No, by "DCR" I meant the DC resistance across "N2" and "N3". The more I think about it, I doubt I2R dissipation will be a problem, but it would be nice to know the DCR.
Be careful when measuring the DCR of a high-inductance device. The intermittent contact with the probes can create an inductive kick that, while not dangerous to you because of the low energy involved, can zap a digital meter. Better to use an analog meter ("VOM") if you still own one. I also keep $3.99 DVMs from Harbor Freight around the house for just such careless duty. Of course, none of them has failed, Murphy being who he is.
You might look at the "Beast" design that Morgan Jones published in the second edition of his "Valve Amplifiers" book. It used a Sowter push-pull choke. I can barely read the part number because it's so small, but I think it reads "8422". Perhaps someone with a psychic connection to MJ (SY, are you reading this?) can answer.
Of course, regular OPTs have been used as ESL plate chokes for years. You will have to terminate the secondary with an RC to damp the resonance, but it's doable.
Be careful when measuring the DCR of a high-inductance device. The intermittent contact with the probes can create an inductive kick that, while not dangerous to you because of the low energy involved, can zap a digital meter. Better to use an analog meter ("VOM") if you still own one. I also keep $3.99 DVMs from Harbor Freight around the house for just such careless duty. Of course, none of them has failed, Murphy being who he is.
You might look at the "Beast" design that Morgan Jones published in the second edition of his "Valve Amplifiers" book. It used a Sowter push-pull choke. I can barely read the part number because it's so small, but I think it reads "8422". Perhaps someone with a psychic connection to MJ (SY, are you reading this?) can answer.
Of course, regular OPTs have been used as ESL plate chokes for years. You will have to terminate the secondary with an RC to damp the resonance, but it's doable.
Can't help with the part number (I don't have a 2nd Edition), but I have used an ESL transformer connected backward as a p-p choke, with a bit of damping resistance across the (former) primary and the stators directly coupled to the tube plates. At the relatively low voltages I tried (1kV), it worked fine.
Checked the DCR and it was lower than i thought it would be, 150 ohm "plate-to-plate", had to check my measurement against an 1% 150 resistor.
About capacitance I measured the panel that has 2*1mm from stator to stator and is 21cm*60cm.
Think this is around 600p and this corresponds to around 12kohm at 20kHz. This means 6kohm for each side of the bridge. Or is it 1200p and 3k/side?
As it isn´t necessary to deliver full swing to 20kHz could it be enough to deliver full swing up to 10kHz?
About capacitance I measured the panel that has 2*1mm from stator to stator and is 21cm*60cm.
Think this is around 600p and this corresponds to around 12kohm at 20kHz. This means 6kohm for each side of the bridge. Or is it 1200p and 3k/side?
As it isn´t necessary to deliver full swing to 20kHz could it be enough to deliver full swing up to 10kHz?
Yes, the DCR is lower than I thought it would too. It won't be an issue then.
If you measured 600pF from stator to stator, then think of it as two 1200pF caps in series. Since it's being driven by a push-pull source, the imaginary center point won't see any signal, resulting in a virtual AC ground. Therefore, each EL34 will be driving the equivalent of a 1200pF cap. At 20kHz, the reactance of a 1200pF cap is 6.6 Kohm. And yes, you won't need to swing full "power" into the load at 20 KHz.
I still think that you'll need some way to assure that the currents are balanced through the former secondary, now primary. Each cathode of the EL34 could have its own precision CCS, with a signal cap linking the cathodes. Or you add a small precision resistor in each cathode, and measure the voltage imbalance between cathodes to drive a servo for correction.
PS: I just checked your website. Nice looking guitar amps!
If you measured 600pF from stator to stator, then think of it as two 1200pF caps in series. Since it's being driven by a push-pull source, the imaginary center point won't see any signal, resulting in a virtual AC ground. Therefore, each EL34 will be driving the equivalent of a 1200pF cap. At 20kHz, the reactance of a 1200pF cap is 6.6 Kohm. And yes, you won't need to swing full "power" into the load at 20 KHz.
I still think that you'll need some way to assure that the currents are balanced through the former secondary, now primary. Each cathode of the EL34 could have its own precision CCS, with a signal cap linking the cathodes. Or you add a small precision resistor in each cathode, and measure the voltage imbalance between cathodes to drive a servo for correction.
PS: I just checked your website. Nice looking guitar amps!
Hi again,
Something like this then:
www.revintage.se/Script300B.pdf
Figured you would like my amps as they are built in the American tradition with American hardware. But with a Nordic twist.......
Thanks for the expanation of the capacitances.
Something like this then:
www.revintage.se/Script300B.pdf
Figured you would like my amps as they are built in the American tradition with American hardware. But with a Nordic twist.......
Thanks for the expanation of the capacitances.
Yes, but NB that I'm only using it as a CT choke. The 1kV rail (limitation was the 811As that I used) meant that SPL was very limited, but this was just a proof-of-concept. Subjectively, and in totally uncontrolled circumstances, I felt that the sound was clearer and a bit more "direct" compared with the original Acoustat interface I was using.
Strikes me I could take my friend Erik Anderssons Audion 15W 300B PP right out of the box and connect the stators to the plates and the HV ground to B+.
The only modification needed would be to terminate the transformer secondary.
Why didn´t I think of that,
?
Any suggestions of where to begin with the RC-termination?
An El Cheapo version could be built around four 6BL7 that has just a tad more Miller cap than one 300B. Also checked the curves to see that they could take 6k load. With two 12DW7 or one ECC83 and any CF-tube, the higher gain would make it work with just one driver gain-stage:
www.revintage.se/Script6BL7.pdf
One could use either the Script stepup, any descent PP transformer or maybe a 115:115V (or 230:230V) mains toroid with the windings in series.
Have two monoblock Andersson 15W PP, 6B4G with missing tubes and equipped with a mains transformer with tappings to give the needed B+. These will do as guinea-pigs.
All the above with the assumption that 500-600V rms is enough to drive the Script-panels.
The only modification needed would be to terminate the transformer secondary.
Why didn´t I think of that,
?Any suggestions of where to begin with the RC-termination?
An El Cheapo version could be built around four 6BL7 that has just a tad more Miller cap than one 300B. Also checked the curves to see that they could take 6k load. With two 12DW7 or one ECC83 and any CF-tube, the higher gain would make it work with just one driver gain-stage:
www.revintage.se/Script6BL7.pdf
One could use either the Script stepup, any descent PP transformer or maybe a 115:115V (or 230:230V) mains toroid with the windings in series.
Have two monoblock Andersson 15W PP, 6B4G with missing tubes and equipped with a mains transformer with tappings to give the needed B+. These will do as guinea-pigs.
All the above with the assumption that 500-600V rms is enough to drive the Script-panels.
revintage said:
Checked a little more and found that 6BX7 probably is a better choice since it has lower capacitance, lower plate resistance and higher Gm. The only thing on the downside is a slightly lower mu of 10 instead of 6BL7s 15.
The reason for the capacite coupling of the stators is that MLs signal-sensing HV-board has the HV related to ground.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.