transformer specs
Hi!
In the schematic there is a 500 written near the output transformer: is it the reflected primary impedance with a secondary load of 8 ohm?
Thanks!
Ciao
Paolo
Hi!
In the schematic there is a 500 written near the output transformer: is it the reflected primary impedance with a secondary load of 8 ohm?
Thanks!
Ciao
Paolo
Hi, I looked at Gabriel's distortion results and I was so curious that I did my own simulation with just basic approximations (references : Ed. Morgan Jones 3rd edition p. 380 , 6c33c russian/english schematic for curves with 60 W dissipation, The Valve Wizard -Single Ended for the power formulas).
I could get to very low distortion for Gabriel's amplifier output stage given 16 watt of max output ... ((Vquiescent - (Vmax + Vmin )/2)) / ( Vmax - Vmin) %
(210 - ((100 + 310 ) /2)) / ( 210 ) = 2.4% and lower, I had 110 V that I arbitrarily reduced to 100 given line tracing errors...
With parameters 210 V , - 75 V bias, 220 ma, 500R primary inductance, I have no idea how it is possible to reach more than 16 watts without increasing the distortion from clipping on the negative biased grid which is getting so negative than there is reverse current beyond around 310 V.
I did some calculation for myself and got 4.5 % 2nd at 13.9 watts output for the lundhal 1627 (650R), bias - 73, HT: 210 V, 180 ma.
I am trying to understand how tube amplifiers works and someday build my own!!!
Gabriel
I could get to very low distortion for Gabriel's amplifier output stage given 16 watt of max output ... ((Vquiescent - (Vmax + Vmin )/2)) / ( Vmax - Vmin) %
(210 - ((100 + 310 ) /2)) / ( 210 ) = 2.4% and lower, I had 110 V that I arbitrarily reduced to 100 given line tracing errors...
With parameters 210 V , - 75 V bias, 220 ma, 500R primary inductance, I have no idea how it is possible to reach more than 16 watts without increasing the distortion from clipping on the negative biased grid which is getting so negative than there is reverse current beyond around 310 V.
I did some calculation for myself and got 4.5 % 2nd at 13.9 watts output for the lundhal 1627 (650R), bias - 73, HT: 210 V, 180 ma.
I am trying to understand how tube amplifiers works and someday build my own!!!
Gabriel
Hi to you all,
Hi Gabriel, I did put the checking point to 0.22 V and I get around 185V at the anode 6C33C. Since I did it without connecting the speakers (as recommended to this measurement) and I hadn't time to pursue I cannot tell you if the sound was better, I will do this evening.
This is probably because I have chosen a secondary with an output of 150V alternate C that gives 212 V CC.
Maybe I should think on applying voltages between 160V and 180V to the anode. But in this case the current should be higher. Something like the coordinates given by navistordave.
If I go this way, the ideal should be to recalculate the bias voltage using the diagram of the tube (see attachment).
In my last listening test, I got the better sound with 176V anode and 275 mA. With a Bias of -65V.
I would like, but I don't know how, to calculate the ideal charge bar and the ideal Bias based on the diagram and the secondary PS transformer voltage which becomes to 212V CC at the output of the power supply. By the way the current for this secondary in the transformer is 600 mA.
To simplify, I get 212V CC and 600 mA after rectifying.
Based on this and the 6c33c tube operating diagram, what is the best Bias voltage, and current and voltage to apply to the anode of the 6c33c with the R17 as it is (see the amps schematics) ???
Please, to those who have technical skills ...
In advance thanks.
José
Hi Gabriel, I did put the checking point to 0.22 V and I get around 185V at the anode 6C33C. Since I did it without connecting the speakers (as recommended to this measurement) and I hadn't time to pursue I cannot tell you if the sound was better, I will do this evening.
This is probably because I have chosen a secondary with an output of 150V alternate C that gives 212 V CC.
Maybe I should think on applying voltages between 160V and 180V to the anode. But in this case the current should be higher. Something like the coordinates given by navistordave.
If I go this way, the ideal should be to recalculate the bias voltage using the diagram of the tube (see attachment).
In my last listening test, I got the better sound with 176V anode and 275 mA. With a Bias of -65V.
I would like, but I don't know how, to calculate the ideal charge bar and the ideal Bias based on the diagram and the secondary PS transformer voltage which becomes to 212V CC at the output of the power supply. By the way the current for this secondary in the transformer is 600 mA.
To simplify, I get 212V CC and 600 mA after rectifying.
Based on this and the 6c33c tube operating diagram, what is the best Bias voltage, and current and voltage to apply to the anode of the 6c33c with the R17 as it is (see the amps schematics) ???
Please, to those who have technical skills ...
In advance thanks.
José
Attachments
Hi,
I could manage to put the 220 mV to the R17 as suggested by Gabriel.
To get this value (very constant) I had to set the Bias to -70.5 and the anode voltage was 183V. The issue with the high frequencies came back, the power at the ear decreased as well as dynamics.
Then I made a series of new tests paying with the Bias voltage, between -60V and -70V.
The best results were when Bias was -64V, current 258 mA and anode voltage 176V. With these values, all became again nice and stronger.
I think that it should be a good thing that I could get the calculation I explained in my previous post.
Best Regards.
José
I could manage to put the 220 mV to the R17 as suggested by Gabriel.
To get this value (very constant) I had to set the Bias to -70.5 and the anode voltage was 183V. The issue with the high frequencies came back, the power at the ear decreased as well as dynamics.
Then I made a series of new tests paying with the Bias voltage, between -60V and -70V.
The best results were when Bias was -64V, current 258 mA and anode voltage 176V. With these values, all became again nice and stronger.
I think that it should be a good thing that I could get the calculation I explained in my previous post.
Best Regards.
José
Yes those were the custom Electra print OPT's. I was limiting the plate dissipation to 45-50W for best tube life/reliability/safety. The new amp will have wider spacing between the output toobz. I will have to re-do the impedance measurements; I think I was getting closer to 700 ohms into the 8-ohm resistive load woth the Hong Kong OPT's.
For 600 ohms and assuming a saturation voltage of 40V with the 45W plate dissipation one gets approximately 180V and 225 mA. Actually running the 160-170V and higher current provides better performance at lower power settings. With the 6KV8 triode-pentode drivers I have the option of running triode mode or pentode mode with a local feedback loop to the plate or cathode of the preceding voltage amplifier stage.
I think it will be cleaner sounding, less distortion with lower HT . Example : 13.6 Watts , 3.55 % 2nd harmonics at bias -40 , 350ma, 140 V HT. Also for this setting the load line is closer to the tube dissipation limit and in a more linear region of the Ia/Va curves.
Also less variation in the HT voltages is required :
183 V variation vs. 230 V variation for a 210 V HT at 180 ma
Also less variation in the HT voltages is required :
183 V variation vs. 230 V variation for a 210 V HT at 180 ma
Hi Jose
Try to use the tube at 210v with 220mA or more if is possible.
the maximum current allowed for this tube at 210v is 280mA but is to much for your power transformer if he has only 600mA for both tubes.
for Gabriel
In the future i will make complete tests on 8 ohm loading with frequency response distortion and power measurement. I will attach some photo with all the mesurement equipment on table
Maybe my last test was wrong, will se.
Try to use the tube at 210v with 220mA or more if is possible.
the maximum current allowed for this tube at 210v is 280mA but is to much for your power transformer if he has only 600mA for both tubes.
for Gabriel
In the future i will make complete tests on 8 ohm loading with frequency response distortion and power measurement. I will attach some photo with all the mesurement equipment on table
Maybe my last test was wrong, will se.
Running a choke input filter allows much higher current to be drawn from the power transformer winding for a given temperature rise. That is because on the condenser input filter the current is drawn only during a small portion of the input cycle, while with the choke input the current draw is over a much greater portion. So the average current during conduction on the choke input power supply is a factor of at least 2-3x.
Another advantage is that the harmonics are lowered, resulting in quiet operation.
I actually corresponded with the Hong Kong guy when I ordered the transformer sets. He had sent me a schematic; I provided a critique (it suffered severe lack of beef in the drive circuit, along with other problems) He replied that the schematic was for newbies, not hard core EE types.
My chassis is almost ready to mount the components, and the PCB 123 fabs are coming in on Monday.
Another advantage is that the harmonics are lowered, resulting in quiet operation.
I actually corresponded with the Hong Kong guy when I ordered the transformer sets. He had sent me a schematic; I provided a critique (it suffered severe lack of beef in the drive circuit, along with other problems) He replied that the schematic was for newbies, not hard core EE types.
My chassis is almost ready to mount the components, and the PCB 123 fabs are coming in on Monday.
Hi all,
Gabriel I have done tests into a wide range of voltage and current. And, again, the best sound, dynamics and spectral image of the music come with lower voltages, between 170 and 176 V to the anode 6c33c.
At bias -60V, anode 170V I reach a current of 319 mA and at bias -64V, anode 176V I get 285mA.
It is difficult to say by ear, but the best is with Bias between -62.5 and -64V.
If I try to go with Bias under -60V, the current goes very high > 320 mA.
Opinions ? .....
Best Regards.
José
Gabriel I have done tests into a wide range of voltage and current. And, again, the best sound, dynamics and spectral image of the music come with lower voltages, between 170 and 176 V to the anode 6c33c.
At bias -60V, anode 170V I reach a current of 319 mA and at bias -64V, anode 176V I get 285mA.
It is difficult to say by ear, but the best is with Bias between -62.5 and -64V.
If I try to go with Bias under -60V, the current goes very high > 320 mA.
Opinions ? .....
Best Regards.
José
I have made good progress building my SE 6S33S stereo amp. As stated before I had the Eastern Audio Xfmr Set; 2 OPTs, the 5H .6A Choke and the 320 Va power. Additional iron installed was a 30H 40mA choke for the bias supply and the 2 Electra Print plate chokes for the 6S33S grid drive. I went with 6EA7's for the drive tubes. Dialing the amp in; I wound up running the 120V primary tap and adding a 330uF cap in front of the 5H choke. Running 250 mA on each tube the voltages are as follows:
6S33S Plate: 188V
Cathode (feeding the OPT Primary) 3.4V (on each tube)
Grid (Also Drive Tube Plate): -60 and -57V
Drive Tube Cathode -235V
Power XFMR is very quiet and runs cool; I will get pix and frequency response plots soon.
6S33S Plate: 188V
Cathode (feeding the OPT Primary) 3.4V (on each tube)
Grid (Also Drive Tube Plate): -60 and -57V
Drive Tube Cathode -235V
Power XFMR is very quiet and runs cool; I will get pix and frequency response plots soon.
Following are the results of the AC tests. The lo mu section of the 6EA7 gain is just 5 so the signal generator was only providing low output power. Following measurements are open loop.
Output Impedance: 1.5 Ohms (of which 0.4 is from the tube; the rest from the OPT)
Output Voltage to 5 Ohms: 4.0V P-P
Cathode Voltage (OPT Primary) 42V P-P
Grid Voltage: 67V P-P
Frequency Response: 10 Hz to 30 kHz
So the drive requirement is on the order of 225v pk
Supply voltage on drive tube = 60 (6s33s bias) + 225 + 50 (6EA7 saturation) + 55 (6EA7 bias) = 390V
Cathode resistor 6EA7 is 2K.
Output Impedance: 1.5 Ohms (of which 0.4 is from the tube; the rest from the OPT)
Output Voltage to 5 Ohms: 4.0V P-P
Cathode Voltage (OPT Primary) 42V P-P
Grid Voltage: 67V P-P
Frequency Response: 10 Hz to 30 kHz
So the drive requirement is on the order of 225v pk
Supply voltage on drive tube = 60 (6s33s bias) + 225 + 50 (6EA7 saturation) + 55 (6EA7 bias) = 390V
Cathode resistor 6EA7 is 2K.
& for fun!Interestingly it appears that the Eastern Audio OPTs are about 10.5:1 step-down on the 8-Ohm tap; about 900 ohms primary impedance. So they are remarkably close to the nominal primary impedance that I was running on my original Electra-Print amp. (about 450 Ohms if I run the 8=Ohm speaker on the 16 Ohm tap) Since my amp is a Cathode Follower setup, the smaller output iron works better; the low impedance drive slides the OPT frequency response lower by about a half an octave. I will complete the dial in of the drive stage (LOVE that mu of 5 drive tube) and the voltage amp (planning on doing about 4-6 dB feedback) and will post the pix and schematics.
Hi all.
Here is a photo of the Duplex 2010 ( two parallel 6C33 per S.E. channel ) output 25W. It uses either two self compensated OPTs or one "double endded" OPT - respectively styles SC-OPT and SC-SCC ). The latter can be switchable manually or remotely, in order to select the sonic stage you prefer, for each musical work.
I will try to enclose also a photo of the OPTs hoping the message will not be too loaded, otherwise I'll send the second photo later.
The tests will be carried on this week and I shall not fail to send you the frequency range and distortion graphs View attachment Duplex_Two Self compensated and one double ended OPT_008.pdf .
Cheers,
Ari View attachment Duplex 2010 rear view 003.pdf
Here is a photo of the Duplex 2010 ( two parallel 6C33 per S.E. channel ) output 25W. It uses either two self compensated OPTs or one "double endded" OPT - respectively styles SC-OPT and SC-SCC ). The latter can be switchable manually or remotely, in order to select the sonic stage you prefer, for each musical work.
I will try to enclose also a photo of the OPTs hoping the message will not be too loaded, otherwise I'll send the second photo later.
The tests will be carried on this week and I shall not fail to send you the frequency range and distortion graphs View attachment Duplex_Two Self compensated and one double ended OPT_008.pdf .
Cheers,
Ari View attachment Duplex 2010 rear view 003.pdf
Here is the second photo.
Ari View attachment Duplex_Two Self compensated and one double ended OPT_008.pdf
Ari View attachment Duplex_Two Self compensated and one double ended OPT_008.pdf
On the OPT's you should get the primary inductance and primary leakage inductance. Even if you do not have an LCR meter these measurements are simple and easy to get. With the LCR you just measure at about 100Hz and directly measure the primary inductance with the secondary open and shorted. If your LCR also has a resistance function then put the secondary resistor (IE 8 Ohms) and directly measure primary impedance.
If you do not have the LCR then a signal generator and a 1K-Ohm resistor are required. Also an AC voltmeter is required. With secondary open adjust sig gen to provide equal voltages across the primary and the 1K resistor. Increase frequency to about 1 kHz and then short the secondary. Measure across the 1K and the primary and calculate the leakage inductance.
The impedance is equal to the square of the secondary voltage divided by the primary voltage times the secondary impedance
Hi all.
Here is a photo of the Duplex 2010 ( two parallel 6C33 per S.E. channel ) output 25W. It uses either two self compensated OPTs or one "double endded" OPT - respectively styles SC-OPT and SC-SCC ). The latter can be switchable manually or remotely, in order to select the sonic stage you prefer, for each musical work.
I will try to enclose also a photo of the OPTs hoping the message will not be too loaded, otherwise I'll send the second photo later.
If you do not have the LCR then a signal generator and a 1K-Ohm resistor are required. Also an AC voltmeter is required. With secondary open adjust sig gen to provide equal voltages across the primary and the 1K resistor. Increase frequency to about 1 kHz and then short the secondary. Measure across the 1K and the primary and calculate the leakage inductance.
The impedance is equal to the square of the secondary voltage divided by the primary voltage times the secondary impedance
Hi all.
Here is a photo of the Duplex 2010 ( two parallel 6C33 per S.E. channel ) output 25W. It uses either two self compensated OPTs or one "double endded" OPT - respectively styles SC-OPT and SC-SCC ). The latter can be switchable manually or remotely, in order to select the sonic stage you prefer, for each musical work.
I will try to enclose also a photo of the OPTs hoping the message will not be too loaded, otherwise I'll send the second photo later.
Hi
Today I made complete test of my 6C33C amplifier
Unfortunately I forget my camera at home and no photo yet, but the amplifier is it at my office and I will make some photos to show you what I measure.
Equipment and measurements conditions.
Function Generator 1 Hz-1 MHz Made in Romania
Output voltage 2.5V rms.
ShibaSoku 760E/2 Automatic Distortion Meter Made in Japan
20Mhz Oscilloscope Made in China
Output loading 8 ohm
OPT Ra 550 ohm
6C33C voltage 210V
6C33C current 220 mA
All the voltage exactly like in Borbely schematic
Input AC voltage regulated 230V
Preamp tubes General Electric NOS 6SN7GT
Driver tubes Russian NOS 6C2C
The measurements were made with NFB
The results
Sine wave 1 KHz
Output 2.83v rms 0.22% bandwidth measurements at this power 10Hz – 100 KHz 0dB
6.3v rms 0.35%
9v rms 0.65%
11v rms 1.35% bandwidth measurements at this power 25Hz – 25 KHz 0dB at this power perfect sine wave without attenuation.
Maximum voltage 14v rms 19%
All this value were read with the Distortion meter
Photos in a few days if I will have free time.
Best regards
Today I made complete test of my 6C33C amplifier
Unfortunately I forget my camera at home and no photo yet, but the amplifier is it at my office and I will make some photos to show you what I measure.
Equipment and measurements conditions.
Function Generator 1 Hz-1 MHz Made in Romania
Output voltage 2.5V rms.
ShibaSoku 760E/2 Automatic Distortion Meter Made in Japan
20Mhz Oscilloscope Made in China
Output loading 8 ohm
OPT Ra 550 ohm
6C33C voltage 210V
6C33C current 220 mA
All the voltage exactly like in Borbely schematic
Input AC voltage regulated 230V
Preamp tubes General Electric NOS 6SN7GT
Driver tubes Russian NOS 6C2C
The measurements were made with NFB
The results
Sine wave 1 KHz
Output 2.83v rms 0.22% bandwidth measurements at this power 10Hz – 100 KHz 0dB
6.3v rms 0.35%
9v rms 0.65%
11v rms 1.35% bandwidth measurements at this power 25Hz – 25 KHz 0dB at this power perfect sine wave without attenuation.
Maximum voltage 14v rms 19%
All this value were read with the Distortion meter
Photos in a few days if I will have free time.
Best regards
Last edited:
Excellent results. However, sorry, I do not use ( read : like ) the overall feed back.
Hereafter the results of the tests made with the ARTA solftware on the
"Duplex" , fitted with the "double ended OPT" . Those with the "self compensated" will follow.
1) with 10V RMS output on 8 ohm ( = 12,5 W )
THD 3,11 % IMD ( 18,5 and 19,5 kHz) 2,9 %
Frequency range at -1dB 20 Hz to 20 kHz almost flat
Z out = 1,4 ohms Damping factor = 5,7
2) with 18V RMS the frequency range, at -3dB is < 15 Hz and > 50 kHz
No overall feed back but just local feedback in the driver stage.
I will continue the tests with the self compensated OPTs before end week.
Cheers,
Ari
Hereafter the results of the tests made with the ARTA solftware on the
"Duplex" , fitted with the "double ended OPT" . Those with the "self compensated" will follow.
1) with 10V RMS output on 8 ohm ( = 12,5 W )
THD 3,11 % IMD ( 18,5 and 19,5 kHz) 2,9 %
Frequency range at -1dB 20 Hz to 20 kHz almost flat
Z out = 1,4 ohms Damping factor = 5,7
2) with 18V RMS the frequency range, at -3dB is < 15 Hz and > 50 kHz
No overall feed back but just local feedback in the driver stage.
I will continue the tests with the self compensated OPTs before end week.
Cheers,
Ari
I was able to take a look at your Amplifier. You should try the 6EM7 octal drive tubes. Same pin-out as those wimpy 6SN7's; They have a mu of 65 triode and a 10 W mu 5 triode with 750 ohms Rp ;-). Cheap and readily available; there is also a Compactron version 6FM7; even cheaper. The big triode is rated at 500V; you could run a 7.5k plate resistor with about 30 mA thru it to drive a 27K grid resistor on the 6S33S. Your ears will thank you!!!
Excellent results. However, sorry, I do not use ( read : like ) the overall feed back.
Hereafter the results of the tests made with the ARTA solftware on the
"Duplex" , fitted with the "double ended OPT" . Those with the "self compensated" will follow.
1) with 10V RMS output on 8 ohm ( = 12,5 W )
THD 3,11 % IMD ( 18,5 and 19,5 kHz) 2,9 %
Excellent results. However, sorry, I do not use ( read : like ) the overall feed back.
Hereafter the results of the tests made with the ARTA solftware on the
"Duplex" , fitted with the "double ended OPT" . Those with the "self compensated" will follow.
1) with 10V RMS output on 8 ohm ( = 12,5 W )
THD 3,11 % IMD ( 18,5 and 19,5 kHz) 2,9 %
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.