I have been looking for a good 6BQ5 push-pull project. I found the followng: OddWatt - ECC802S SRPP / EL84 (6BQ5) Push-Pull Tube Amp
here: https://diyaudioprojects.com/Tubes/EL84-Push-Pull/
My question regards the B+ on the various projects.. In the original thread, this project is referenced:
Many here are likely familiar wth these projects.
Here is my question. In the referenced Leibowitz project the B+ is 260v with 250v applied to the plates. In the 'final' version of this amp, updated 9 October 2012, the B+ is only 210v. Why the drop n voltage? Seems like a lot.
Roger
here: https://diyaudioprojects.com/Tubes/EL84-Push-Pull/
My question regards the B+ on the various projects.. In the original thread, this project is referenced:
Compact Hi-Fi Power Amplifier by Melvin Leibowitz
found here: https://diyaudioprojects.com/Technical/Papers/Compact-Hi-Fi-Power-Amplifier/Many here are likely familiar wth these projects.
Here is my question. In the referenced Leibowitz project the B+ is 260v with 250v applied to the plates. In the 'final' version of this amp, updated 9 October 2012, the B+ is only 210v. Why the drop n voltage? Seems like a lot.
Roger
As in away from that amp.
Ifn you want to do a PP EL84/6BQ5 i would suggest on eof the three outstanding PP amplifiers here: Baby Huey, Red-Light District, and El Cheapo.
We did a Cloass A triodeb variation nof the El Cheapo. One of my best 2 amplifiers (but very modest power — 3.2 w at clipping). The othe amp is aFIrstwatt SIT-3.
You have to work hard to make a bad sounding EL84 amp, but the Oddwatt seems tome to be a waste of resources. ie self-splitter throws away half the power, and the tube sections in the totem pole could be ued much more effectively
dave
There are lots of worthy PP 6BQ5 EL84 amplifiers.
The Oddwatt is one of them.
Simple (one of the most simple)
Easy to wire
Quality parts (use those 8k pp 25 Watt Edcor and UL taps . . . and get that perfect looking 1kHz square wave without using global negative feedback.
Class A output stage
The only negative feedback are the output tube cathode to cathode, and the Ultra Linear mode.
The "beauty" of a tube amplifier is in the eye of the beholder.
Simplicity, Cost, Power, Distortion, Weight, Looks (and WAF), ease of building, ease of troubleshooting, lack of trouble from oscillations, etc.
Oh, someone said the OddWatt is wasteful. So is almost every tube amplifier.
So go build a Class D Amp for efficiency, or at least build solid state amplifiers.
i have built self inverting amplifiers. For the pp output stage current source: used an LM317, an NPN transistor, and a 5Hy 200mA Choke with a series resistor for bias.
All those amplifiers "sounded" very good.
How many of you have ever heard a decently designed and well built self inverting amplifier with good parts, and good output transformers?
You can bore out a 305 V8 engine block for more power, but why do that to your family car?
Just my $0.03
The Oddwatt is one of them.
Simple (one of the most simple)
Easy to wire
Quality parts (use those 8k pp 25 Watt Edcor and UL taps . . . and get that perfect looking 1kHz square wave without using global negative feedback.
Class A output stage
The only negative feedback are the output tube cathode to cathode, and the Ultra Linear mode.
The "beauty" of a tube amplifier is in the eye of the beholder.
Simplicity, Cost, Power, Distortion, Weight, Looks (and WAF), ease of building, ease of troubleshooting, lack of trouble from oscillations, etc.
Oh, someone said the OddWatt is wasteful. So is almost every tube amplifier.
So go build a Class D Amp for efficiency, or at least build solid state amplifiers.
i have built self inverting amplifiers. For the pp output stage current source: used an LM317, an NPN transistor, and a 5Hy 200mA Choke with a series resistor for bias.
All those amplifiers "sounded" very good.
How many of you have ever heard a decently designed and well built self inverting amplifier with good parts, and good output transformers?
You can bore out a 305 V8 engine block for more power, but why do that to your family car?
Just my $0.03
Last edited:
Me not talkig about that kind of waste. More a cae of te same tubes (resources) could, IMO, be used more effectively.
dave
Joe Roberts: "I never met a 2A3 amp I did not like". Thanks Joe, for everything, including publishing "Not Your Father's Dyna".
How about a bad sounding PP EL84 amplifier?
Me: "I never met a push pull EL84 amp I did not like".
The worst PP EL84 amplifier is the one that did not get built.
(the Poster was afraid it would not be good, or would be too complex to work without having to troubleshoot and tweak out the oscillation, for example).
How about a bad sounding PP EL84 amplifier?
Me: "I never met a push pull EL84 amp I did not like".
The worst PP EL84 amplifier is the one that did not get built.
(the Poster was afraid it would not be good, or would be too complex to work without having to troubleshoot and tweak out the oscillation, for example).
The ooddwatt looks easy to build. Ise it really only half the power? Does the SS CCS add any 'flavor ' to the sound? I want to avoid global feedback, as I have n way to test distortion or waveform.
And no one has addressed my original question about B+ voltage.
And no one has addressed my original question about B+ voltage.
Last edited:
Rogerayotte,
Your original question does deserve to be answered:
I see the following on the Original OddWatt PP EL84:
LM317 CCS with a 15 Ohm current set resistor. The current is 83mA (41.5mA per EL84). 0.0415A.
The B+ to the output transformer center tap is 240V. Perhaps there is a 5 Volt drop through the 1/2 primary winding; so 235V to each EL84 plate.
There is about 8V bias at the EL84 cathodes.
235V - 8V = 227V plate to cathode.
227V x 0.0415A = 9.42 Watts plate plus screen dissipation. That is good, it is not near the maximum spec.
I can not imagine using less B+ voltage than 240V at the output transformer center tap.
Perhaps another version of the amp uses more than 41.5mA.
If I consider a 12 Watt dissipation at 210V, the current would be 57mA, way too much for an EL84 quiescent current.
But I would not design an amplifier to stand 12Watts in an EL84 plate plus screen.
Yes, I have owned EL84 amplifiers that did stand 12 Watts or more, they just run to hot for me.
There are probably lots of different mods to the original OddWatt PP EL84.
But I can not see anything wrong with the Original circuit in Post # 1.
Do not worry about the "sound" of a solid state CCS. I have used them in cathode circuits, and as plate loads.
If a solid state CCS "causes" a bad sound, the amplifier design did not take proper account for the CCS specs:
Minimum current
Maximum current
Minimum Voltage
Maximum Voltage
Excess voltage x current that exceeds the chip's maximum power dissipation
A heat sink that does not get rid of enough heat.
And . . .
Improper design of the amplifier stage that uses the CCS, such as the wrong bias voltage, and range of voltages with maximum signal to the amplifier.
An alternative:
A Choke can be used as a CCS, it usually requires a resistor in series to get the right bias voltage at the cathodes. That is an easy calculation.
One thing about a choke CCS is that there never is a minimum voltage.
But solid state CCS Do have a Minimum voltage, they do not operate at less than that voltage.
If the quiescent voltage is enough for a solid state CCS, it might not be enough voltage when the signal is applied, the cathodes swing up and down at 1/2 times the grid voltage swing.
Biasing using a Choke as a CCS:
At 8 Volts Bias, and 0.083A, the total resistance of the choke plus series resistor is:
8 / 0.083 = 96.4 Ohms.
A choke with a DCR of 65 Ohms, requires 31.3 Ohms. Use a 30 Ohm resistor in series with the choke.
Here I go again, I look at others projects, and then want to build one of them for myself.
A PP EL84 and Choke CCS could be a good idea.
Attention to details prevents such problems.
Your original question does deserve to be answered:
I see the following on the Original OddWatt PP EL84:
LM317 CCS with a 15 Ohm current set resistor. The current is 83mA (41.5mA per EL84). 0.0415A.
The B+ to the output transformer center tap is 240V. Perhaps there is a 5 Volt drop through the 1/2 primary winding; so 235V to each EL84 plate.
There is about 8V bias at the EL84 cathodes.
235V - 8V = 227V plate to cathode.
227V x 0.0415A = 9.42 Watts plate plus screen dissipation. That is good, it is not near the maximum spec.
I can not imagine using less B+ voltage than 240V at the output transformer center tap.
Perhaps another version of the amp uses more than 41.5mA.
If I consider a 12 Watt dissipation at 210V, the current would be 57mA, way too much for an EL84 quiescent current.
But I would not design an amplifier to stand 12Watts in an EL84 plate plus screen.
Yes, I have owned EL84 amplifiers that did stand 12 Watts or more, they just run to hot for me.
There are probably lots of different mods to the original OddWatt PP EL84.
But I can not see anything wrong with the Original circuit in Post # 1.
Do not worry about the "sound" of a solid state CCS. I have used them in cathode circuits, and as plate loads.
If a solid state CCS "causes" a bad sound, the amplifier design did not take proper account for the CCS specs:
Minimum current
Maximum current
Minimum Voltage
Maximum Voltage
Excess voltage x current that exceeds the chip's maximum power dissipation
A heat sink that does not get rid of enough heat.
And . . .
Improper design of the amplifier stage that uses the CCS, such as the wrong bias voltage, and range of voltages with maximum signal to the amplifier.
An alternative:
A Choke can be used as a CCS, it usually requires a resistor in series to get the right bias voltage at the cathodes. That is an easy calculation.
One thing about a choke CCS is that there never is a minimum voltage.
But solid state CCS Do have a Minimum voltage, they do not operate at less than that voltage.
If the quiescent voltage is enough for a solid state CCS, it might not be enough voltage when the signal is applied, the cathodes swing up and down at 1/2 times the grid voltage swing.
Biasing using a Choke as a CCS:
At 8 Volts Bias, and 0.083A, the total resistance of the choke plus series resistor is:
8 / 0.083 = 96.4 Ohms.
A choke with a DCR of 65 Ohms, requires 31.3 Ohms. Use a 30 Ohm resistor in series with the choke.
Here I go again, I look at others projects, and then want to build one of them for myself.
A PP EL84 and Choke CCS could be a good idea.
Attention to details prevents such problems.
Last edited:
@rogerayotte
There is an outstanding analysis of the OddWatt circuit here -
https://tubecad.com/2014/06/blog0293.htm
Scroll to the heading "Gratuitous SRPP Circuits" about 1/2 way down.
There is an outstanding analysis of the OddWatt circuit here -
https://tubecad.com/2014/06/blog0293.htm
Scroll to the heading "Gratuitous SRPP Circuits" about 1/2 way down.
rogerayotte,
I added more info to my Post # 10.
You may want to read it again.
I am not a fan of SRPP. Others are. OK, use what works and satisfies.
I did use one SRPP to drive a 45 triode tube. I used the 45 self bias voltage to elevate the SRPP tube filaments (need to keep the filament to cathode voltage of the top SRPP tube within the spec, both at quiescent state, and peak signal Voltage swing too). And yes, that elevated voltage had a very large capacitance to ground, to keep the signal off the filaments.
If I do build a self inverting PP EL84, I will probably use a Choke CCS. I already did that for a PP 6CK4.
Whatever you decide, Have Fun designing, building, and listening!
I added more info to my Post # 10.
You may want to read it again.
I am not a fan of SRPP. Others are. OK, use what works and satisfies.
I did use one SRPP to drive a 45 triode tube. I used the 45 self bias voltage to elevate the SRPP tube filaments (need to keep the filament to cathode voltage of the top SRPP tube within the spec, both at quiescent state, and peak signal Voltage swing too). And yes, that elevated voltage had a very large capacitance to ground, to keep the signal off the filaments.
If I do build a self inverting PP EL84, I will probably use a Choke CCS. I already did that for a PP 6CK4.
Whatever you decide, Have Fun designing, building, and listening!
Logic800, thanks. I guess I was just curious about why the stated B+ was dropped so much. And thanks also for commenting on actual experience, andnot supposition.
Roger
Roger
I whole heartedly agree with the “I’ve never heard an EL84 amp I didn’t like” notion.
A couple of years ago I set out to build an PP EL84 amp that used the cheapest no name parts that I had sitting around amongst my collection of unused stuff. It sounded so good I was greatly annoyed.
Tiny and cheap Edcor input transformer/phase splitter. 6N23P input tubes with Cascode Ixys depletion mosfet CCS at 8mA on each plate, diode string for bias, cap coupled with generic polypropylene caps to UL connected to a pair EL84s into a Hammond 1620 OPT wired with a 4ohm secondary configuration. EL84s bias with Blumlein Garter arrangement, “top” resistor for each cathode bypassed with 220uF generic electrolytic cap. Power supply was a 240V isolation transformer, CLC, separate 6.3V heater transformer. Vak was ~240Vdc iirc Ik per EL84 was 40mA. Probably around 6W to 8W channel.
It had a fast open sound, maybe not the last word in refinement (probably due to the cheap iron used), a bit “loose” in the lower registers (low damping factor, the lack of addition fb no doubt), fun to listen to music through, the opposite of analytical. Played rock and roll, classical and jazz with equal aplomb.
One has to try really hard to make a poor sounding EL84 amp IMHO.
Don’t get uptight about how much power you are trying to get out of them. With speaker efficiency that starts with a 9x dB@1W@1m, nominal 8ohm, they are fine, unless you are a head banger. Or spend a lot of time listening to pipe organs or death metal, or have low efficiency monkey coffins, which probably means you need a different starting point for an amp.
Side note and to your initial query 220Vak on EL84’s works fine. Result is lower power, but they still sound great and the tubes will last a very long time. 240 to 260 Vak is probably their sweet spot though.
My 2c, ymmv etc etc
@rogerayotte. Measurements: B+ 224V, with each EL84 anode 217 and 216V respectively. Amp was warm at idle, no input, with dummy speaker loads. Noting that the published circuit diagram for the type 'D' version has B+ marked as 235V. Hope this helps.
@rogerayotte I have built the EL84 miniblocks (https://diyaudioprojects.com/Tubes/EL84-Mini-Block-Amps) by Bruce Heran as a dual mono, which basically is the Oddwatt with separate power transformers per channel.
I really like the result very much. I recommend you go for it.
B+ is 235 in mine, as in the latest schematic iteration.
I really like the result very much. I recommend you go for it.
B+ is 235 in mine, as in the latest schematic iteration.
I guess the difference between one transformer powering both channels (as in my case) vs. one for each channel. Just as a point of reference, in 2020 the complete set of parts cost me c.£600, sourcing the chassis from Italy, and the recommended set of transformers from a dealer in Germany.