• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

valve buffer gainclone

Status
Not open for further replies.
Jump to Latest
#2
If you are making a stereo amp and want to use only one tube then the only one you can use is the ECC82. If you have pairs, then in theory you could use any of them, properly connected, and of course, with the proper heater supply.
Your CH84 is probably an ECH or UCH84 but the first letter got rubbed out (be careful, the markings tend to rub off easily on some tubes!)
 
#5
jaste said:
thanks! Since I´m totaly new at working with tubes, can you give me any advise on how to wire up the tube? Which pin does what?

Thank you in advance!
Click to expand...

:att'n:

If you've never worked with tubes before, make sure you read the sticky thread on high voltages and safety. Unlike most solid state equipment, tubs run at LETHAL voltages and can kill the unwary. Work with a companion, preferably an old guy who knows his way around the dangers, but at minimum, someone who can quickly turn off the power and call an ambulance.
 
#7
Not really, EL86 would be the tube equivalent to a low voltage, high current medium power transistor - it could be used in a buffer but that would be only if you really had nothing else to use. Even as a tube power amp it requires relatively non-standard operating conditions, which is why it is not very popular in that application, and why it is cheap. The EL84 would be the 'usual' tube for that power level, just check on prices for those and you'll know what I mean.

Perhaps the best tube to use as a buffer from your table would be the 6DJ9/ECC88 (and many other tubes from that same family!), although a 6AQ8/ECC85 would also do OK. The good thing about the 6DJ8 is that, as tubes go, it is a low voltage tube, which will reduce the potential of someone getting fatally electrocuted, and, it sounds good too 🙂
 
#8
Many thanks, I see Franz, used the ECC88 as well, doesnt seem too hard to get, even here.

These are the ECC88 types I can get, could you recommend any specific one maybe, based on brand knowledge or eperience...

ECC88/6DJ8 Various cheapies
6DJ8EG EI
E88CC Gold pins Mullard, Siemens
E88CC/6922EH SOVTEK
E88CC/6922EH SOVTEK
E88CC/6922 SOVTEK
ECC88 Mullard, Philips
E188CC/7308 USA
CCa Gold pins Siemens
6N1-P Svetlana
 
#9
7308 seem to be preferred (but also NOT cheap) although I had very good results with Siemens E88CC (normally also not at all cheap). You may add PCC88 to your list as a cheap one that may actually still be quite decent. In most cases you can put a PCC88 instead of an ECC88 and have it work perfectly (though substitution the other way is less likely to work). I have had used Ei PCC88s out of old TV sets work as good as some NOS even though they were probably run right to the limit for who knows how many years, so assuming they still do them like that or better, Ei might be a good but acceptably priced choice. You may want to look into getting a few cheapos just to experiment with till you get everything right.
 
#11
Try ECC86 too

Hi,

I've used ECC86, which was specifically designed to run on low
voltage, in my chip/tube hybrid amp. Works great, but I haven't
compared with ECC88 yet. ECC86s tend to be very cheap. The
ones I got were marked Haltron; I haven't compared brands though.
The ECC86s operate at a max voltage of 30V across them and 2-4
mA or so (much like a small JFET like the 2SK170GR, which is also
nice), whereas an ECC88 only just starts to be usable at 30V
(grid current may start to become a concern, or so I've heard).
The '86 and the '88 have the same pinout, heater voltage, and usually
about the same heater current, so the types can be swapped without
circuit changes (within their common range of max ratings, of course).
With the ECC86, I would make sure there is at least 12V across the
valve at all times and run about 2mA quiescent.

Morgan
 
#12
ECC86 is a rather special triode, I did not mention it because it tends to be almost impossible to find in some places of the world - including where I am. I have not seen a single one on offer or even just seen one, used or not. It was originally intended for use in car radios and even the ones I have seen had vibrator power converters and ordinary tubes.
 
#13
Just a quick comment on ECC88 types. I measured quite a few of them and found:

Siemens ECC88, 887 date code, 8 tubes (16 triode sections) measured, all wildly out of specification (mu of 20-23).

CCa, 4 tubes (8 sections) measured, lowest distortion of any ECC88 type, all microphonic.

GE 6KN8, 20 tubes (40 sections) measured, high mu version (mu = 45), low microphonics, moderate distortion.

Amperex E88CC (various, 20 sections total measured) had low distortion, though a bit higher than CCa, low microphonics. Used tubes with high hours pulled from Tek scopes measured the same as NOS. My first choice for a preamp tube.
 
#15
ECC88

Trio = 3.
Each C in ECC88 stands for a triode, 3 pins, tri(3)-diode
Tri(3)-ansistor also have 3 pins B, C, E - base collector emitter

Each triode has G, S, E - gate source emitter = B, C, E

The other 2 pins, to make it 8 pins, is for heater.
A tube needs to be warmed up to work.
So these 2 heater pins is connected to a separate heater circuit.
When tube gets warm, hot, it begins to glow
and can start doing work with those other 2x3 pins.

So for a stereo, 2 channels, you only need 1 tube ECC88
because have 2 triodes inside.
 
#17
lineup said:
ECC88
...Each triode has G, S, E - gate source emitter = B, C, E
Click to expand...

That would be A(node) - also known as P(late), G(rid) - NOT gate!, C(athode) - also sometimes K instead of C.
The closest equivalent is a N channel JFET, S(ource) ~~ C(athode), G(ate) ~~ G(rid), D(rain) ~~ A(node).

The difference of course is that the Cathode needs to be hot (several hundred deg. C) to start emitting electrons, so you have two extra pins (in the case of the ECC88) that power the heaters for boh cathodes, since ECC88 has two triodes in it.

The ECC88 has 9 pins, the remaining 9th pin (2x3 for triodes + 2 for heaters = 8) is a shield structure between the two triodes, that is usually grounded in an application such as yours.

European tube markings are usually quite logical. In the case of the ECC88, this is how it works:
E = 6.3V heater
C = triode, CC = double triode
8 = series 8, this usually means noval socket tube, noval = 9 pins (but there are some exceptions)
8 = type number 8 (here you need a databook to see what it is actually, with the rest of the markings you can get a fairly good idea, especially if you have the actual tube).

The US equivalent would be 6DJ8, of which only the first number (heater approximately 6V) gives you some real data, the last number means 8 elements are brought out to pins, heater counts as one even though it uses 2 pins - so, heater, A1, A2, G1, G2, C1, C2, shield, total = 8.

For numbered equivalents, such as 6922, you need a book to decode what they are...

A typical small triode needs about 10-30 seconds to start working right after heater power is applied, although, it can take several minutes for the tube to get to full operating state, this also depends on heater characteristics as well as heater supply mode (constant voltage or current).
 
#18
SY said:
Just a quick comment on ECC88 types. I measured quite a few of them and found:

Siemens ECC88, 887 date code, 8 tubes (16 triode sections) measured, all wildly out of specification (mu of 20-23).

CCa, 4 tubes (8 sections) measured, lowest distortion of any ECC88 type, all microphonic.

GE 6KN8, 20 tubes (40 sections) measured, high mu version (mu = 45), low microphonics, moderate distortion.

Amperex E88CC (various, 20 sections total measured) had low distortion, though a bit higher than CCa, low microphonics. Used tubes with high hours pulled from Tek scopes measured the same as NOS. My first choice for a preamp tube.
Click to expand...

Interesting, I had 4 E88CC by Siemens, which were used and not only spot on with mu, but also had the best match between sections.
I also have a number of Mullard 6DJ8s which came used out of two scopes, with very high hours, that also measured as NOS, except two which were used in a gyrator circuit and obviously driven well into maximum current ratings. On the other hand i have a NOS Amperex 6922 that has a good 25% difference between sections and mu is on the lowish end on the weaker one - and a NOS Amperex 7308 which is absolutely perfect.
I have also had a box of 144 East German (?) NOS PCC88 marked RSD (although i suspect they are probably made in Poland or Chech/Slovakia), which were absolutely horrible. About 1 in 10 had proper specs, with things like one section having a mu of 17 and another a mu of 22 in the same tube coming in 5 out of 10 cases. It turned out that heaters were the problem - in this particular tube the sections have heaters connected in series, and in many you could actually see they glowed with different shades of red. Heater voltages for the required 300mA current varied wildly, from 5 to 10V. Usually upping the current to about 330mA got them to the low limit of spec, but I got rid of the lot, i think I only kept a couple selected ones.
 
#19
There is evidently a difference between Siemens E88CC and ECC88! I've got some old VS Russian 6922 which show horrible mismatch due to heater imbalance, but I only had 2 of them, so who knows if they're all like that?

FWIW, the only RSD tubes I have are ECC81 and they are also just terrible. Only the GE JAN and 6201 measured worse...
 
#20
franzamp.gif


Can you help me identify the pins, I marked them as ABC on the picture...
 
Status
Not open for further replies.
Top Bottom