That's interesting but what does a plot of THD vs Frequency have to do with frequency response? I want to know what the output power vs frequency looks like.
Please cover 8 9 and 10 with some plastic caps it is nerve wracking .. I would hate to lose one of my favorite youtube channels .. other than that you have always had excellent videos on tube circuitry .. Thanks for all the hard work Dave
I really do appreciate that... :-) It did finally occur to me that I really should have placed, at least, a piece of tape (preferably red...) over the HV terminals, 8,9,and 10. Usually, when I am working along, I am very safe in my work practices but making videos is a huge distraction - not a good excuse - and as can be seen in this video, mistakes are much more likely to happen. 800 volts could certainly be a bad experience and possibly deadly but when I deal with multi KV level, I can honestly say, I am much more focused and aware and careful as I know it can, and would likely be, Death On Contact. I truly appreciate your comment and concern.
I try never to short the secondaries so there are no sparks by making sure I have unplugged the primary of the transformer and also not get electrocuted by the HV windings.
Been watching your series on transformers. I'll probably end up getting one of those LCD, ESR testers. Seems pretty handy. Question. How important is using the correct speaker load for an output transformer? How much can it be off and how will it affect the OPT? I notice that you were testing 4, 8, and 16 ohm taps with an 8 ohm load. Thanks for your info!
I can't say for sure what the tolerable extreme is in impedance mismatch but one incident I can recall is years ago a gentleman brought his amplifier to me and the EL34's were bad. I ordered new ones and about a week later he brought it back so I sent the tubes back to TubeDepot and they send me new ones. Another week goes by and same thing. Well, I sent those EL34's back and they sent me another pair (Thanks TubeDepot) but this time I told him to bring his speaker cabinet to me. He had four 12" speakers at 16 ohms all wired in series for 64 ohms. I rewired the four for 16 ohms and as far as I know those EL34's are still running. The moral of that story is, there is definitely an extreme that we cannot go to in impedance mismatch and hope to keep output tubes working or possibly shorting the output transformer.
I have a pair of those MkIII... beat up pretty bad by previous owners... Been planning on building larger new chassis, more power supply filtering, and a cascode differential front end.. also offering balanced input.
The output transformer is very good up to 50 watts. I don't recommend the 6AN8 driver as I have around 20 of those tubes and only a couple perform really good and no one makes them anymore. There is a drop-in replacement for the driver board sold by Triode Electronics called the Poseidon board that I have used in several amplifiers that performs nicely. If you have two and the output transformers are good and can be painted nicely you can have a top-notch 100 watt stereo amp.
When I get a mystery transformer I like to use a scope to look at all the secondary windings, and DOT the terminals for AC polarity relative to the input. Later when I use the transformer, I have not only the volts on windings, but also the polarity of each.
Could someone help me with my Western Electric 91A rep. Amps. I have 2 pair, one of the pair goes up to 25Khz and the other pair nearly 12Khz as a maximum. How is that possible? At first I thought it must be the output transformer but tested it with another transformer of the same type and still the amp is not passing the 12Khz. Please do you have some suggestions, what could this be?
I think the different frequency waveforms would add and subtract both in frequency and amplitude so you would end up with sum and difference plus the two fundamental frequencies. It could end up pretty strange...
I can see that could come in handy if the THD was really high but in our case, it was always below 3% so it really wasn't that bad. The single idea of the measurement was to determine if frequency response was affected by this mismatch. It was to some extent but not dramatically. On the other hand, musicians might like the subtle difference in sound that is created by this mismatch.
The logarithmic scale is useful for verifying low distortions, since it is possible to easily view from 0.01% to 10% in the graph. Note that it is always used in specialized publications. It was just a suggestion of better visualization. Thanks for the answer.
You mention that 8 ohms is the proper load for the amplifier and output transformer. Shouldn't maximum power transfer occur when the impedences are properly matched? Why does power output increase for a 16 ohm load?
Because it is lowering the impedance the output tubes see this drawing more current through those tubes this why the power and distortion increases the impedance load is doubled on the 4 ohm tap upping the impedance the tubes see and lowering current through the tube and also lowering distortion and power but this is only the case for this amp and these output tubes and sometimes increasing Z may actually increase distortion but and interesting effect reguardless.
I'll bet that when you put the 8 ohm load on the 16v tap, the amplifier went into severe clipping which explains the really bad THD. All you are really doing is changing the load on the plate circuit. More current or less. It explains why the 4 ohm tap had a better THD. Good example on the transformer winding phasing. Series, in phase adds voltage. Out of phase subtracts voltage. Often called "buck-boost". As long as the winding insulation is rated for the total voltage to ground, you can use low voltage windings even from different transformers to buck or boost a B+ voltage. You just have to be careful not to use two different transformers that have a difference in phase shift across the primary to secondary. With single phase transformers the phase shift primary to secondary is usually 0 3-phase power transformers with delta connected primary and wye connected secondary have a 30 degree phase shift. If all else fails, get out the oscilloscope. But there's lots of tricks with windings. I think in the with of days engineers were more inventive in using what they had.
Actually he does care, his last few videos the camera did't move at all... boring. I asked him if he would "bounce" the camera so there was some action.. Thanks again Dave
I think it's not a big deal unless you are viewing on a 100" screen, the up side is that we get to easily view all the equipment settings and angles. It would take a lot of effort and time to get all the same shots from a tripod.
"If there are sparks, you are doing something wrong!" I LOVE IT Dave!
David, Thanks so much for your videos. I love them the way they are. Don't worry about A-holes who demand it all without contributing one Iota.
That's interesting but what does a plot of THD vs Frequency have to do with frequency response? I want to know what the output power vs frequency looks like.
Thanks, guess that answers the question. The 8 ohm load on the 4 ohm tap wasn't too different, but on the 16 ohm tap it wasn't too good. Cheers, DA.
back in the 70s I had a fresh pair of MKIII running Speakerlab's copy of PWK's Klipschorn - slightly modded PAS
I am definitely a fan of Klipsch and listen to a pair of Klipschorns daily that I purchased new in 1996.
Please cover 8 9 and 10 with some plastic caps it is nerve wracking .. I would hate to lose one of my favorite youtube channels
.. other than that you have always had excellent videos on tube circuitry .. Thanks for all the hard work Dave
I really do appreciate that... :-) It did finally occur to me that I really should have placed, at least, a piece of tape (preferably red...) over the HV terminals, 8,9,and 10. Usually, when I am working along, I am very safe in my work practices but making videos is a huge distraction - not a good excuse - and as can be seen in this video, mistakes are much more likely to happen. 800 volts could certainly be a bad experience and possibly deadly but when I deal with multi KV level, I can honestly say, I am much more focused and aware and careful as I know it can, and would likely be, Death On Contact.
I truly appreciate your comment and concern.
Can you tell me why there are no sparks ?
I try never to short the secondaries so there are no sparks by making sure I have unplugged the primary of the transformer and also not get electrocuted by the HV windings.
Been watching your series on transformers. I'll probably end up getting one of those LCD, ESR testers. Seems pretty handy. Question. How important is using the correct speaker load for an output transformer? How much can it be off and how will it affect the OPT? I notice that you were testing 4, 8, and 16 ohm taps with an 8 ohm load. Thanks for your info!
I can't say for sure what the tolerable extreme is in impedance mismatch but one incident I can recall is years ago a gentleman brought his amplifier to me and the EL34's were bad. I ordered new ones and about a week later he brought it back so I sent the tubes back to TubeDepot and they send me new ones. Another week goes by and same thing. Well, I sent those EL34's back and they sent me another pair (Thanks TubeDepot) but this time I told him to bring his speaker cabinet to me. He had four 12" speakers at 16 ohms all wired in series for 64 ohms. I rewired the four for 16 ohms and as far as I know those EL34's are still running. The moral of that story is, there is definitely an extreme that we cannot go to in impedance mismatch and hope to keep output tubes working or possibly shorting the output transformer.
I have a pair of those MkIII... beat up pretty bad by previous owners... Been planning on building larger new chassis, more power supply filtering, and a cascode differential front end.. also offering balanced input.
The output transformer is very good up to 50 watts. I don't recommend the 6AN8 driver as I have around 20 of those tubes and only a couple perform really good and no one makes them anymore. There is a drop-in replacement for the driver board sold by Triode Electronics called the Poseidon board that I have used in several amplifiers that performs nicely. If you have two and the output transformers are good and can be painted nicely you can have a top-notch 100 watt stereo amp.
Have you done the same type of testing with a 4ohm load on all (4,8, and 16 ohm) taps?
When I get a mystery transformer I like to use a scope to look at all the secondary windings, and DOT the terminals for AC polarity relative to the input. Later when I use the transformer, I have not only the volts on windings, but also the polarity of each.
thx, can you talk more about the tools in this vid? i am a newb and some of that stuff is new to me. like the device u have going on at 9 mins
have you considered writing a book about your electronics projects?
Could someone help me with my Western Electric 91A rep. Amps. I have 2 pair, one of the pair goes up to 25Khz and the other pair nearly 12Khz as a maximum. How is that possible? At first I thought it must be the output transformer but tested it with another transformer of the same type and still the amp is not passing the 12Khz. Please do you have some suggestions, what could this be?
What will happen if you have two transformers with different frequency connected
I think the different frequency waveforms would add and subtract both in frequency and amplitude so you would end up with sum and difference plus the two fundamental frequencies. It could end up pretty strange...
I saw in some channel that it is possible to put the vertical scale in logarithmic by clicking the legend THD+N in the software Pete Millett.
I can see that could come in handy if the THD was really high but in our case, it was always below 3% so it really wasn't that bad. The single idea of the measurement was to determine if frequency response was affected by this mismatch. It was to some extent but not dramatically. On the other hand, musicians might like the subtle difference in sound that is created by this mismatch.
The logarithmic scale is useful for verifying low distortions, since it is possible to easily view from 0.01% to 10% in the graph. Note that it is always used in specialized publications. It was just a suggestion of better visualization. Thanks for the answer.
You mention that 8 ohms is the proper load for the amplifier and output transformer. Shouldn't maximum power transfer occur when the impedences are properly matched? Why does power output increase for a 16 ohm load?
I am confused about that too, slightly higher ohms should not matter so much on a tube amp.
But his graph went completely wild.
Because it is lowering the impedance the output tubes see this drawing more current through those tubes this why the power and distortion increases the impedance load is doubled on the 4 ohm tap upping the impedance the tubes see and lowering current through the tube and also lowering distortion and power but this is only the case for this amp and these output tubes and sometimes increasing Z may actually increase distortion but and interesting effect reguardless.
"LCR"
I'll bet that when you put the 8 ohm load on the 16v tap, the amplifier went into severe clipping which explains the really bad THD. All you are really doing is changing the load on the plate circuit. More current or less. It explains why the 4 ohm tap had a better THD.
Good example on the transformer winding phasing. Series, in phase adds voltage. Out of phase subtracts voltage. Often called "buck-boost". As long as the winding insulation is rated for the total voltage to ground, you can use low voltage windings even from different transformers to buck or boost a B+ voltage. You just have to be careful not to use two different transformers that have a difference in phase shift across the primary to secondary. With single phase transformers the phase shift primary to secondary is usually 0 3-phase power transformers with delta connected primary and wye connected secondary have a 30 degree phase shift.
If all else fails, get out the oscilloscope.
But there's lots of tricks with windings. I think in the with of days engineers were more inventive in using what they had.
Camera bounce too much. Guess you don't care how it looks to us.
Actually he does care, his last few videos the camera did't move at all... boring. I asked him if he would "bounce" the camera so there was some action.. Thanks again Dave
I think it's not a big deal unless you are viewing on a 100" screen, the up side is that we get to easily view all the equipment settings and angles. It would take a lot of effort and time to get all the same shots from a tripod.
@@Eric-uf8zx That's two for camera bounce! How about actually planning a video.