Everyone is assuming that the output transformer is the culprit, this is not necessarily the case, there is an input transformer as well and the two transformers in concert are responsible.
If the feedback winding on the OPT is tapped changing to a different tap may both reduce gain and increase bandwidth.
Jack Elliano (Electra-Print) could probably wind a pair of replacement transformers for this amplifier at something approaching reasonable cost, again you need to know root cause of the roll-off which imho in real world terms may not be as bad as it looks. This approach would at least allow the use of 8 ohm speakers directly.
If the feedback winding on the OPT is tapped changing to a different tap may both reduce gain and increase bandwidth.
Jack Elliano (Electra-Print) could probably wind a pair of replacement transformers for this amplifier at something approaching reasonable cost, again you need to know root cause of the roll-off which imho in real world terms may not be as bad as it looks. This approach would at least allow the use of 8 ohm speakers directly.
Well, as you say, using my little Crown amp as the driver, instead of the driver amp, gives me only -3dB at 20KHz, and I think that is acceptable. So, it seems to me now, that the main "culprit" is the driver amp.
To answer cyclecamper's question, the p-p impedance of these 805's in this configuration is about 6K ohms.
And I should say, that if I wanted to use these for myself, they'd need to work with 4 ohm speakers. I was using the 8 ohm speaker as a suitable target for a potential sale.
To answer cyclecamper's question, the p-p impedance of these 805's in this configuration is about 6K ohms.
And I should say, that if I wanted to use these for myself, they'd need to work with 4 ohm speakers. I was using the 8 ohm speaker as a suitable target for a potential sale.
Hang on... you must remove the pin 5&6 connection entirely to know anything real about the output iron. DO NOT short those pins!! Leave them open.
Keith, these will "work" with 4 ohm speakers, since you have 200+ watts potentially available, you may not get the full output power with less than optimal loading. They will "work" and make a lot of sound.
The RCA sheet calls for 6700 ohms at 1250vdc and 8200 at 1500vdc.
Lower than I recalled.
To run them for "real" I'd think in terms of what it would take to drive the interstage transformer on the 805 chassis. The answer to that is that it will only take <5watts and then to match the impedance of that transformer.
You could do that any number of ways...
The original driver chassis can be sold as complete amps... fwiw.
I think the question that you really need to answer revolves around what you really want here? Do you want a collectable and working amplifier with the old original look? Do you want a modified old amplifier that has been more or less optimized to modern standards? Do you want a high power tube amp that is new and modern?
I'd suggest that if you want to sell it I'd put it up as-is as the person who will purchase it will take the time to research it and/or know what it is already.
The idea of an autoformer for the output is interesting...
_-_-bear
PS. did you have a load on the secondary of the output iron?? necessary... the fluctuation of B+, does that mean it went down and stayed down, or went up and down? Did you look at the B+ with a scope (watch out it may be more than a typical scope with 10x probe is designed for)?
Keith, these will "work" with 4 ohm speakers, since you have 200+ watts potentially available, you may not get the full output power with less than optimal loading. They will "work" and make a lot of sound.
The RCA sheet calls for 6700 ohms at 1250vdc and 8200 at 1500vdc.
Lower than I recalled.
To run them for "real" I'd think in terms of what it would take to drive the interstage transformer on the 805 chassis. The answer to that is that it will only take <5watts and then to match the impedance of that transformer.
You could do that any number of ways...
The original driver chassis can be sold as complete amps... fwiw.
I think the question that you really need to answer revolves around what you really want here? Do you want a collectable and working amplifier with the old original look? Do you want a modified old amplifier that has been more or less optimized to modern standards? Do you want a high power tube amp that is new and modern?
I'd suggest that if you want to sell it I'd put it up as-is as the person who will purchase it will take the time to research it and/or know what it is already.
The idea of an autoformer for the output is interesting...
_-_-bear
PS. did you have a load on the secondary of the output iron?? necessary... the fluctuation of B+, does that mean it went down and stayed down, or went up and down? Did you look at the B+ with a scope (watch out it may be more than a typical scope with 10x probe is designed for)?
Last edited:
Pins left open, wires shorted (green to orange).
Yes, if there's a realistic and economically reasonable path to doing so.
See Edcor EM3600.
Yes, 32 ohm load. Wildly fluctuating between say 200 and 350 volts, up and down. No scope on the HV, wasn't taking the risk to have it run that long. Scope on the output showed the signal also fluctuating in level, didn't bother to take the time to measure it.
I'm thinking now, since the power amp OPT seems to have adequate FR, my choices are: 1) modify the existing driver amp or 2) use or build another driver amp. Building a tube based driver amp may be fun and doable...
One thing I haven't seen yet is the frequency response fom the power amp when the output is wired for 100-Volt (~48ohm load)
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
It will take a driving amplifier of considerably over 5 watts to operate cleanly. The driving amp's load is a pair of forward conducting diodes with a small threshold (for 805's). A good design will throw away a lot of drive to maintain linearity under these conditions. Maybe a good place to start are the various G2 drive sweep tube schemes folks have been trying. 805's will have similar issues.
All good fortune,
Chris
All good fortune,
Chris
From your reports that a higher load impedance gave you a flatter frequency response on the high end.
RCA calls for 6 watts of drive at 1250vdc and 7 watts at 1500 for over 300 watts output.
5 watts is close enough. A pair of 2A3 are likely sufficient or 6V6, etc...
If the drop is NOT in the interstage xfmr (still to be determined) I'd consider driving that with a P-P cathode follower circuit, assuming the voltage swing necessary for the grid drive can be achieved, again RCA calls for a bit over 200volts peak grid to grid.
The wiring diagram is in post #26 for 100 ohms... I'd presume he has the specs/diagram for the Kenyon xfmr or estimated the turns ratio based on your tests, thus the impedance?
_-_-
5 watts is close enough. A pair of 2A3 are likely sufficient or 6V6, etc...
If the drop is NOT in the interstage xfmr (still to be determined) I'd consider driving that with a P-P cathode follower circuit, assuming the voltage swing necessary for the grid drive can be achieved, again RCA calls for a bit over 200volts peak grid to grid.
The wiring diagram is in post #26 for 100 ohms... I'd presume he has the specs/diagram for the Kenyon xfmr or estimated the turns ratio based on your tests, thus the impedance?
_-_-
When wired for 70 volts the secondary windings are in parallel. When wired for 100 volts the secondary windings are in series. That makes for a big difference in inductance.Yes, but I needed about 200 ohms before it was significantly better. I did try 50 ohms, which was a little improvement, but I didn't take notes. Can you say how you know that is the proper way to wire this OPT for 100 volts? It's a good thing to knowThanks.
I don't "know" that is the proper way to wire this. You already know how to test transformers so test the changes before applying power.
Don't forget to move the feedback connection also.
HK
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.