hello,
I've got a strange problem in the powersection of my 36 watt. And I can not grasp the reason why it is happening or what is causing it. (man... the houres that I've spend on trying to figure it out....)
I've used the schematic that is posted on the site off 18watt.com. The connections of the powersection of the amp (screen, anode, cathode, etc...) are exactly the same.
It seems that only one side of the push-pull section is conducting.
I'll try to explain: if you have the OT, I get 333v at the central tap (where the B+-line is connected). One side of the OT (left side) is 297 volts. This is the side where the 2 powertubes are conducting. The other side (right side) of the OT stays at 333 volts, meaning no current is going through these two tubes.
I have about 24 volts on the cathodes. Ive used 180 ohm resistors, so 133 mA of idle current is going through 2 powertubes. Way to much according to the datasheet of JJ EL84.
Another strange thing is that it's not always the same 2 tubes that are conducting. When the amp is switched off and on again, it can happen that the other side (right side of the OT so to speak) is conducting and the left side is 'dead'.
If I raise the cathode resistors (say from 180 ohm to 280 ohm), the voltage goes up keeping the idle current at 133 mA. So baising this thing is impossible.
I never get all the 4 pentodes to conduct at the same time. Only one paire off tubes.
Has anyone experienced this type of strangeness? Or has anyone have a clue of why this could be happening?
I've changed the powertransformer (now a hammond 275v - ct - 275v at 200mA), I've resoldered all of the joints, I've looked over all of the wires, groundings, etc... but I can not find the problem. I've even swapped powertubes in the hope that some kind of order of powertubes could make this thing going, but no dice. I've measured the OT for any broken windings, but everything seems ok.
I really could use some help here. All possible suggestions are welcome.
one half off push-pull works, other half is dead
Moderators: pompeiisneaks, Colossal
Re: one half off push-pull works, other half is dead
What are you using for screen resistors? Any grid stop resistors? There might be oscillation going on here. How much filtering do you have?
Re: one half off push-pull works, other half is dead
Screen resistors are 1k / 5 watts.
Grid stop resistors are 8k2.
I don't quiet understand what you mean by 'filtering'.
If you refer to the elco's used in the B+ line: two 32µF/450V elco's and two 22µF/450V elco's. I don't use a filterchoke. It's a resistor: 1k/10watts.
Grid stop resistors are 8k2.
I don't quiet understand what you mean by 'filtering'.
If you refer to the elco's used in the B+ line: two 32µF/450V elco's and two 22µF/450V elco's. I don't use a filterchoke. It's a resistor: 1k/10watts.
Re: one half off push-pull works, other half is dead
I just gotta ask, do your grids have a DC reference to ground? How?
Re: one half off push-pull works, other half is dead
Yes, I was speaking of the main filter caps.
I'm guessing that there is oscillation happening IF different sides of the output section work / don't work. Sounds like you really need to scope things out here.
For grins, you could always try real high grid stoppers (on the the tube) say 100k, just to see if that helps isolate problems.
A signal generator and a scope would help with checking the OT too, but a 20 v p-p signal is a long way from B+ with a signal (if there is breakdown inside the OT at high voltage).
Jana has a good point about the ground ref for the grids too. If you do try high grid stops like I suggested above, make sure your total grid to ground resistance isn't above 220k or so.
John
I'm guessing that there is oscillation happening IF different sides of the output section work / don't work. Sounds like you really need to scope things out here.
For grins, you could always try real high grid stoppers (on the the tube) say 100k, just to see if that helps isolate problems.
A signal generator and a scope would help with checking the OT too, but a 20 v p-p signal is a long way from B+ with a signal (if there is breakdown inside the OT at high voltage).
Jana has a good point about the ground ref for the grids too. If you do try high grid stops like I suggested above, make sure your total grid to ground resistance isn't above 220k or so.
John
-
Andy Le Blanc
- Posts: 2582
- Joined: Sat Dec 22, 2007 1:16 am
- Location: central Maine
Re: one half off push-pull works, other half is dead
How are you setting the amp up to bias? Is there a load on the output?
What's is the output tranny? Does it do this with the inverter tube removed?
I've not had an amp exhibit this the way you've described. I want to think
that it might be the inverter circuit. The power side of an amp isn't too
complicated, there's only so many things to check, and it shouldn't "flip-flop".
It could be a very low frequency motorboat, which might mean that the filter
capacitors might be below a critical value???
What's is the output tranny? Does it do this with the inverter tube removed?
I've not had an amp exhibit this the way you've described. I want to think
that it might be the inverter circuit. The power side of an amp isn't too
complicated, there's only so many things to check, and it shouldn't "flip-flop".
It could be a very low frequency motorboat, which might mean that the filter
capacitors might be below a critical value???
lazymaryamps
Re: one half off push-pull works, other half is dead
Hi,
I could be on to something here.
The ref resistors to ground for the grids are 470k.
But... I've noticed I soldered them before the grid caps (so they are between the anode of the drivertube and the gridcap to the pentodes). And they should be attached after the caps (DC blocked by the caps - so nothing is loading the signal now). Maybe that's the cause??
I'll fix it this evenening, and let you know.
I hope this is the issue that I'm having.
Perhaps I'll lower the grid to ground resistance to 220k in total as you suggested John_P_Wi. Is there a reason why this should be maximum 220k?
I could be on to something here.
The ref resistors to ground for the grids are 470k.
But... I've noticed I soldered them before the grid caps (so they are between the anode of the drivertube and the gridcap to the pentodes). And they should be attached after the caps (DC blocked by the caps - so nothing is loading the signal now). Maybe that's the cause??
I'll fix it this evenening, and let you know.
I hope this is the issue that I'm having.
Perhaps I'll lower the grid to ground resistance to 220k in total as you suggested John_P_Wi. Is there a reason why this should be maximum 220k?
-
Andy Le Blanc
- Posts: 2582
- Joined: Sat Dec 22, 2007 1:16 am
- Location: central Maine
Re: one half off push-pull works, other half is dead
Its a piece of tube data , Grid-No.1-Circuit Resistance. It depends on the type of bias being implemented. The max El84/6bq5 rating is 0.3M for fixed bias, 1M for cathode bias. The other grid resistors, connected between these and the grids, are the grid stops. You need to add these to find the total circuit resistance.
That would do it.But... I've noticed I soldered them before the grid caps (so they are between the anode of the drivertube and the gridcap to the pentodes). And they should be attached after the caps (DC blocked by the caps - so nothing is loading the signal now). Maybe that's the cause??
lazymaryamps
Re: one half off push-pull works, other half is dead
Violetv,
Andy explained it well about the grid leaks to ground value. When something is amiss, I like to lower values as a testing method to help rule out oscillation.
Changing these values will change the feel and tone of the amp, as Andy pointed out, don't exceed the tube datas maiximum values to ground.
John
Andy explained it well about the grid leaks to ground value. When something is amiss, I like to lower values as a testing method to help rule out oscillation.
Changing these values will change the feel and tone of the amp, as Andy pointed out, don't exceed the tube datas maiximum values to ground.
John
Re: one half off push-pull works, other half is dead
Hi,
thanx for the hint / advice about the grid resistors. That solved the problem alright. My amp now works like a charme. !!!
I never would have thought that the problem was in that area of the amp. I've been soldering an re-soldering the poweramp, measuring everything there is to measure and pulling haires like a mad man. I was on the verge of buying another output-transformer, because I thought that might be the problem.
I didn't come across anything about the grid to ground resistor. I've been reading books from the librairy, searching the net, doing the math and drawing loadlines like suggested in many articles on baising. But nothing has pointed me in the direction of grid resistors.
A very important lesson learned. Thanx again.
thanx for the hint / advice about the grid resistors. That solved the problem alright. My amp now works like a charme. !!!
I never would have thought that the problem was in that area of the amp. I've been soldering an re-soldering the poweramp, measuring everything there is to measure and pulling haires like a mad man. I was on the verge of buying another output-transformer, because I thought that might be the problem.
I didn't come across anything about the grid to ground resistor. I've been reading books from the librairy, searching the net, doing the math and drawing loadlines like suggested in many articles on baising. But nothing has pointed me in the direction of grid resistors.
A very important lesson learned. Thanx again.
Re: one half off push-pull works, other half is dead
FWIW 220k-270k is a more normal value (than 470k) for output tube grid load resistors
Re: one half off push-pull works, other half is dead
[quote="tubeswell"]FWIW 220k-270k is a more normal value (than 470k) for output tube grid load resistors[/quote]
Noted. I'll have a try to see how this influences the feel of the amp.
Noted. I'll have a try to see how this influences the feel of the amp.