"Classic" version bare PCB" How to substitute LM334 for resistor and capacitor?

Hello, My name is Fábio and recently I bought a "classic" version bare PCB.
I live in Brazil and didn't find the LM334. I have found the LM234.
Via datasheet they are the same family, only changing the temperature range.
But I'm having problems. These components seem very fragile.

The LM234 or LM 334 are my great problem. Suddenly it stops working, the voltage on pin 1 and 6 (of ECC82) goes to 80V and the voltage in pin 9 goes to 1V.
I bought 10 from China via ALI EXPRESS.

https://pdf1.alldatasheet.com/datasheet-pdf/view/22753/STMICROELECTRONICS/LM334.html
https://www.ti.com/lit/ds/symlink/lm334.pdf?ts=1674803089952&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FLM334

Is it possible to replace the LM334 (in the outside of the tubes) using a traditional method, using a resistor and capacitor?

Another problem I got. The range of the bias is too short. I've getting 240mA with the  minimum bias -39V
Talking with Roy, he help me and I change the 27k resistor for 15k, so I can get -28V
My tubes are Russian Tung-Sol, EL34B.

Best Wishes
FAB

There are a lot of Chinese parts that are just garbage.

An engineer who once worked for me and is now a bigwig at Analogue Devices says that AD will no longer accept return parts for failure test unless the customer can prove provenance.
The copies are so good that until the plastic is removed and the die examined, they are impossible to tell apart from genuine parts.

Ditto on capacitors, resistors, transistors, switches, ICs, etc. etc.

Buy GENUINE parts from Digikey, Mouser, Anthem, etc.

I have hundreds of these in stock and of course they are also available from Mouser and DigiKey.
Most of the ICs sold from china are fakes and do not work correctly.

IMO, the LM334 is a poor choice for tube bias. Its current sweet spot is 1mA. @ 10mA the performance curve is near vertical. It suffers from self-heating and ambient temperature fluctuations. It can be much improved by temperature compensation as outlined in the datasheet.

Better performance may be obtained with a discrete circuit or a different part.

Seamus wrote:
IMO, the LM334 is a poor choice for tube bias. Its current sweet spot is 1mA. @ 10mA the performance curve is near vertical. It suffers from self-heating and ambient temperature fluctuations. It can be much improved by temperature compensation as outlined in the datasheet.
Better performance may be obtained with a discrete circuit or a different part.

Hello Seamus, thank you your answer. All of the chinese LM334 are very bad and do not work.
Due that chinese components, i) I think my quadricapacitor burned out; ii) my GZ34 started sparking, and iii) my choke filter may be is burned out too It (seems not).

Do you have a schematic wiring circuit to show me what i need to do. LM334 is not a good way. For me, it's the weak link of the chain.
https://servimg.com/view/20477940/2

tubes4hifi wrote:I have hundreds of these in stock and of course they are also available from Mouser and DigiKey.
Most of the ICs sold from china are fakes and do not work correctly.

Hello Roy (or Bob), I'm having a lot of problems because of this component.
i) I think my quadricapacitor burned out becaude them; ii) my GZ34 started sparking becaude them , and iii) my choke filter may be is burned out too It (seems not).

How I need to do to you send me two or four.
Best wishes

FAB

What amp are you using?

Do you have the schematics?

What diagnostic tools do you have?

Seamus wrote:

What amp are you using?

Do you have the schematics?

What diagnostic tools do you have?

Answers

AMP: ST70 Bob Latino PCB Classic Version (3 x 12Au7 working as driver and phase inverter)
Schematic: Bob Latino PCB "classic" version bare PCB"  http://www.tubes4hifi.com/ST70.htm
Diagnostic tools: Osciloscope, multimeter, Wave function generator HeatKit from 10 to 100KHz, 100W lamp in series with socket that can be bypassed and Wattimeter.

Are my answers  satisfatory?

My Idea is mod the circuit to a traditional Long tail Pair Phase Inverter (similar to the schematic of the picture), but using the 12AU7.
But I need the correct values of the resistors and capacitors for use the 12Au7 tubes. Do you have any Idea? Schematic?



Best wishes and thanks

Based on Guitar Phase inverter Amps , I've found these values for 12AU7


But, I don't know if it  is correct

Phase inverter calculator
https://www.ampbooks.com/mobile/amplifier-calculators/long-tailed-pair/calculator/

fabio_ab wrote:Based on Guitar Phase inverter Amps , I've found these values for 12AU7


But, I don't know if it  is correct

Phase inverter calculator
https://www.ampbooks.com/mobile/amplifier-calculators/long-tailed-pair/calculator/

My question is: Instead of LM334, can I built a Long Tail Pair Phase inverter using resistors as the schematic below?
Note: same gain in each triode, resitors are very easy to find. Very easy to implement.

I´m lost. What is the advantage in assume a LTPI with LM334 instead a traditional way?

Best wishes, but I bwolud like read someone convincent.

Not sure about that calculator as it leaves all the resistors the same for tubes of vastly different gains and Rp. That is generally not done.

The 'advantage' of a CCS is that with a well matched tube, there is no need have a balance resistor on the phase splitter. If the phase splitter tube has mismatched sections, which is more probable than not, perfect balance is not achieved.

OTOH, without the correct test equipment, the probability of adjusting an amp with a phase splitter balance trimmer for perfect balance is small.

With good LM334, the circuit should work well provided the rest of the amp is correctly wired.


There are literally THOUSANDS of designs and opinions.
Without experience, and knowing the parameters of the other parts of the amp, the likelihood of success is unknown.

Thanks Seamus.

I built the circuit I've posted above.
I measured 34V RMS on Pin 1 and the same valuue after the 0.22uF capacitor going to first EL34 (but without DC as must be) .
On Pin 8 I have the oscilation phase inverted, but I can't see the amplified oscilation on pin 6 and obvious none oscilation after 0.22uF.


I'm almost there! I think It's possible

I apreciate very much your help.
Best wishes.

.

[quote="fabio_ab"]

Seamus wrote:What amp are you using?

Do you have the schematics?

What diagnostic tools do you have?

Schematics is ST70 - VTA70 amplifier. I don't know if I can post the complete schematic here. I Only present my mods.
Diagnostic tools: you can see at the picture. Osciloscope, Heatkit wave generator. Two multimeters. Home made Wattimeter. Load Resistence  of 8.2Ohms
`


Take a look at the  resistors of 82k and 100k (not written over)  instead of 30k.

Output power at 600Hz.



The reap on the top the wave is because of the phanton resistence load.
18V RMS on a 8.2Ohms resistor. Almost 40W.
Tube current 40mA. Plate voltage 440V. 17.6W each tube.

Not bad without LM334

I think my aproach is working fine and the first time I saw a long tail phase inverter with ECC82 and well balanced.
Thanks for all

Congratulations!

4 resistors 1 capacitor, and you have a robust phase inverter "in phase with the time" of ST70.

IMO  : One more problem I found  about this board is that the bias circuit is wired on the PCB.
But, it's dificult to make a mod without extra holes in chassi

Best wishes Seamus