mean something else?

Also, be VERY careful with this method in general, since it's REALLY easy to plug shorts which cause the magic smoke to escape.

I'm still working on it btw., but I coded myself into a corner and have to rework it deeply. Since I also work on software at work, I have to admit that my motivation is somewhat impaired at the moment.
 
Yes--I noticed that too after I wrote the reply. Must be another error I made somewhere...

Something I do when setting the pins is to keep 'mains voltage' test pressed--this tells me if something wrong--deviation on meter from set 120, indicating more current draw. Its worked for me....

Ive been trying to generate a card for 12AX7, but it seems to object to the cathode connections to pins, Ive tried the various Noval sockets, but only the A5 comes up all greens.
Odd, as the 6H1P is very similar only heaters different, and I have a card that does both sections--you just change over 3 pins for the sections........

Another thing--My L1-3 seems to have voltmeter and micro-ammeter built in using the main meter, with external jack sockets and a switch between V/uA The range is on the 'Insulation' rotary switch, going from 1.5v to 500v in steps.

Ive no idea how to access that function though....
(sorry about cruddy pic, this pic is from seller and before I cleaned it.)
L1-3.jpg
 
Last edited:
What you have there is likely an "MILU-1".

Found this reference right now, but they don't know much about the specialities yet: https://www.radiomuseum.org/r/frunze_milu_1il_1.html

I, unfortunately, don't know anything about it, either. But there is one plug, the 28/II which is officially not used, but on the L1-3 and L3-3 I encountered it DOES connect for socket wiring purposes (Line F to G1, IIRC). Maybe they used it differently on the MILU-1.

It being near to the current range setting may imply that it was once used to connect the meter to this special ranges and they later repurposed it.

So use my software with care and check this out maybe.
 
Absolutely--With care! Its a nice machine, It had a few o/c resistors, etc but Ive sorted most if not all its horrors. MILU-1, OK--interesting. have to see what we can find out...

Ive got in PDF format a set of original cards, Ive printed out a few relevant ones, it works with those fine.

Not sure where I'm going wrong--but cant print cards from your software...
--print button presses, but nothing happens--Gotta be me doing summit silly....
 
Cerker,

Tried a few times to print out a test-card, but nothing happens--The print button presses on the page, but nothing happens at all. Re downloaded and installed program, still no-go. Its the card I tested earlier by copying down numbers, I know it works, but cant print it out.

Tried getting a ticket on sourceforge--but comes back 'Forbidden' dont have permissions, when I go to create one,-- even though logged in. In both cases--What am I doing wrong?
 
Right,,,,

Getting somewhere--Sorta!....

Version 191229b doesnt work at all. Something missing from software.
Version 191228b appears to work--But wont print out.
Version 191227b seems to function And does appear to print--but is older software, Will need to confirm print size etc tomorrow.

None seem to have what can be called an installer. It must be opened by accessing the 'exe' file from the list of software components in the file named L3-3Konverter-snapshot. and is lost on closing. Re-opening must be done by accessing the 'exe'

A shame--As its otherwise a great product

Still cant buy a 'ticket'..........
 
Today I found issue during measurement 5c4s recitifier. After aprox 10pcs measured smoke appear.... I think that my L3-3 is dead. But later on, everything works fine. But, if you look closely to schematic (and into device), load resistor for 5c4s is 5k2, which is combined with resistor chain R20-26. If you have cathode current 125mA (which is needed), on R26 is 200V, so 25W power dissipation. And factory assembled is 10W version :) After 2-3 minutes temp on it is around 250C. So not recommend to measure 5c4s and similar recitifiers too long.
 
At all:
Sorry .. I really don't have neither the time nor the "drive" to work at the software at the moment.
The problems are most likely from my change to a newer Qt version with MinGW compiler instead of VS2017 which I didn't get to work anymore.

Other projects to do and to be honest .. lots of programming at work, so currently I can't stand doing it at all at home. It was a little project I didn't expect to take off like that at and I had a lot more of spare time when I first started it.

Maybe I will resume it in a few months (or years) when things get slower again. If not, the source code is available, feel free to fork.
 
Not sure if you've managed to repair your L3-3 in the mean time or not.
My L3-3 has issues with the uAmeter / isolation tester and as it turns out there is an issue with the right hand "parameter" switch or to be more specific there is an issue with the switch's wiring.

It appears that the isolation used on the wiring has deteriorated over time and there are severe leakage currents. I've checked this by disconnecting the wiper of R123 and disconnecting the two wires going to the uAmeter. Then I've connected one lead of a multimeter to one of the wires and another one to chassis ground. It is possible to get varying readings between 2k-10Mohm just by poking around the wiring of the right hand switch. This is particularly annoying as there are high voltages lines going to this switch as well and shorting the anode supply or internal 250V supply to ground will result in costly damage.

Needless to say, that I'm very much not looking forward to rewiring the switch as there is no space to maneuver with the soldering iron and many heavy gauge connections going in all directions.

I'll try to source a spare switch somewhere, copy the wiring and then install the whole assembly.
Hi Alex, did you manage to repair uAmeter? I have exactly the same problem with my L3-3. I can't calibrate the uAmeter so that when setting the "measurement" switch it shows zero, and when setting "calibration" it shows the point "120". Only I have no problem with the wires. I checked the wires exactly as you suggested and nothing. This is not the reason why the meter cannot be calibrated. I even changed to another uAmeter with a similar internal resistance. Unfortunately, the problem is still the same ... Do you have any new ideas on what could be the reason for the inability to calibrate? Have you solved this problem yourself?
 
I worked many years with L3-3, I still have a working one but I use it only to test rectifier tubes, it uses real conditions for rectifier test - better test than simulated low DC test. For other tubes I built a Roetest, compared many features and faster test.
Some time ago, before Roetest, I bought 2 testers L3-3 "as is", lottery but the price was acceptable, checking them one looks perfect like new but missing main transformer and wire wound resistor attached on transformer, the second one looks used, some oxide inside, at first look I noticed that the ceramic switch need replacing. First idea was to move the transformer from the second one to the good looking one but there are a lot of wires to identify, all of them white, not so easy. Having the Roetest I gave up repairing/restoring them thinking to sell parts from them. If someone interested PM me. Also I have few versions of schematics, documentation, cards found online in time and 1 or 2 original user manuals that I can scan. And 2 complete set of tubes that I think I won't ever use.
In time I used L3-3 testing thousands of tubes and I repaired few so I have some experience.
 
Yes--I noticed that too after I wrote the reply. Must be another error I made somewhere...

Something I do when setting the pins is to keep 'mains voltage' test pressed--this tells me if something wrong--deviation on meter from set 120, indicating more current draw. Its worked for me....

Ive been trying to generate a card for 12AX7, but it seems to object to the cathode connections to pins, Ive tried the various Noval sockets, but only the A5 comes up all greens.
Odd, as the 6H1P is very similar only heaters different, and I have a card that does both sections--you just change over 3 pins for the sections........

Another thing--My L1-3 seems to have voltmeter and micro-ammeter built in using the main meter, with external jack sockets and a switch between V/uA The range is on the 'Insulation' rotary switch, going from 1.5v to 500v in steps.

Ive no idea how to access that function though....
(sorry about cruddy pic, this pic is from seller and before I cleaned it.)
View attachment 1029882
Left down corner are 2 connections B + and -, I think these and the switch 1.5V to 300V can be used as a voltmeter.
 
What you have there is likely an "MILU-1".

Found this reference right now, but they don't know much about the specialities yet: https://www.radiomuseum.org/r/frunze_milu_1il_1.html

I, unfortunately, don't know anything about it, either. But there is one plug, the 28/II which is officially not used, but on the L1-3 and L3-3 I encountered it DOES connect for socket wiring purposes (Line F to G1, IIRC). Maybe they used it differently on the MILU-1.

It being near to the current range setting may imply that it was once used to connect the meter to this special ranges and they later repurposed it.

So use my software with care and check this out maybe.
I think from MILU-1 to L3-3 was a continuous production, of course with some modification, improvements. MILU-1 from 1961 till 1970, then L1-3 then L3-3.
http://www.rw6ase.narod.ru/index1/pribor/izmiritel/milu1.html