I have a push pull el34 valve/tube amplifier for which the design shows it needs a 5 wire primary.
I have found a cheap 3 wire primary transformer, can I use this ?
I obviously need to connect grid2 to this anode which currently is done using 270 resistors.
Will this be ok ?
What other options are there ?
Triode mode ? pentode mode ?
What grid 2 resistor value should I use ?
I have found a cheap 3 wire primary transformer, can I use this ?
I obviously need to connect grid2 to this anode which currently is done using 270 resistors.
Will this be ok ?
What other options are there ?
Triode mode ? pentode mode ?
What grid 2 resistor value should I use ?
OK, look up "ultralinear" power amp circuits. They use an output transformer with the standard center tap and the end wires going to the plates, but they also have taps part way between center tap and the end for each screen grid. That is your five wire primary.
To use a common center tapped primary, you have to then make an alternate power source for your screens. Look at just about any tube guitar amp schematic for examples of that. Well, except ultralinear ones.
To use a common center tapped primary, you have to then make an alternate power source for your screens. Look at just about any tube guitar amp schematic for examples of that. Well, except ultralinear ones.
It seems that you have wired the EL34s up as triode-strapped pentodes, that will lower the output power quite a bit, if you are looking for more power, then you should create a separate node on the power supply after the B+ to supply the screen grid directly. Just make sure that its voltage is lower than the plate voltage at idle. If you know the specs for your output transformer, we can give you more feedback (no pun intended ;-).
First a silly question: Is the 3 wire transformer suitable for EL34s?
Typically 3400 ohm primary impedance, rated for about 50W output power?
What voltages are we talking about? In pentode mode (both screens are being fed from the same HT supply) most designs call for 400-450V for both screens and anodes. In Ultralinear mode (screens connected to taps on the output transformer) some designs go up to 500V. The sound will be cleaner, but at the expense of a little less maximum power. Triode mode is even cleaner, but you loose half of the power compared to pentode mode. Preamp gain cannot compensate for the loss of gain in the output valves: it is the maximum output power that is limited.
Take good care of the g2 HT supply: classic designs use a choke to filter the g2 supply. Two reasons: the g2 supply needs to be clean. Any power supply ripple will be amplified and show up at the speakers. The other reason is that the g2 current in EL34's increases a lot when the signal goes up. When the g2 supply sags under signal conditions, that will increase distortion. You have to find out yourself if it is the kind of distortion that you like or not.
Beam tetrodes like the 6L6 draw a lot less g2 current, so there you can filter with a big resistor instead of an (expensive) choke.
Typically 3400 ohm primary impedance, rated for about 50W output power?
What voltages are we talking about? In pentode mode (both screens are being fed from the same HT supply) most designs call for 400-450V for both screens and anodes. In Ultralinear mode (screens connected to taps on the output transformer) some designs go up to 500V. The sound will be cleaner, but at the expense of a little less maximum power. Triode mode is even cleaner, but you loose half of the power compared to pentode mode. Preamp gain cannot compensate for the loss of gain in the output valves: it is the maximum output power that is limited.
Take good care of the g2 HT supply: classic designs use a choke to filter the g2 supply. Two reasons: the g2 supply needs to be clean. Any power supply ripple will be amplified and show up at the speakers. The other reason is that the g2 current in EL34's increases a lot when the signal goes up. When the g2 supply sags under signal conditions, that will increase distortion. You have to find out yourself if it is the kind of distortion that you like or not.
Beam tetrodes like the 6L6 draw a lot less g2 current, so there you can filter with a big resistor instead of an (expensive) choke.
Do you know the primary impedance of your output transformer?
My guess would be around 2800 ohm, since Fender mainly uses 6L6 outputs.
Four EL34s running at 350V into 1700 ohms using cathode bias will give you around 70W. If your OPT is indeed 2800 ohm, the power would be closer to 50W than to 70W.
First step to get more power is to change the cathode bias to fixed bias.
After that, the quick 'n' dirty way to get more power is to use 8 ohm speakers on the 16 ohm tap or 4 ohm speakers on the 8 ohm tap. This translates to halving the primary impedance that the EL34's see. Assuming the OPT is 2800 ohms to 16/8/4 ohms, you could use it as 1400 to 8/4/2. According to the Telefunken datasheets, this would give you 88W. This is with 350V on the anodes. If the OPT has a high dc resistance you will loose some precious Volts there.
Of course your power transformer must be able to deliver enough current: 0,6A continuously at full power, more when you overdrive the EL34's.
My advice: Just try and see where it takes you
The other way to get to 100W would be to get a 450V supply.
One other thing you need to ask yourself is: do you really need 100W. I have seen '100W' Marshalls that put out no more than 80W.
My guess would be around 2800 ohm, since Fender mainly uses 6L6 outputs.
Four EL34s running at 350V into 1700 ohms using cathode bias will give you around 70W. If your OPT is indeed 2800 ohm, the power would be closer to 50W than to 70W.
First step to get more power is to change the cathode bias to fixed bias.
After that, the quick 'n' dirty way to get more power is to use 8 ohm speakers on the 16 ohm tap or 4 ohm speakers on the 8 ohm tap. This translates to halving the primary impedance that the EL34's see. Assuming the OPT is 2800 ohms to 16/8/4 ohms, you could use it as 1400 to 8/4/2. According to the Telefunken datasheets, this would give you 88W. This is with 350V on the anodes. If the OPT has a high dc resistance you will loose some precious Volts there.
Of course your power transformer must be able to deliver enough current: 0,6A continuously at full power, more when you overdrive the EL34's.
My advice: Just try and see where it takes you
The other way to get to 100W would be to get a 450V supply.
One other thing you need to ask yourself is: do you really need 100W. I have seen '100W' Marshalls that put out no more than 80W.
The output transformer is 2000R.
I thought I was using fixed bias with 120R connected to two cathodes ?
Unless you mean fixed current ?
My power transformer supplies 200VA so about 900mA.
The output transformer only has a 4R tap so I cant do much there.
The basis of the circuit is from a 100watt Simms-Watts PA I used to have but pentode connected instead of ultra linear. But I don't use a negative voltage bias I just rely on a 130R double cathode resistor to get the bias.
The Simms-Watts PA was extremely loud for 100 watts. I couldn't stand anywhere near the speaker without hurting my ears when it was on full power.
I thought I was using fixed bias with 120R connected to two cathodes ?
Unless you mean fixed current ?
My power transformer supplies 200VA so about 900mA.
The output transformer only has a 4R tap so I cant do much there.
The basis of the circuit is from a 100watt Simms-Watts PA I used to have but pentode connected instead of ultra linear. But I don't use a negative voltage bias I just rely on a 130R double cathode resistor to get the bias.
The Simms-Watts PA was extremely loud for 100 watts. I couldn't stand anywhere near the speaker without hurting my ears when it was on full power.
It shows ultra linear on diagram but the connections are just pins on a pcb.
It can be wired for triode mode, ultra linear or pentode mode.
It can be wired for triode mode, ultra linear or pentode mode.
An externally hosted image should be here but it was not working when we last tested it.
Fixed bias means a negative supply voltage is applied to the grids. The voltage can be adjustable, but it is still considered fixed. it is fixed because it does not vary with the action of the circuit.
What you have is cathode biased tubes. It is not fixed, because the voltage developed across the resistor depends upon the tube current.
In other words in a fixed bias amp, I could remove the B+ from the tube entirely and the bias voltage would remain. If I removed the B+ from the cathode biased tube, the bias voltage would disappear.
Hi Guys
If you want 100W why are using cathode bias? That is the limiting factor here. Maybe you should read about how tube power amps work.
Wire up a bias supply and assure that the bias pot provides a sweep from around 10% of absolute B+ to 15%. If B+ is 470V, then -47V to about -60V or so. This assures that you can turn any tube sample 'off' and have complete control over how 'on' it is. Ideally, individual bias pots and current-sense resistors for safe ground-referenced bias measurment and adjustment.
Another detail is that you can wire the tubes as triodes and this is a good tone for some players. Make sure the screen resistor is ALWAYS in series with pin-4. basically, treat the other end of the screen-stop as the "flexible" connection point.
Despite what some readers believe, there is NO specific OT primary impedance that is "ideal" for each tube type. The common big-bottle tube types used in MI all work into a very widely overlapped range of loads quite happily. This means that if there is heater current to support EL-34s, then you can readily plug in 6L6, 6550, KT-66/77/88 and all the equivalents, and in most cases 6V6 also. Just bias the new tube properly for the voltage environment.
Our FAQ and our TUT-series books explain all of this in detail.
Have fun
Kevin O'Connor
London Power/Power Press
If you want 100W why are using cathode bias? That is the limiting factor here. Maybe you should read about how tube power amps work.
Wire up a bias supply and assure that the bias pot provides a sweep from around 10% of absolute B+ to 15%. If B+ is 470V, then -47V to about -60V or so. This assures that you can turn any tube sample 'off' and have complete control over how 'on' it is. Ideally, individual bias pots and current-sense resistors for safe ground-referenced bias measurment and adjustment.
Another detail is that you can wire the tubes as triodes and this is a good tone for some players. Make sure the screen resistor is ALWAYS in series with pin-4. basically, treat the other end of the screen-stop as the "flexible" connection point.
Despite what some readers believe, there is NO specific OT primary impedance that is "ideal" for each tube type. The common big-bottle tube types used in MI all work into a very widely overlapped range of loads quite happily. This means that if there is heater current to support EL-34s, then you can readily plug in 6L6, 6550, KT-66/77/88 and all the equivalents, and in most cases 6V6 also. Just bias the new tube properly for the voltage environment.
Our FAQ and our TUT-series books explain all of this in detail.
Have fun
Kevin O'Connor
London Power/Power Press
Hi Guys
You seemed to miss the point.
You can definitely cathode bias the tubes if you wish, or if this provides the tone you want. You will just not get 100W this way.
Also, you need more than -50V to assure the tubes will be under your control.
With respect to cathode biasing: at the very least you should use two Rk's for the four tubes. Each Rk should be shared by a push-pull pair to promote dynamic balancing. You can take this a step further and have individual Rk's, losing dynamic balance but allowing tubes to be pulled safely for power reductions and/or asymmetric tones.
Without Ck the output will be cleaner but lower in power. With Ck added to any of the above, the amp will transition to class-AB at some power level.
How much power is spent in Rk is of little concern to output power. Rather, it is relevant to placement of Rk and Ck, and in sizing Rk power-wise.
Have fun
Kevin O'Connor
You seemed to miss the point.
You can definitely cathode bias the tubes if you wish, or if this provides the tone you want. You will just not get 100W this way.
Also, you need more than -50V to assure the tubes will be under your control.
With respect to cathode biasing: at the very least you should use two Rk's for the four tubes. Each Rk should be shared by a push-pull pair to promote dynamic balancing. You can take this a step further and have individual Rk's, losing dynamic balance but allowing tubes to be pulled safely for power reductions and/or asymmetric tones.
Without Ck the output will be cleaner but lower in power. With Ck added to any of the above, the amp will transition to class-AB at some power level.
How much power is spent in Rk is of little concern to output power. Rather, it is relevant to placement of Rk and Ck, and in sizing Rk power-wise.
Have fun
Kevin O'Connor
Hi Guys
I thought since you had a Fender OT and wanted 100W that you would be using more typical supply voltages. Looking back at your previous post I see there is only 350Va. That would put the fixed-bias grid voltage around -35V or so. With cathode bias, more like -19V. Your 20V reading across Rk lines up with this.
Note these voltages are with respect to the cathode, which is the tube's reference point.
You might get about 60Wrms with this arrangement, mostly limited by the supply voltage. You lose the cathode bias voltage right off the top as far as supply utility goes. Then there is the internal resistance of the tube that further restricts how close the plate voltage can be pulled towards the cathode voltage.
60W through normal guitar speakers in large cabinets will be very loud. If you need it louder, add more speakers and arrange them to focus the sound. SPKR shows cabinet layouts for best sound on stage or at home.
Have fun
Kevin O'Connor
I thought since you had a Fender OT and wanted 100W that you would be using more typical supply voltages. Looking back at your previous post I see there is only 350Va. That would put the fixed-bias grid voltage around -35V or so. With cathode bias, more like -19V. Your 20V reading across Rk lines up with this.
Note these voltages are with respect to the cathode, which is the tube's reference point.
You might get about 60Wrms with this arrangement, mostly limited by the supply voltage. You lose the cathode bias voltage right off the top as far as supply utility goes. Then there is the internal resistance of the tube that further restricts how close the plate voltage can be pulled towards the cathode voltage.
60W through normal guitar speakers in large cabinets will be very loud. If you need it louder, add more speakers and arrange them to focus the sound. SPKR shows cabinet layouts for best sound on stage or at home.
Have fun
Kevin O'Connor
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.