4 x EL84 (6BQ5) in Push-Pull - Power Supply

Eric Lawson · 3rd April 2010, 02:44 AM

Hello all, I'm a relative "newbie" from Australia.

I've been building amps for a few years now - mostly 2 x EL84 in push-pull for 15W guitar amps for myself and a couple of mates. I've come up with a great formula over that time, using 1 x ECC83S (12AX7) and 1 x ECC82 (12AU7) in the pre-amp with a TMB tone-stack. I included Master Volume and Gain controls to provide a bit of extra versatility. I've settled on a bridge rectifier based power supply for the sound I was chasing. I'm pretty happy with the 2 x EL84 power supply I've developed (check out the attachment) - but confused when it comes to adapting something similar to a 4 x EL84 amp. Do I just beef up the VA on the power transformer secondary to 87 (280V @ 310mA? That is, add an extra 65mA each (max) for the 2 extra EL84's or would I also need to beef up the the first stage RC network. If so - any ideas? I'm just trying to make the new version as reliable as the old one. If you'd like to see the 15W power amp or my proposed 30W power amp, I can post those as well.

Sorry to rave on, but any help would be appreciated.

torrence · 3rd April 2010, 05:08 AM

Hi Eric,

I presume when you say 4 x EL84 you are planning to operate in parallel push-pull (PPP). Roughly speaking, the power supply will need to deliver twice the current, so indeed your transformer should be able to handle twice the load, but the voltage drop across your resistors in the CRC network will also be twice as much. The best way to think about the design is that you want two power supplies in parallel to give the same performance as what you have now driving two tubes. This would mean double the capacitance and half the resistance in the CRC network.

In truth, this may be neither practical nor necessary. Try out PSUD to compare the ripple and voltage sag under load for different configurations.

Note also that your output transformer will need to have half the input impedance to have similar performance to your current design.

-Eric

bob91343 · 3rd April 2010, 05:30 AM

If the tubes are running class A you probably won't have to upgrade the driver stage. But if you are running AB or B, you will need a stronger driver. Otherwise the grid current will overload it.

When paralleling tubes it's wise to include parasitic suppressors in the grid and plate leads, say 100 Ohms for a wild guess.

Eric Lawson · 07:07 AM

Thanks for your help so far Torrence and Bob91343.

Yes, I forgot to mention that the 4 x EL84's would be in parallel push-pull. I've added 2 more attachments showing the current 2 x EL84 power amp and the 4 x EL84 power amp as it will probably be built. I can't go much further with the turret board until I settle on the power supply components.

What do you guys think?

Jaap · 3rd April 2010, 07:50 AM

Quote:

Originally Posted by Eric Lawson

What do you guys think?

I think that the 10K resistors (1 watt) in your psu are too hot.

torrence · 3rd April 2010, 08:10 AM

Quote:

Originally Posted by Jaap

I think that the 10K resistors (1 watt) in your psu are too hot.

Actually, I didn't notice this before. Surely this is a typo. A current of 50 mA across 10k would drop 500V...

Eric Lawson · 5th April 2010, 02:04 AM

Quote:

Originally Posted by torrence

Actually, I didn't notice this before. Surely this is a typo. A current of 50 mA across 10k would drop 500V...

Well - with reference to the first attachment (EDLSUP15.pdf) - point A measures 360V @ 50mA, point B is 263V @ 8.4mA, point C is 231V @ 3.5mA and point D is 217V @ 1.2mA. It's no typo - in practice, I installed all 10K 2W resistors. The amp works extremely well. Please check the attachment.

For the 30W (4 x EL84) version, after torrence's coments, I was thinking of putting an extra 47uF Electrolytic before point A and changing the A-B resistor to 470R 5W as well as increasing the last 2 Electrolytics (C&D) to 15uF. Do you feel it necessary to increase the second (point B) 47uF Electrolytic as well? I guess the new board will be about 25% longer to accommodate the extra components.

Thanks again for the tips guys!

Eric Lawson · 5th April 2010, 07:28 AM

Taking your various suggestions into account, this is what I had in mind for the 30W (4 x EL84) power supply. Is it worth a try?

taj · 5th April 2010, 06:10 PM

Quote:

Originally Posted by Eric Lawson

The amp works extremely well. Please check the attachment.

Eric, what does the amp sound like? Vox AC30 type? What kind of music are you playing?

..Todd

Brit01 · 5th April 2010, 06:21 PM

Nice soldering work there Eric. Very neat. I need to take this approach.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Unity-Gain EL84 Push-Pull Power Buffer	Kashmire	Tubes / Valves	9	11th February 2010 02:27 AM
6BQ5 Push-Pull amplifier	YFW	Tubes / Valves	6	7th June 2007 10:18 PM
EL84/6BQ5 Ultra-Linear Push-Pull Recomendation	diegot	Tubes / Valves	17	5th April 2007 11:41 AM
modify parallel push-pull EL84 to single push-pull	chungtat	Tubes / Valves	12	3rd November 2005 11:25 PM
Double EL84 Push-Pull supply	Elkaid	Tubes / Valves	44	30th January 2004 06:26 PM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read

3rd April 2010, 05:08 AM	#2
torrence diyAudio Member Join Date: Feb 2008 Location: Oregon	Hi Eric, I presume when you say 4 x EL84 you are planning to operate in parallel push-pull (PPP). Roughly speaking, the power supply will need to deliver twice the current, so indeed your transformer should be able to handle twice the load, but the voltage drop across your resistors in the CRC network will also be twice as much. The best way to think about the design is that you want two power supplies in parallel to give the same performance as what you have now driving two tubes. This would mean double the capacitance and half the resistance in the CRC network. In truth, this may be neither practical nor necessary. Try out PSUD to compare the ripple and voltage sag under load for different configurations. Note also that your output transformer will need to have half the input impedance to have similar performance to your current design. -Eric

3rd April 2010, 05:30 AM	#3
bob91343 diyAudio Member Join Date: Mar 2010	If the tubes are running class A you probably won't have to upgrade the driver stage. But if you are running AB or B, you will need a stronger driver. Otherwise the grid current will overload it. When paralleling tubes it's wise to include parasitic suppressors in the grid and plate leads, say 100 Ohms for a wild guess.

5th April 2010, 06:21 PM	#10
Brit01 diyAudio Member Join Date: Jul 2008	Nice soldering work there Eric. Very neat. I need to take this approach.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)