Dips, swells weird distortion help!

[markammon1979]

Hi I wonder if anyone can help me. I recently converted an old audio amp into a 5E3 deluxe. It works fantastically apart from after half an hour or so playing at near top volume the sweet distortion sound goes weird! The amps sound dips, swells and the distortion sound becomes just plain weird. I have changed all the tubes (all 50's 60's NOS) apart from the rectifier tube. The rectifier is an old 5y3gt made in the 50's by GE. Could this be the problem? Is the rectifier playing up when it gets too hot? I do not have another to test. The only other rectifiers I have are a NOS GE 5u4 or a Mullard GZ34. Can either of these be used for testing purposes?

[Paul P]

...The only other rectifiers I have are a NOS GE gu4 or a Mullard GZ34.
Can either of these be used for testing purposes?

I wouldn't use your Mullard GZ34 for anything other than in an amp that is
known to be in perfect order and probably would only use it for special
occasions at that. I don't think there are many of them left and they're
very expensive if you can find one. Seems to be the Holy Grail of rectifier
tubes in many people's eyes.

Paul P

[markammon1979]

I also have a 5u4GA rectifier. Could that be substituted for testing purposes?

[markammon1979]

voltages too high?

I checked the voltage on the amp yesterday and the power tubes were measuring about 420v which I believe is very high. I've put a resistor in to lower the voltage to about 340v and I'll see how that goes. Could the high voltage have been causing this?

[tubeswell]

Just over 400V might be okay if your Cathode resistor is big enough. Tube performance is not just the product of plate voltage. Rather it is the difference between the cathode and plate voltages that determines the plate dissipation (as well as the Cathode voltage and Cathode resistance).

Cathode Voltage / cathode resistance (in Ohms) = tube current (in mA).
Tube current (in mA) x (diff. in plate and cathode voltages) = plate dissipation (W).

Check your voltages at idle when you just started it up, and then again at idle when you got it going nice and warm, and see if they are any different. More likely could be either resistors or coupling caps in the circuit are overheating and changing value or malfunctioning when its hot. Resistors need to be specd to a certain power rating for the job. What are the Wattage ratings of your resistors throughout the amp?

Or could be a coupling cap is leaking too much DC into the grid of the next stage - to check - take the 'tail end' only of the coupling cap (i.e. the end opposite from the previous stage plate) out of the circuit and 'gator clip an voltmeter on then flip the amp on to idle. It should only be showing a low VDC reading (5V - 6V at most). If it is around 12V or higher I'd ditch it.

[markammon1979]

dropping resistor

Hi, I've done some testing. I've found out that the transformer is producing 50v over what it should be (not the correct 5E3 transformer) so my voltages were incorrect all ober the amp. To get the voltage down I have put a dropping resistor connected between the HT supply and the DC side on the rectifier. I know have correct voltages and a much better sounding amp. The problem is that the dropping resistor is getting pretty hot. It is a 250ohm 20w resistor. Is this ok? Is putting a resistor in an ok thing to do?

[PRNDL]

I recently converted an old audio amp into a 5E3 deluxe. It works fantastically apart from after half an hour or so playing at near top volume the sweet distortion sound goes weird! The amps sound dips, swells and the distortion sound becomes just plain weird.

Did you put in new electrolytics, or use the old ones?

[tubeswell]

Hi, I've done some testing. I've found out that the transformer is producing 50v over what it should be (not the correct 5E3 transformer) so my voltages were incorrect all ober the amp. To get the voltage down I have put a dropping resistor connected between the HT supply and the DC side on the rectifier. I know have correct voltages and a much better sounding amp. The problem is that the dropping resistor is getting pretty hot. It is a 250ohm 20w resistor. Is this ok? Is putting a resistor in an ok thing to do?

A more efficient way of reducing the B+ voltage is putting one or more zener diodes reverse biased from the center tap of the PT high voltage secondary winding to ground. When you use a resistor in the B+, you also reduce the current getting through, which will be making your resistor heat up (as the resistor is having to otherwise dissipate the energy that is being impeded). The back-biasing sets the HT winding at a slightly negative voltage w.r.t. ground, thus lowering the overall + voltage being put out by the secondary w.r.t. ground.

There are various opinions about what power rating to go for on the Zeners - safer to go bigger rather than smaller I would say. (5W minimum)

See:

http://www.aikenamps.com/BackBiasing.html

[markammon1979]

It's had all new electolytics.

[PRNDL]

50 volts over what's expected in B+ is common and won't cause the problems you describe.
6V6's can easily handle 400V as long as they are biased correctly (they sound better IMHO).
50 / 370 is about 12%. Problems in tube amps usually result in huge differences.

I experience the same problem when diagnosing amps ... it's easy to get sidetracked on minor issues.

The first steps are usually:
- measure all voltages (B+ and tube anode/cathode/grid)
- measure all resistors and replace those >10% off
- measure the resistance of electrolytics. They should rise to K or M ohms or the value of resistors in parallel.

[Bruce / Mission Amps]

Hi, I've done some testing. I've found out that the transformer is producing 50v over what it should be (not the correct 5E3 transformer) so my voltages were incorrect all ober the amp. To get the voltage down I have put a dropping resistor connected between the HT supply and the DC side on the rectifier. I know have correct voltages and a much better sounding amp. The problem is that the dropping resistor is getting pretty hot. It is a 250ohm 20w resistor. Is this ok? Is putting a resistor in an ok thing to do?

Yes I agree the lower voltages sound better in these amps and I'd work to get it there too.
I hope I'm just not asleep here, but let's do a little math here because something doesn't look right to me :

You can drop 50v across a 250 ohm resistor?... what kind of current would that take?
50v/250ohms = 200ma.... WHAT?
How can a little 5E3 draw 200ma?
200ma into a 250 ohm resistor is .2a^2 x 250 = 10 watts so even though hot, your 20 watter should be fine.
BUT, the 5E3 amp only makes about 12 watts so 25%-35% of your PTs power is getting used up in a hot resistor.

The standard 5E3 PT from the 50s was only about 100ma-110ma.
Most of these amps will draw around 75ma to 85ma out of the rectifier at idle.
If that was the case, your 250 ohm resistor should only drop the voltage by less then 20vdc. Where is that other 30v drop coming from?

To loose ~50v with zener diodes you'd only need about 3 to 6 18v-12v 5 watt zener diodes in the center tap of the PT high voltage secondary.
At about $1 each I think it is easier using a few 5w axial zener diodes then buying and mounting a big TO3 packaged 50 watt 50v reverse biased zener with a socket and all that rot. But that's your choice and the TO3 looks neater.
The 5 watters will be fine .... 18v drop with 250ma = 4.5 watts... but your 5E3 shouldn't be drawing more then around 135ma even when cranked to the hilt.
Also, when using power resistors to drop B+, don't put one or more after the rectifier, use two of them.. one on each Hi-V AC leg to the rectifier.
I think it is more effective.

[markammon1979]

Still problems...

Hi, thanks for the advice Bruce. Math was never my strong point! But even after all that it is still acting weird at the 20- 30 minute timescale. I decided to make a new board up with new components to transplant into the amp. And what do you know... it still does it! The only components not to have changed are the tube sockets, transformers, input jacks and pots. I have now recorded an audio track to show you guys what it sounds like. The first part of the track is when it is just turned on. The second part is after 30 minutes. the audio is at: http://cid-966258abf718898d.skydrive...em%20sound.mp3

Have a listen and see what you think!