DIY mini T2 Build Thread

[micon21]

jamesmking,

Please write the gist of the 6.3 volts.  I'm sorry, I don't understand everything.  When you turn on your amplifier, you don't supply 6.3 volts at once to legs f5 and f4 of the 6922, but much less?
And what happens at this time with the EL34 lamp heating?

[jamesmking]

On 2/16/2022 at 8:27 PM, micon21 said:

jamesmking,

Please write the gist of the 6.3 volts.  I'm sorry, I don't understand everything.  When you turn on your amplifier, you don't supply 6.3 volts at once to legs f5 and f4 of the 6922, but much less?
And what happens at this time with the EL34 lamp heating?

I run the heaters at about 6VAC which can increase the life of the valves. I have resistors on the 6922 heater transformer outputs because the transformers where outputting about 6.6VAC which is too much.

(You need separate transformer windings for the 6.3VAC heaters on the el34s in each channel and separate from the 6922s heaters. The el34s have their heaters at about at about -400V.) The EL34s draw far more current and the voltages without any dropper resistors are about 6VAC for the transformers I purchased.

On switch on all valves are provided with heater voltages simultaneously from the low voltage transformer and the opamp dc balance and dc offset servos are powered on from the +-15V golden reference lv or simple lv in the golden reference hv board.

Only the high voltages are delayed in order to allow the valves cathodes to heat up and develop a space charge before the high voltages are activated. I use a one minute delay. All other voltages pass through the delay board without being switched or delayed.

Here is a simplified block diagram for the wiring I used (there are other ways of doing this).

 

Edited May 19, 2022 by jamesmking

[micon21]

I assumed that the circuit with the two chips also deals with the gradual heating of the lamps, from about 2v  and up to 6.3v/ 15....30sес.

But it turns out that this is not the case.

https://ibb.co/NW7GQwf

then it is not clear why so many radio parts are involved.

sorry.

 

[jamesmking]

25 minutes ago, micon21 said:

I assumed that the circuit with the two chips also deals with the gradual heating of the lamps, from about 2v  and up to 6.3v/ 15....30sес.

But it turns out that this is not the case. then it is not clear why so many radio parts are involved.

sorry.

 

two chips?? do you mean the op-amps on the amplifier boards - they are for the dc servo and balance servo. The heaters will pull more current when you first switch on since the heaters resistance will be lowest when they are cold. As the heaters heat up the resistance of the heaters goes up and so they draw less current until they reach a stable current draw.

or do you mean the chips on the golden reference low and high voltage boards? they are 10V high quality voltage references used as part of the voltage regulation and control circuits. The opamps on the golden reference lv boards are used to compare the voltage reference against the output voltage (divided down by a set of series resistors) and adjusts the output voltage accordingly.

the other chips on the amplifier boards are also voltage references (the same as in the golden reference lv and hv boards) and are they are used to control the constant current anode/plate supplies to the el34s. The final chip on the amp boards is an opto coupler to provide high voltage isolation between the el34s and the Dc offset servo.

 

Edited February 16, 2022 by jamesmking

[micon21]

I bought the EL34 a long time ago, but I haven't bought the tube  6922 yet.

Please recommend a brand for the 6922.

 

P.S.

I know a good specialist, he is not greedy and also shows his ideas.   The master in 2019 offered an interesting variant of the amplifier for electrostats.  https://ibb.co/SRpSDnJ

But unfortunately I could not buy 10m90s so I decided to try to repeat the mini T2, where in the amplifier itself 10m90s is not necessary.

[jamesmking]

5 hours ago, micon21 said:

I bought the EL34 a long time ago, but I haven't bought the tube  6922 yet.

Please recommend a brand for the 6922.

 

P.S.

I know a good specialist, he is not greedy and also shows his ideas.   The master in 2019 offered an interesting variant of the amplifier for electrostats.  https://ibb.co/SRpSDnJ

But unfortunately I could not buy 10m90s so I decided to try to repeat the mini T2, where in the amplifier itself 10m90s is not necessary.

Everyone has their own tastes in sound. In my experience the manufacturer of the valves can radically change the sound of stax valve amps. I mostly listen to classical and jazz and want a warm detailed sound. I am very intolerant of hard bright metallic treble. I like dynamics - both micro and macro and am fairly tolerant of sound stage width, depth and imaging. I listen to many historic recordings and I don't like overly analytical sound balances that rip apart old recordings... but everyone has their own priorities and listening preferences.. So please don't spend money on 6922s just based on my experiences.

 

I tried gold pin electro harmonix 6922. They sounded ok but I had one go noisy and the two triode halves on almost everyone of the 12 I purchased were very miss matched (the seller did not offer matched triodes) and they almost all tested  very weak <50% on my avo IV valve tester and they quickly got weaker with use. I don't know if I hot a bad batch but the online shop only offered a 3 month warrantee. I have not tried buying any from another shop, until proven otherwise I consider them an unreliable waste of money. I wonder if they are actually 6922s and not something else.

I tried some NOS philips e88cc - tested much stronger and better balanced triodes but I found them to not have a lot of bass and somewhat bright for my setup.

At the moment I am using genalex gold lion 6922. Fairly well matched triodes and all tested close to 100% on the avo. They are quite dynamic and have good deep bass the treble is extended, but they are not cheap. 

 

for EL34s I tried electro harmonics - awful, transistor sounding hard bright metallic and ran very hot. cheap and useful for testing on initial start-up but not for listening.

I tried grove tubes EL34s these sounded ok but the price recently went up by 25% turning them from a cheap bargain to more expensive then most other non nos options.

currently I am running tung-sol el34s. In the uk they are cheaper than the groove tubes and have deeper better bass, more dynamics and imaging. However I did have one fail - the anode sparked fortunately no damage was done to the diy T2. Again the uk shop only offered a 3 month warrantee and the failure was after 3 months and two weeks...

I now buy all my valves from partsconnexion - which even without sale prices and with fast  including shipping it is still cheaper than buying in the uk, they offer 6 month warrantee and often have 20% off sales....

 

many transistors required for stax amp builds have been out of stock for 6 months or more and probably will be out of stock for another 6 months to a year.. its the great transistor and semiconductor shortage.... you just have to be patient or perhaps someone on the forum has spares they could sell you.

 

Edited February 17, 2022 by jamesmking

[Pars]

Not Stax T2, but when I used to have a Counterpoint SA5.1 preamp which used 6922s, the favorites that I used in it were some Amperex PQ white label 6922s. I would guess these are a) hard to find and b) expensive as fuck now. I also used the EH 6922 gold pins and liked them as well, but not as good as the Amperex.

[Kerry]

I tend to use Amperex for the T2 but also like the Jan Phillips 6922 for a change of pace.

[audiostar]

On 11/23/2020 at 10:58 PM, JoaMat said:

Lately the left channel humming has increased to become even more bloody annoying. 

But… today I cured it (pure luck - trial and error approach). I have two umbilicals between psu and amplifier. Both carrying 580V bias and then joining at the stax connector. So, I removed left 580V bias wire in amplifier and the hum disappeared. To my ears the amplifier is now dead silent.

To circumvent this loop issue and to not sacrifice the dual mono nature of this amplifier, maybe the left Stax connector should be fed by the Bias from the left umbilical and the right Stax connector fed from the Bias coming from the right umbilical. Would be my solution to this.

[JoaMat]

Seems to be a good solution. Thank you.

[micon21]

  

Thank you mwl168 for the opportunity to get the mini T2 boards. the package from mwl168 arrived on february 22nd, and on the 24th my country was attacked by the enemy. and everything went into the background, the country was fighting for survival. thanks very much to many countries, especially the USA, very much helped and helped. now there is already hope that my country will survive, and i hope that i can solder a mini T2 and listen to the potential of this version.  

Translated with www.DeepL.com/Translator (free version)

[JoaMat]

I’m happy to hear you from you, @micon21 and I wish you and your country all the best in the fight with Mr. Putin. You will survive. In the end the good will defeat the evil.

[Sherwood]

On 5/10/2022 at 1:58 PM, micon21 said:

now there is already hope that my country will survive.

Слава Україні, і удачі з Кишинева.

[Helium]

Not sure if DIY forum where people come together to share interests is an appropriate place for such things. There is too much hostility and hate in life nowadays, don't bring it in here.

[n_maher]

1 hour ago, Helium said:

Not sure if DIY forum where people come together to share interests is an appropriate place for such things. There is too much hostility and hate in life nowadays, don't bring it in here.

Take a breath and stop issuing orders when you are not empowered to do so.  I see nothing wrong with the posts here.

[Helium]

Ok, no one is issuing orders (except that taking a breath can be regarded such). Why such tone?

[Sherwood]

On 5/16/2022 at 3:03 PM, Helium said:

There is too much hostility and hate in life nowadays

Don't I know it.  Still, so much death right next door influences every part of life.

 

On 5/16/2022 at 3:03 PM, Helium said:

don't bring it in here

That is grammatically an imperative, so, an order.

I have no fight with you, Helium, I'm just expressing a little solidarity with a fellow forum participant in my neighborhood.  That can't possibly be what passes for hate.

[Voltron]

On 5/16/2022 at 1:04 PM, Helium said:

Ok, no one is issuing orders (except that taking a breath can be regarded such). Why such tone?

I would have called you a cunt for your prissy comment and imperious demand, so n_maher  and Sherwood have let you off lightly by Head-case standards. So take a breath or take a hike. 

[jamesmking]

3 hours ago, micon21 said:

Hello everyone.

I finally had the opportunity to do a mini t2. to begin with soldered only one board, that is, one channel.

iMaybe someone made a record of the voltages at different points on the board. 

normally I would be happy to help, but I am at my parents and don't have access to a mini T2 to measure until I return to my home.

im not sure of your level of experience, so my apologies if some of this is obvious

0. be aware working on a powered amp exposes you to the risk of electric shock and you should only be doing this if you know how to be safe and have a good quality multimeter.

1. I would build and check the psu is working without it connected to the amp. I assume you have build golden reference high voltage supplies. Advice on these can be found on the thread for the gr hv.

2. before i would think about powering the amp board I would carefully inspect the solder joints looking for bad joints and solder bridges with a handheld magnifier or better a microscope.

3. I would verify as far as possible the component placement, orientation vales are correct. If you have a set of very fine spring loaded probes like those sold by probe master you can check for continuity/shorts between components referencing the schematic.

4. i would strongly suggest that testing is initially done on using variac and slowly bringing up voltages to their normally working levels, measuring psu rails and looking out for smoke, bad noises etc.

5. i would strongly suggest using cheap valves for initial power ons until you are confident the amp is stable and working correctly.

 6. I would monitor the dc offset (dc voltage between + audio output and ground) and dc balance (dc voltage between - audio output and + audio output)

7. I would not think about connecting expensive headphones until I have put in some audio test signals and looked at the output on a oscilloscope and had the amp running for at least a few hours.

8. if you have a temperature probe, its also worth measuring the transistor case temperatures and the temps of any high wattage resistors.. 

if you still need specific voltages I should be able to help after jan 9th

good luck

 

 

Edited December 28, 2022 by jamesmking

[Pars]

One other thing I would add to James' excellent check steps would be to check the SMD resistors to make sure they aren't shorted. I had this happen with a couple of 0805 resistors on a mini dynalo, and could not see anything under magnification (OptiVisor 10X lenses). This might not be an issue with 1206/1210 packages though, or as much of one. Mine were all hand soldered. Using paste and a stencil might make this a non-issue.

[micon21]

THANK YOU for having someone willing to help me. 
I have a lot of experience in electronics.  i got my first static headphones in 2015   KOSS 950, used, I made a transistor version. I did not like the sound of 950, I bought a Perun, then bought used SR-007 and soldered a tube amplifier. I currently have   009S. used.

I can not buy amplifiers for 2000...3000$ in Ukraine 200$, it is like in the usa 4000...5000$.

Very glad I found your theme on the mini T2. all that is necessary for this version was bought.
But the principle of the mini T2 is not clear to me, as a strange idea. for this I did not buy a board for the power supply. I did not buy the case. understand me correctly. do not want to spend money when you can not be sure that everything will work.   
I made all the stabilizers on a simpler circuit, with much less cost for it. i feed the incandescent separately, manually for now. after 60 seconds i feed all the other power.   +220, +400, -460 do not come at once, they come gradually, with a rise in 100 seconds. 

[micon21]

The circuit I showed as an example. for mini Т2 voltages are different. I made three of these, of course first adjusted the necessary voltages.  

I have an oscilloscope, it will tell me if the board is working or not.  

there is a jumper on the board, SERVO  ENABLE, should it be shorted the first time it is turned on? if it is not connected, what happens?

 

 

Edited December 29, 2022 by micon21

[jamesmking]

5 hours ago, micon21 said:

The circuit I showed as an example. for mini Т2 voltages are different. I made three of these, of course first adjusted the necessary voltages.  

I have an oscilloscope, it will tell me if the board is working or not.  

there is a jumper on the board, SERVO  ENABLE, should it be shorted the first time it is turned on? if it is not connected, what happens?

 

I understand correctly, the signal goes like this?

the automatic servo is limited in the range it can compensate for. So removing the jumper allows you to manually adjust so the servo does not need to do much work then you can add the jumper to activate the automatic servo.

 

[jamesmking]

3 hours ago, micon21 said:

after full power on, on the output, tweaking 10k, makes it possible to make a minimum of +3.5v and -4.1v. then fed the signal 1000Hz  1v - on the output received a very small signal amplification, no more than 5 volts...

have you designed the amp side as single ended i.e. rca input or balanced i.e. xlr input?

if you are running single ended then the - input side needs connecting to ground.

assuming that is ok next check the input (i.e. grids) on the 6922s that connect to the inptut and check you actually have a signal on both.

next check the input i.e. grid of the other 6922s to check you have an input to them

next check the grids of the el34s to check they have an input

this should give you an indication of where your signal is disappearing.

I assume you scope is set for 10M input and NOT 50ohm... also I assume you scope probes are set to x10 so you don't overload the front end of your scope

I also assume you have checked all the dc voltage inputs to the amp board and they are correct.

have you got a volume pot on the input or are you feeding 1Vrms directly into the amp?

if i remember correctly the min t2 is about x1000 gain

 

 

 

Edited December 29, 2022 by jamesmking

[jamesmking]

7 hours ago, micon21 said:

 

 

I understand correctly, the signal goes like this?

no thats just the feedback loop from the output to the input valve. the actual audio signal path is far more complicated.

the output is provided by the el34 valves. Powered from constant current devices connected to their anodes/plates.