I suppose then...
If I were to use an OPT with a secondary designed for a higher load impedance - using 300R on a 600R tap, to get a lower reflected impedance to the plate...
I'd expect that the resonances I saw in an OPT like the SE one I have mentioned, would be increased in amplitide and Q; if there is a resonance that can be excited by the valve.
If I were to use an OPT with a secondary designed for a higher load impedance - using 300R on a 600R tap, to get a lower reflected impedance to the plate...
I'd expect that the resonances I saw in an OPT like the SE one I have mentioned, would be increased in amplitide and Q; if there is a resonance that can be excited by the valve.
That's ok, when contributions are essentially merde, then the author isn't really trying to help. And none want that kind of "help".
Let me qualify myself.
I make mistakes and speak in error too often for my liking, but I am humble enough to realise this.
Dropping BS comments, well that's kinda flys against the spirit of the forum, dont you think?
regardless of anything else, quick reference to the OPT mentioned would have cleared up the tap impedance..
Last edited:
So, I have set about repeating tests with different series resistance, placed at the signal generator output.
I have repeated with Rs = 10k, 56k and 100k.
I have only plotted 10k and 56k but the trend shows in the raw numbers for the 100k configuration.
Broadly speaking, the half impedance point at high frequencies is at around 60kHz with Rs = 10k.
With Rs = 56k, this half impedance point drops in frequency to about 47kHz.
I expect that to be lower in frequency again, with Rs = 100k.
Given the size of the change, with Rs = 56k, perhaps then, the loss of HF in this case isnt too bad at all.
Assuming Rs = Rsource seen by the OPT then the response might be very good indeed; I consider 56k to be high impedance...others may not.
Any thoughts on my assumptions? Given these cost £5 and I need a slightly lower turns ratio, then I may give the 230: 18V variant a try.
I have repeated with Rs = 10k, 56k and 100k.
I have only plotted 10k and 56k but the trend shows in the raw numbers for the 100k configuration.
Broadly speaking, the half impedance point at high frequencies is at around 60kHz with Rs = 10k.
With Rs = 56k, this half impedance point drops in frequency to about 47kHz.
I expect that to be lower in frequency again, with Rs = 100k.
Given the size of the change, with Rs = 56k, perhaps then, the loss of HF in this case isnt too bad at all.
Assuming Rs = Rsource seen by the OPT then the response might be very good indeed; I consider 56k to be high impedance...others may not.
Any thoughts on my assumptions? Given these cost £5 and I need a slightly lower turns ratio, then I may give the 230: 18V variant a try.
Good question...
I suppose that would be the anode ra in parallel with the plate to grid feedback resistor and CCS apparent resistance...
The anode is CCS loaded, and dynamic impedance is about 150k to 180k, but other than that, I'll just have to set up the output stage without OPT and substitute in a couple of suitable resistance loads, say 15k and 30k, and calculate the output impedance from the result?
Got me thinking that...
I suppose that would be the anode ra in parallel with the plate to grid feedback resistor and CCS apparent resistance...
The anode is CCS loaded, and dynamic impedance is about 150k to 180k, but other than that, I'll just have to set up the output stage without OPT and substitute in a couple of suitable resistance loads, say 15k and 30k, and calculate the output impedance from the result?
Got me thinking that...
I did a quick and dirty calculation early in the build for output impedance with the LNFB in place, by testing with different loads on the OPT, and I think the result was about 80 Ohms.
While it isn't low by any means, for driving 300 Ohm headphones, that measurement, if it was accurate, and my memory serves me correctly, is quite good, DF of about 4?
If I scale that perhaps accurate, perhaps garbage, number by the impedance ratio...
80 x 49 ...about 4k output impedance at primary.
While it isn't low by any means, for driving 300 Ohm headphones, that measurement, if it was accurate, and my memory serves me correctly, is quite good, DF of about 4?
If I scale that perhaps accurate, perhaps garbage, number by the impedance ratio...
80 x 49 ...about 4k output impedance at primary.
Measured output impedance
I have taken 2 or 3 measures of Volts out versus load resistance.
Vout was measured with 3 different load resistances, 250, 300 and 350 Ohms, resulting in V = 2.77, 2.95 and 3.15 respectively.
250R to 300R, gives Zout of 145 Ohms
250R to 350R, gives Z out of 183 Ohms
Using the latter figure, and multiplying by N^2
183*49 = 9kOhms output impedance of the output stage/source impedance feeding the OPT.
Seeing this, has bolstered my confidence that the test value for source resistance of 10k, was close to reality - and the response in the impedance test should also be close to reality.
Of course the downside is that DF is slightly less than 2.
I have taken 2 or 3 measures of Volts out versus load resistance.
Vout was measured with 3 different load resistances, 250, 300 and 350 Ohms, resulting in V = 2.77, 2.95 and 3.15 respectively.
250R to 300R, gives Zout of 145 Ohms
250R to 350R, gives Z out of 183 Ohms
Using the latter figure, and multiplying by N^2
183*49 = 9kOhms output impedance of the output stage/source impedance feeding the OPT.
Seeing this, has bolstered my confidence that the test value for source resistance of 10k, was close to reality - and the response in the impedance test should also be close to reality.
Of course the downside is that DF is slightly less than 2.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.