• 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.

Gu50 output problems

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I was building a kt88 50 watt amp.

As my test quad of kt88 turned out to be toast, i needed new valves.

I was repelled by the ugliness of most kt88 and the price of others. (i have little money to spend)

Somebody offered me hit stack of GU50 pentodes for a reasonable price. I am not too bright with tubes yet, but after some advice from others i decided to alter the amp for the tubes

I made a 250Volt power supply for the G2 as the datasheet said.

I powered it up and something strange is happening:

I hooked up a tone generator, and the opt started to sing with the frequency applied...

i measured the current through the tubes, and one tube has a hard time stabilizing the current as the other seems to draw no current at all.

The opt worked fine with the kt88 tubes...

I have a B+ of about 500V, a g2 of 250 volt, i can regulate bias between -18 and -65 volts. The phase splitter is a long tail ecc99.

when i turn down the volume, the singing does too. But then at the background a "tick" apears. The tick varies in frequency when bias voltages are changed.

Does anybody have a clue what happens here?

I checked all the connections a 100 times, checked every solder joint.

i attached some of the datsheet

Bas
 
Bas, let's do some troubleshooting. A schematic would be very helpful, but I'll throw a few things out to try in general:

Pull the tubes. Using GREAT CARE, check the voltages at the sockets. Are the output tubes getting the proper bias voltage at their grids?

Disconnect the driver stage. If it's capacitively coupled, pull the caps from the driver end and ground that end of them. The idea is that you want the grids to be at AC ground but to keep the DC bias on the grids. Do things bias up OK with no singing? If so, you've eliminated the output stage and power supply. If nothing biases up properly, you've got some checks to do. Be suspicious of socket contacts.

If everything is OK to this point, verify that the driver stage is still working, then we'll assume you've got some positive feedback happening somewhere.
 
thanks!

I got it working now

some nasty odds here. I bought a lot of 20 tubes, supposedly good, new ones. As i discovered, some are in terrible shape! I changed the pairs 3 times. 3 times i got 1 bad and one good tube!

Now it is playing relatively good. I suspected some old stock couplin caps, so i changed them wit extremely cheap film caps of which i know the sound suffers terribly. I'll change them back and hope i get some decent sound.

I read somewhere that somebody made these tubes run in ultralinear. What would be the consequence if i hooke the g2 to the ultralinear tap?
I connected the g3 to the cathode.

I'm still very fresh on pentodes, but i'm excited i got them to work! (plus they look ultracool in their cast aluminium enclosures)

made a quick shot with my phone cam

Bas
 

Attachments

  • 560i0179.jpg
    560i0179.jpg
    38 KB · Views: 1,672
they are. This is my first amp i built, but i am faily shure these are fantastic matches to the GU50's

I have crappy speakers, the amp is still a little hummy, but i have NEVER heard such a detail in bass. The low end is do defined, that i can actually hear the amount of guitars playing in songs that i thought it was impossible to hear! (deftones)

My triodes are faily microphonic. Do you have an idea what a ood brand is? I ma planning on stocking up on different barnds to try out.

Bas
 
Now that the thread is on a roll, why not ask something else?

I have a loud, nasty hum in my right channel.

No obvious reasons to me. All kind of things could go wrong, but it is the exact mirror of the left channel

(i have been tinkering with this for weeks)

I am really on the verge of despair.

any tube switching is useless, pulling wires, touching wires, shielding wires, nothing works. DC is as clean as the other channel, input noise is comparable.

I had the same problem when it was running with kt88's.

I am about to jump to the conclusion that the wood i used is hummy or my right ear, as i can not find anything wrong in the amp.

I can measure the hum on the coupling caps. the waveform is like__/|__/|__/|_

Both "hums" are in opposite phase after the phase splitter.

As my first stage is dc coupled, i cannot disconnect it to see if the hum is from there (or can i?)

Could i disconnect the coupling caps?

I need a way of eliminating to find the spot where it goes wrong..

BAs
 
How are you powering the heaters? 7193s don't seem to like heater - cathode potentials much. I had a similar problem which i solved by grounding one side of the heater supply (not sure why only one channel was affected - put it down to slight variations between tubes).
Things won't be helped in your circuit by the fact that the ECC99 cathodes are at at fairly high voltage. You need to check the maximum heater - cathode potential allowed with these tubes. (you might need to use a separate heater supply for the ECC99s)
Whatever you do, the heater supplies need to be tied to either ground or something at a similar potential to the cathode. Don't just leave them floating or you will get hum problems
 
Ok..

Swapping the 7193 with a 6sn7 yields the same problem, and it' s heaters are referenced to ground with a center tap.

Th ecc 99's are referenced to gnd with a 1uF cap on the centertap, dunno why, but thatis what the schematic said. Tying it directly to ground made no diference. Both channels come fom the same winding..

The GU50's all have their seperate 12,6 volt dc supply (couldn't get 3A regulators, so used a 1,5per tube)

they are run from the same winding. How do i reference those? Would i get problems if i would hook em to gnd with a voltage devider? Or is grounding the centertap before rectifying enough?

Bas
 
VanguardVillage said:
this tube is very famous in our country,for it was for military use.it's called ry50/fu50.i believe the gu50 is not as good as ry50/fu50, wihch is one of the best tubes china had built.the other is fu5(=805),which is almost the best 805 in the world.


I'd assume the chinese version is made on (perhaps modified) russian tooling anyways.

The soviet version is indeed a duplicate itself.

German army was using it already in 1935 (?)

datasheet
 
beamnet said:
thanks!

I got it working now


I read somewhere that somebody made these tubes run in ultralinear. What would be the consequence if i hooke the g2 to the ultralinear tap?
I connected the g3 to the cathode.

I'm still very fresh on pentodes, but i'm excited i got them to work! (plus they look ultracool in their cast aluminium enclosures)


Bas


I wouldn't advise getting to UL taps until everything else in the amp works the way it should and the expected power is achieved and is clean......The reason is connecting g2 to UL can stir up more trouble with an RF tube like LU50. I've done several designs not using this tube but other high gm tubes which often oscillated...(and that was without global feedback) .....The solution is a g2 to anode snubber..... but estimating the correct R/C values requires experience.


richj
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.