DIY Audio Projects Forum

Hi, Everyone, every now and then I run into a new situation. This is one of them. A diyer built a clone Champ 5C1. Standard except for using an EL84 instead of a 6V6. It should be fine. It starts out fine then after a modest time, arcs from the anode of the EL84 to the chassis. The chassis is like many older designs used as signal ground. It will actually carbon track between them. All voltages check out and the output transformer was the old impedance (17K ...gads) and replaced with the newer 4.5K version. The socket has been replaced as well. I figure that on overdrive the resulting wave is somehow getting really large. But why? How to stop it? Any thoughts.

Good listening
Bruce

_________________
Some of my DIY Tube Amplifier Projects:

• PoddWatt: 5751 / EL84 Push-Pull DIY Tube Amplifier Project
• OddBlocks: KT88 Push-Pull DIY Tube Amplifiers (Mono Blocks)
• ForeWatt: ECC802S (12AU7 / ECC82) DIY Tube Preamplifier Project
• GrooveWatt: Tube (Valve) RIAA Phono Preamplifier Project

Have you tried a ceramic socket? This shouldn't track to earth.

Hi,

How far is the arc jumping? can you show a picture of the tube socket.

If its clearing more than a couple of mm then it must be a very high voltage, think spark plug gap in a car.
So this can only be back E.M.F from the output Tx, can't see what else would generate it.

I have seen transient suppressors used something like 415V.
I have also seen reverse diodes across output tubes<<but if they fail you can trash the output Tx.

But it makes me wonder what would "chop" the DC in the output Tx primary to generate such a high voltage.
I wonder if the tube is shutting down with cathode cap charge up, might be worth trying a zener across the cathode cap just above the cathode voltage to see if it stops it. Or try removing the cathode cap to see if it still arcs.

Anyway if these kind of voltages are appearing across the tube, I'm surprised the tube hasn't flashed over inside. Unless it is and that's what is causing the HV from output Tx to Gnd.

Regards
M. Gregg

_________________
What is the sound of one hand clapping?

Something else to try,

Try putting a 1K grid stopper on pin 9 and pin 2.
And a 1meg from input grid to ground just for interest.<<it is interesting to see the effect of the vibration breaking the volume wiper from the track causing a high speed oscillation on the input grid. Or a 1 meg across from the volume wiper to ground should also work.

I'm guessing these voltages may exceed the output Tx insulation values so it might not last long.

Regards
M. Gregg

_________________
What is the sound of one hand clapping?

Hi, So far no fix. I keep wondering if there is some wiring error or perhaps abuse of the amp going on. Possibly severe overdriving?

Good listening
Bruce

_________________
Some of my DIY Tube Amplifier Projects:

• PoddWatt: 5751 / EL84 Push-Pull DIY Tube Amplifier Project
• OddBlocks: KT88 Push-Pull DIY Tube Amplifiers (Mono Blocks)
• ForeWatt: ECC802S (12AU7 / ECC82) DIY Tube Preamplifier Project
• GrooveWatt: Tube (Valve) RIAA Phono Preamplifier Project

Hi,

The only thing that can cause this HV on the anode is from the OP Tx.
Either the quiescent current is being chopped or there is another input into the speaker terminals.

Is the arching happening when you are testing it or just when the "owner" is using it out of sight?
I find it hard to believe the input to the Tx from the output tube is chopping the transformer current.

The only other time I have seen anything like this is with a broken primary winding on the op Tx.
Or in extreme cases a failed power transformer with shorted windings

Regards
M. Gregg

_________________
What is the sound of one hand clapping?

Just for interest

what happens if you put a 1 K power resistor across the speaker terminals? with the speaker attached.
So the 1K is in parallel with the speaker coil.

Its very interesting in this situation, because you have a tube rectifier so you have twice the voltage across the series power B+ winding which is the only other source of high voltage but still not enough to jump a couple of mm unless its being inverted to give higher voltage otherwise I would look at the rectifier.
The reason is because you say it happens on overdrive so its possible for vibration of the cabinet to cause intermittent fault of power tube etc.
But I assume you have just changed and checked the tubes as a matter of course.

Is it an old amp been revamped or a new build?
Its interesting shame you can't catch a short video of it happening.
If all else fails try substituting the op Tx primary with a resistor and thrash the amp to see if the happens then that would eliminated the OP Tx.
But as a guess its not going to vibrate the cab...strange.

Regards
M. Gregg

_________________
What is the sound of one hand clapping?

Hi, Thanks for the suggestions. It is a new build diy amp. By a novice. The output transformer has been replaced with a higher quality one. Both do it. It seems only in overdrive. The best guess I have is that it is being overdriven so much that the tube is going into somthing like cut off and then back on. That type of action seems to result in spikes as the energy in the core is released. So my latest thought and suggestion to the diyer is to put a capacitor from the anode to ground. Roughly 0.01uf or so ceramic. It might just dull the spikes enough to stop the arcing. I suspect that the builder is trying to extract more output from the amp than it is capable of. Not IMO an unusual thing for some musicians.

Good listening
Bruce

_________________
Some of my DIY Tube Amplifier Projects:

• PoddWatt: 5751 / EL84 Push-Pull DIY Tube Amplifier Project
• OddBlocks: KT88 Push-Pull DIY Tube Amplifiers (Mono Blocks)
• ForeWatt: ECC802S (12AU7 / ECC82) DIY Tube Preamplifier Project
• GrooveWatt: Tube (Valve) RIAA Phono Preamplifier Project

Some thoughts...

The EL84 is not a beam power tube (BPTs) but a true pentode. BPTs are very tolerant of a high Vg2 voltage, true pentodes are not. BPTs are designed for the condition Vg2=B+, pentodes are generally designed for the condition Vg2 ≈ 1/3 to 2/3 B+. The Champ 5C1 design called for a 6V6 which is a BPT. The 5C1 voltages were also above max for the 6V6 and whereas the official 6V6GT max voltages are only 15V higher than the EL84 (315v vs 300v), the BPT is much more likely to handle the over voltage condition.

It is also likely that the bias point of the EL84 is significantly lower current than the 6V6 in this circuit. This leads to strong cutoff in overdrive conditions and the residual magnetic flux in the core pushes the primary voltage VERY high when the current is cutoff. I would think that this is a design problem not an equipment failure.

Note the bias differences between the 6V6 with Vg2=285v and the EL84 with Vg2=300v (≈5% delta).
The significant drive of the 6SJ7 is likely driving the EL84 into cutoff rather quickly. In my opinion, the choice to simply insert an EL84 into the 5C1 design without an evaluation of the B+ voltages and the operating point was ill advised. I suspect that if one inserted a 6V6 (or even a 6AQ5) in this amp, it would operate satisfactorily.

You could shunt the primary side of the transformer to equalize the load impedance across frequency with a 5kΩ to 10kΩ resistor in series with a 0.1µf capacitor (as shown in the following paper by Tulauskas) but I would think that a redesign for a more realistic bias point would be in order.
In summary, I believe that the issues with this amplifier stem not from a component or tube failure, but from a fundamental design flaw introduced when the EL84 was substituted for the 6V6 call for in the original design. Without a reassessment and redesign for the power tube substitution, I find it unlikely that the amplifier will function as desired.

Just my thioughts on the matter. YMMV.

_________________
Matt
It's all about the Glass!
http://www.CascadeTubes.com
Cascade Tubes Blog

Hi Matt,

Thanks for the info in the PDF it reminds me of the Zobel Ideas.
The old valve record players used the same tone control ideas in the 1950's.

Some really interesting info in the PDF.
Its interesting to note in years gone by how they matched the output Tx to the speaker load but in this day and age we tend to plug all sorts of speakers into the amps without much thought.

Regards
M. Gregg

_________________
What is the sound of one hand clapping?