|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
141.239.172.61
I just built a SE kt90 amp with 6j5 driver tube. The amp uses fixed bias. I guess i have 2 problems. first my bias voltage is not able to go low enough to get the power tubes to their proper cathode current. The bias circuit takes -60v to 4.7k resistor to 50K pot which is grounded with a 5k resistor. The wiper is connected thru a 13k resistor which goes to the grid of the power tube. This arrangement yields between -30 to -45 volts of bias voltage. I need to reduce the negative voltage to raise the current of the power tube. so, i am not sure which resistor/s to change to have a more effective bias voltage.
The next part of my problem is that i believe the plate voltage of the 6j5 tube is too low at 60vdc. The B+ to the plate 100k plate load resistors is 250v. I am following the RH universal amp version 2. I have built this amp many times now with mostly 6sl7 driver tube. And also a version with switchable 6sn7 and 6sl7 driver tube. I am following the exact same circuit as the switchable 6sl7 to 6sn7 except using the 6j5, which is 1/2 of a 6sn7. Not sure what is going unless it has to do with the fixed bias .
Well, i kind of ran out of ideas of what to do here so any suggestions would be most appreciated. cheers, Dak
Edits: 10/17/16Follow Ups:
I played with the bias circuit, removing the 5k ground resistor. This allowed the bias voltage to be less negative and thus raising the cathode current. In one channel the bias current i was able to adjust to 80ma which moved the tube out of cutoff and allowed the music through. So, i know that the bias system just needs to be tweaked to work ok.
The next part is the 6j5 driver tube. The music that is coming from the amp is very weak which i believe is the driver stage underperforming. While playing with the bias i also checked the voltages on the 6j5 and these are the values. Anode voltage is only 45vdc, roughly for both channels. Cathode voltage is about 1.7vdc from a 380 ohm resistor. The plate resistor is 105k while the B+ supply is about 275vdc. This resistance must be too large so i am going to change it. Any suggestions? I am thinking 33k which is pretty standard for 6sn7.
The complicating factor is the 6550 plate to 6j5 plate feedback resistor in this circuit which is 100k. I guess, i will install the 33k and see how it affects the driver circuit.
Edits: 10/20/16
190 volt drop across 100k is 1.9ma of plate current.
A 6j5 will not be very happy with only 1.9ma. of current. 8ma. would be more "normal".
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Actually, the driver stage is an "RH" type , i.e., there is plate fb from the output tube anode to the anode of the 6j5. The fb resistor is also 100k. So, wont that make the plate load 1/2 of just the single 100k resistor?
Significantly, when i touch the input grid of the 6j5 with my DMM probe there is no sound. nothing, no click, hum, or buzz. zip, zilch, nada. I measured about 60vdc on the plate, and IIRC, 2vdc on the cathode, but i can't find my notes about that measurement, so not sure.
The odd thing is that i built this same circuit with 6sn7 instead of 2 6j5 and it worked fine. Pretty much the same part values and the same circuit but , that amp did not have fixed bias, it was cathode biased.
Edits: 10/19/16 10/19/16
"Significantly, when i touch the input grid of the 6j5 with my DMM probe there is no sound. nothing, no click, hum, or buzz. zip, zilch, nada."
That's really strange.
I assume you bias supply is referenced to ground.
I think you said the grid to bias pot resistor is only 13k?
This is not an answer to why you are getting no sound but why such a low value grid resistor?
There has to be something mis-wired.
Do you get anything using you meter at the output tube grid?
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
bias supply is referenced to ground via 5.6k resistor
grid supply resistor of 6550 is 51k
When measuring the volts on power tube input grid (about -30vdc) there is definite tick when contact is made. That would indicate that the output tube stage is ok
What is the value of the cathode resistor for the 6j5 and what is the cathode voltage?Did you check/change the coupling cap? It might be open.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 10/20/16 10/20/16
The voltage drop is just a resistance ratio with -60 in and you want -20 out you need 2/3 of the resistance above and 1/3 below. A 50K pot will do this given your other resistors.
At full right rotation you will have 4.7K / 55K or 92% (-55 Volts)
At full left rotation you will have 54.7K / 5K or 8.3% (-5 Volts)
A 5K pot will give you somewhere the range you are getting. You might have a different value resistor that you think as well
At full right rotation you will have 4.7K / 10K or 68% (-40 Volts)
At full left rotation you will have 9.7K / 5K or 34% (-20 Volts)
Check the values of your resistors and the pot. Some of them are wrong.
Hi Chip, almost 100% sure that it is 50k Allen Bradley linear. I have the wiper going to the 13K resistor to the grid. Does it matter which leg is going to ground? I did not think it matters.
I will recheck the resistor values. It is not the first time that i have messed up with that. cheers.
Edits: 10/18/16
a 5K or 10K-pot will give you finer control, in the C- range needed to bias the tubes. Once, you get the set resistors closer to the ideal negative voltage. Play with the circuit-to-ground resistor's value (prolly going to be around 20-30k-ohm), until you get close to nominal C- voltage (prolly around minus 55-60VDC). Then, insert the 5k or 10k-pot.
Good luck!
But, whether or not that is the right value to achieve my parameters, i am not sure as i am pretty much following my other fixed bias amp which had a 50k pot for the voltage adjustment. i admit that the correctly working circuit is not the same as this circuit. This new circuit i chose because it has a fail safe feature. I don't know if it has been vetted by other users. Theoretically, does anyone see why it can not work? I left many of the resistors in the bias circuit unsoldered in anticipation of changes. cheers and thanks for all the suggestions.
solve one problem at a time. Measure voltage at each junction in your fixed bias circuit then draw it out, resistances and voltages. The bias circuit draws almost no current so the voltage out will be the resistance ratio. You may have an unsoldered connection making a poor connection
I just saw this. Isn't the amp cathode-biased, Class "A?"
No, i am doing it with fixed bias. Same thing you just bias the amp up with a meter which measures the current on the cathode. One can dial in the amount of bias you want. Also you can have different power tubes and simply and easily adjust the current to suit the different tube's needs.
Are you in intending to run this SE amp in Class "A?"Thanks!
Edits: 10/19/16
you may get more feedback if you post this over at Tube/DIY where that is what they talk about over there.
so i figure i would post here. Maybe some of the not the usual suspects will have some keen insight?
Tube DIY is about those who build their own tube amps and discuss circuitry, diagnostics and assembly challenges like yours.
Engineers and tinkerers, not just users who want to compare their favorite brand of tubes.
And does it matter that you all are exposed to some of the practical side of tubes? If this is a big no, no, i would not take offense if the moderator moves the discussion to a more appropriate venue. thanks for the comments.
to get input from the more builder oriented group. :)
A picture IS still worth a thousand words. Draw up a schem of one channel and of the psu, take pictures of same with your phone camera, and post them here. Much more likely to get helpful info that way.
Edits: 10/18/16
I'd manipulate the 5k-ohm resistor on the 50K pot. You might want to change that setup to, say, a 25 to 30k-ohm circuit-to-ground resistor and a 10K-ohm pot. Prolly give you more sensitive range of control.
You have a schematic of this amp? Pictures worth a thousand words, kinda thing.
8^)
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: