The ultimate DIY? A Stax SRM-T2!

[GeorgeP]

6 minutes ago, mypasswordis said:

Not sure what the Gold Aeros are, but regular 6DJ8s shouldn't be used with the T2. They have a max plate voltage of 130V and the top tube sees about 130V. 

Weren't they using both 6922 and 6dj8 at the same time?

[Whitigir]

I guess the EL35 Pentode was really intended for the T2 and not the Beam Tetrode.  There are 5 connections in the scheme ? While the Beam Tetrode has only 4 elements ? Speaking of which, better turn of T2 until KG confirm about this 

 

Edited November 1, 2017 by Whitigir

[mypasswordis]

Not sure of the "they" you are referring to but I am using 6922s, which are confirmed to work on Kerry's T2. Those have a max plate rating of 220V. Maybe there are more robust versions of 6DJ8 out there that can go above the normal max, but rather than do hours of research it was simpler for me to just use 6922s. 

[Whitigir]

6 minutes ago, mypasswordis said:

Not sure of the "they" you are referring to but I am using 6922s, which are confirmed to work on Kerry's T2. Those have a max plate rating of 220V. Maybe there are more robust versions of 6DJ8 out there that can go above the normal max, but rather than do hours of research it was simpler for me to just use 6922s. 

You did use 6922 in Kg T2 successfully ? Which one are you using.  Many tube retailers list 6922/6dj8/ECC88 in the same category 

Edited November 1, 2017 by Whitigir

[mypasswordis]

2 minutes ago, Whitigir said:

You did use 6922 in Kg T2 successfully ? Which one are you using

I just looked up the datasheet for JJ and it looks like they are calling the 6DJ8 the same as 6922 and the max voltage is 220V so you are fine. I am using JAN Philips 6922. 

[Laowei]

More tube weirdness. The JJ 6CA7 is a Beam Pentode, with the beam forming plates connected  to pin 1, similar electrically to the pentode EL34’s G3.

The other big bottle 6CA7s are Beam Tetrodes, with the beam forming plates internally wired to the cathode (pin 8). Pin 1 is NC in this case.

In the GG, G3 is tied to the cathode on the PCB. So no difference electrically between using any EL34 or 6CA7 in that position. In the T2 circuit, G3 is tied to the plate and draws current, so may have stressed the JJ 6CA7 beyond what it could handle. The internally connected EH 6CA7 doesn’t see that. 

Edited November 1, 2017 by Laowei

[Whitigir]

5 minutes ago, Laowei said:

More tube weirdness. The JJ 6CA7 is a Beam Pentode, with the beam forming plates connected  to pin 1, similar electrically to the pentode EL34’s G3.

The other big bottle 6CA7s are Beam Tetrodes, with the beam forming plates internally wired to the cathode (pin 8). Pin 1 is NC in this case.

In the GG, G3 is tied to the cathode on the PCB. So no difference electrically between using any EL34 or 6CA7 in that position. In the T2 circuit, G3 is tied to the plate and draws current, so may have stressed the JJ 6CA7 beyond what it could handle. The internally connected EH 6CA7 doesn’t see that. 

I think you just perfectly explained why my T2 cooked my JJ 6CA7, and not the EH 6CA7, while GG played real nice with JJ6CA7 and I loved it

it also seems the explanation, the Beam Tetrode has the second harmonic minimized.  Hence why I am hearing the EH6ca7 more balanced than any of the EL34 I threw on the T2

Edited November 1, 2017 by Whitigir

[GeorgeP]

16 minutes ago, mypasswordis said:

Not sure of the "they" you are referring to but I am using 6922s, which are confirmed to work on Kerry's T2. Those have a max plate rating of 220V. Maybe there are more robust versions of 6DJ8 out there that can go above the normal max, but rather than do hours of research it was simpler for me to just use 6922s. 

Fair enough. By "they", I meant stax, but just wasn't sure about the 6922. They certainly used 6dj8 tubes, as did a number of folks here if one were to search through this thread. Maybe the 6dj8s will expire sooner than 6922, but I have had no issues with NOS Matsushita 6dj8 and 7dj8 for at least 4 years (I also used 1980 JAN Philips 6922, but found the amp to not be as quiet as with some other tubes).

[Whitigir]

Well, it seems I better be safe than sorry.  Only go for true EL34 from now on.

[kevin gilmore]

so on the input tubes

the input tube has a plate voltage of roughly +75 volts and a cathode voltage of -3 volts

the cascade tube has a plate voltage that would be a max of 250v, but its a current source, and is typically +200V and a cathode voltage of +75 volts and the maximum voltage swing is about 15v

so both tubes well within specifications

closer look at the original pictures and the original tubes are el34G

they look absolutely identical to the Richardson tubes currently made in china which are labeled as 6ca7

Edited November 1, 2017 by kevin gilmore

[Whitigir]

In a sense, EL34G is the real EL34 and is Pentode, and not Beam Triode ? Ouch....I run that beam triode for a few hours 

[mypasswordis]

200-75=125. I don't know, it's just one of those things that wouldn't ever stop nagging in the back of my head, and 6922s cost the same so why make life more complicated than it already is imo. But I'm sure 6dj8s will last longer than people will take to get bored/start tube rolling. 

[kevin gilmore]

those JJ tubes have only 1 hole in the plate per side, the originals have 3 holes and a different configuration of the plate, more square than rectangular.

my T2 is still using the same tubes as when I demo'd it at CJ010.

I recently did a 100 to 120V conversion on a real T2 for someone on HF. It has the original tubes, 22 years old.

(pictures can be supplied if anyone has a real T2 and wants to convert voltages, better than $250 in shipping both ways)

The gold aero's are labeled 6dj8

 

Edited November 1, 2017 by kevin gilmore

[Whitigir]

Is Sylvania ECC88 NOS recommended ? I read Amperex was giving people problem and killing Transistors ?

[nopants]

if you’re talking about the jan 6922 with green print on the tube i’ve had good results with them

[Whitigir]

1 minute ago, nopants said:

if you’re talking about the jan 6922 with green print on the tube i’ve had good results with them

Okay, going to swap out these JJ and put them in together with Winged C el34, and no more tube rolling! Thank you so very much 

[Whitigir]

I tried the Green label JAN 6dj8, and eventhough it was Balanced from the store.  It was no good.  The balances went crazy from -2V to +3V in a blink that the DMM couldn’t even keep up.  The offset on Left channel was reading 25V and the one in the right channel was reading 177V....holy cow.  I turned off and took them out as soon as I could.....luckily....no smell...

Stuck back on the JJ goldpin ECC88 and the Balanced started right out below 500mV and keep going down steadily into +30 and -30mV on Left, and +150 to -150mV on right.  Balances was dropping steadily when started out 4.5V on Left, and 6.5V on right and drop quickly, probably within couple minutes and it dropped to 0.8V on Left and 1.5V on right, and keep dropping down.

I think these JJ goldpin are meant to be in there.  Better stick to what works as T2 is not something to be playing around with.  The next thing is probably a quad of Xf2 as everyone is saying they sound beautiful.  I have no choice but to get a Quad to get the best out of my T2 and while sticking to what work best.

Edited November 2, 2017 by Whitigir

[mypasswordis]

I doubt those are the same tubes that I got, which look like this: http://www.thetubestore.com/Tubes/6922-E88CC-Tube-Types/JAN-Philips-6922-Low-Noise

Stick with what works for you, though.

[Whitigir]

1 minute ago, mypasswordis said:

I doubt those are the same tubes that I got, which look like this: http://www.thetubestore.com/Tubes/6922-E88CC-Tube-Types/JAN-Philips-6922-Low-Noise

Stick with what works for you, though.

Ah, these are Philips  gawd, a lot of confusion.  I guess, the T2 really plays nicer with 6922 rather than 6dj8 ? 

[mypasswordis]

The JJ gold pin are 6922 so you are totally fine with regards to voltages, so just stick with what works. 

[GeorgeP]

2 hours ago, Whitigir said:

I tried the Green label JAN 6dj8, and eventhough it was Balanced from the store.  It was no good.  The balances went crazy from -2V to +3V in a blink that the DMM couldn’t even keep up.  The offset on Left channel was reading 25V and the one in the right channel was reading 177V....holy cow.  I turned off and took them out as soon as I could.....luckily....no smell...

Stuck back on the JJ goldpin ECC88 and the Balanced started right out below 500mV and keep going down steadily into +30 and -30mV on Left, and +150 to -150mV on right.  Balances was dropping steadily when started out 4.5V on Left, and 6.5V on right and drop quickly, probably within couple minutes and it dropped to 0.8V on Left and 1.5V on right, and keep dropping down.

I think these JJ goldpin are meant to be in there.  Better stick to what works as T2 is not something to be playing around with.  The next thing is probably a quad of Xf2 as everyone is saying they sound beautiful.  I have no choice but to get a Quad to get the best out of my T2 and while sticking to what work best.

I suppose there are two possibilities - 1) your JJ tubes are out of wack and you set the batteries taking into account their faults (and the sylvanias are fine), or 2) you need to look for a new tube reseller because this one is no good. I find a lot of resellers only test one side of the dual-triode, and I have seen some dual triodes measure new on one side and poor on the other side - this would cause problems in the t2. Not saying that is what happened here, but what you described shouldn't happen if the tubes are good.

If your tubes are good, you should be able to switch them without having to make any adjustments to the batteries. The only time I have had an issue was when the two halves of the dual triode were not reasonably close.

And in answer to your email, I have had bad tubes in my t2, but nothing where the bad tube took out any transistors.

 

 

[Whitigir]

I bought these off Tubedepot, and always checked the option of Balance and matching.  The only way to know is to buy new Quad matched elsewhere and try again, but if 1 of the JJ is out of whack, then swapping sides will prove that right ?

Even with JJ swapped side, the balanced and offset just swapped over.  I think you are correct as that this seller does not do Dual triode matching.  However, I will still gonna buy those Philips but from Other place this time

Edited November 2, 2017 by Whitigir

[joehpj]

I was thinking if I should use regulated heater for the 6922. Those tubes amplify small signals and the amp itself has very high gain.

And is there any problem to use same pair of leads for heating up the 4 front tubes?

[mypasswordis]

You just have to make sure everything can handle 1.5A. The regulators I am using can only do 1A each so I have to use two.

[JoaMat]

Managed to squeeze in the eight diodes under the lf353 op-amp.

Batteries resistors killing some 730 volt in series with the LEDs, or in my case lt1021, replaced with 10m90s.

Works great in layout software. In real life -