A Non-Floating Signal Tracer is a problem, unless . . .
The circuit / product you are troubleshooting has a grounded chassis, and a grounded circuit.
You will get lots of hum if the circuit / product is not grounded, or worse yet, you will have a shocking result.
Safety first!
Prevent the "Surviving Spouse Syndrome"
A Floating Signal Tracer, is battery powered (Tube or Solid State).
In terms of using a pair of back to back transformers for B+, that may be "floating" to DC, but there will be capacitive leakage currents that may cause you to hear hum no matter where you probe the circuit / product that you are troubleshooting.
Ground Loops abound.
Just my opinions
The circuit / product you are troubleshooting has a grounded chassis, and a grounded circuit.
You will get lots of hum if the circuit / product is not grounded, or worse yet, you will have a shocking result.
Safety first!
Prevent the "Surviving Spouse Syndrome"
A Floating Signal Tracer, is battery powered (Tube or Solid State).
In terms of using a pair of back to back transformers for B+, that may be "floating" to DC, but there will be capacitive leakage currents that may cause you to hear hum no matter where you probe the circuit / product that you are troubleshooting.
Ground Loops abound.
Just my opinions
Yeah, that was a typo covered a couple posts back in the thread. The tube is a 6AW8A, triode/pentode combo. I got the basic circuit together and ran DC tests, which looked good. Current levels well within tube specs, etc. In AC tests, the triode seemed to amplify with no issue but the pentode section has issues. Troubleshooting will have to wait until I get more time....
dtbradio,
For a pentode (when connected as a pentode), with transconductance Gm, and whith a plate load impedance RL . . .
. . . The gain is Gm x RL
Pentode section:
6AW8 Gm 9,000 uMhos
6AW8A Gm 9,500 uMhos (9.5 mA change / volt of signal on g1)
Example:
Using a 6AW8A, and an output transformer of 5k Ohms to 8 Ohms with an 8 Ohm load:
0.0095 Amp x 5000 Ohm = 47.5 the gain from g1 to the plate.
A 5000 Ohm to 8 Ohm transformer has a voltage loss of 25.
47.5 / 25 = 1.9 gain (from pentode g1 to the 8 Ohm tap).
Use those same calculations, and just change the output transformer numbers to what you have, and see what your gain should be for your circuit.
For a Triode wired pentode, the gain calculations are different. Are you triode wiring the 6AW8A?
For a pentode (when connected as a pentode), with transconductance Gm, and whith a plate load impedance RL . . .
. . . The gain is Gm x RL
Pentode section:
6AW8 Gm 9,000 uMhos
6AW8A Gm 9,500 uMhos (9.5 mA change / volt of signal on g1)
Example:
Using a 6AW8A, and an output transformer of 5k Ohms to 8 Ohms with an 8 Ohm load:
0.0095 Amp x 5000 Ohm = 47.5 the gain from g1 to the plate.
A 5000 Ohm to 8 Ohm transformer has a voltage loss of 25.
47.5 / 25 = 1.9 gain (from pentode g1 to the 8 Ohm tap).
Use those same calculations, and just change the output transformer numbers to what you have, and see what your gain should be for your circuit.
For a Triode wired pentode, the gain calculations are different. Are you triode wiring the 6AW8A?
Here's the rough schematic, pin numbers omitted, tube = 6AW8A. Will be trying 6BA8A as well, same pinout and close-enough specs. I haven't added bypass caps to the cathode resistors yet, but intend to once the rest of the circuit checks out.
No bypass cap across the output tube 1k cathode resistor = very low gain.
Also, with a 10k screen resistor, that may cause low gain, and distortion too.
0.1 uF coupling cap, and 820k g1 resistor has a - 3dB low frequency response of 2Hz (-1 dBk at 4Hz).
That is a much lower frequency than the output transformer, and the speaker can reproduce.
820k does not substantially lower the gain of: a 47k triode load resistor that is in parallel with the triode rp.
The question is, does the pentode section of the 6AW8A have a specified maximum g1 resistor value?
Do not exceed the value of that specification for the pentode g1.
Also, with a 10k screen resistor, that may cause low gain, and distortion too.
0.1 uF coupling cap, and 820k g1 resistor has a - 3dB low frequency response of 2Hz (-1 dBk at 4Hz).
That is a much lower frequency than the output transformer, and the speaker can reproduce.
820k does not substantially lower the gain of: a 47k triode load resistor that is in parallel with the triode rp.
The question is, does the pentode section of the 6AW8A have a specified maximum g1 resistor value?
Do not exceed the value of that specification for the pentode g1.
Last edited:
For cathode-bias ops, both sections specify 1M as max grid resistor value. I'm well under in both cases. As for the 10K on G2, I chose that because I've seen that used on a number of schematics for similar pentode circuits, and I have no other frame of reference for what to use on G2. The spec sheet talks about not exceeding 3.5mA on G2, so I guess in my case 68K would make more sense unless I'm not understanding something.
Do you recommend blocking resistors in this simple circuit?
PS, I changed the .1uf to .01uf. Smaller size cap, made more sense in this circuit. May go to .0047, but don't want to cut lows TOO much.
Do you recommend blocking resistors in this simple circuit?
PS, I changed the .1uf to .01uf. Smaller size cap, made more sense in this circuit. May go to .0047, but don't want to cut lows TOO much.
Last edited:
Tested it again with no audio input, and I have a nice 4-volt nearly-perfect sawtooth at the plate of the pentode. This is with the 68K at G2 now. Will come at it again tomorrow, closing up the shop for the night.
Is it bad manners to quote myself, lol? Figured out that the sawtooth was amplifier power supply ripple. Added an 8K resistor and 47uf cap, and the ripple from the pentode plate dropped to around 50mV. More to come as I get time.
Project sidelined for the moment. My improvised HV transformer melted down for some reason. Was watching cathode voltages for any signs of problems/over-current when I smelled "it". Everyone knows what "it" is. The unmistakable yet indescribable smell of over-heated transformer varnish, copper, and assorted other transformer constituents. Kind of like the smell coming off of the angry constituents of some politicians....
The reactance of such capacitor must be less than the parallel resistance of the internal screen dynamic resistance (may be found in the tube datasheet or approximated to the screen voltage divided screen current [this DC resistance is several times smaller than the dynamic screen resistance but at less as firsts times is does the job]) and the dropping resistance at the smaller frequency that the amp must be capable of manage. But don't exceed too much the cap value, because an excessive value will delay the tube to operate and the anode transformer may saturate (and arcing when desaturates) at very slow frequencies or when plugging external sources to the amplifier being it powered.
You can also use a regular filament transformer and using it in reverse from the regular heater supply, to make a nice 120VAC bias supply.
When hooking up two filament transformers back-to-back, be sure they do not get too hot. Sometimes the cheapest-made transformers will overheat this way, even with no or low load because they were made with as little wire turns as possible.
You don’t need to back to back Anteks to get HT and filament, because their 100 and 200 VA 230v units are priced low enough that you don’t need to resort to this. If you can get suitable transformers for only $42 or $52 you just use them. Not much savings over two 20 or $26 units (At half the VA). When you find surplus Stancors for $6.95 THATS when you stick a pair back to back.
A DC DC step up might be just what is needed :
https://www.ebay.com/itm/2656699942...RXhW1UpMYqFweQe8JkRKk7bg==|tkp:Bk9SR9rCm6HGYQ
https://www.ebay.com/itm/2656699942...RXhW1UpMYqFweQe8JkRKk7bg==|tkp:Bk9SR9rCm6HGYQ