• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Fun with junk EL84 SE

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi all,

I have some spare time these days, so I dug out this beat up EL84 SE amp I found on ebay for next to nothing a while back. I want to use the transformers and build something for fun. As I don't really "get" design, I decided to just copy the circuit and replicate it as is and see what it sounds like. I have a couple of questions I wanted to ask:

1. The second grid of the EL84 (pin 9) is connected straight to the HT supply - no resistor in-between. Does that make sense to you? I have looked at many EL84-based designs on the web and have never seen this...
2. Thinking about adding a switch to allow me to run the EL84s as triodes. Would this be as simple as taking the second grid off the supply and connecting it to the relevant node of the output transformer, or do I need to think about it more carefully?
3. The heaters for the valves (2 EL84s, one ECC83 and the rectifier) are run directly from the transformer. Do you think I should change this to some simple DC or even regulated supply, or not worth the fuss?

Thank you!
Nikos
 
1. The second grid of the EL84 (pin 9) is connected straight to the HT supply - no resistor in-between. Does that make sense to you? I have looked at many EL84-based designs on the web and have never seen this...

Yes, my PP EL84 amp had B+ connected to the screens, approx. 325v IIRC.

2. Thinking about adding a switch to allow me to run the EL84s as triodes. Would this be as simple as taking the second grid off the supply and connecting it to the relevant node of the output transformer, or do I need to think about it more carefully?

Connect screen to plate (anode) with a 100-470R, 1-2W resistor. Simple as that.

3. The heaters for the valves (2 EL84s, one ECC83 and the rectifier) are run directly from the transformer. Do you think I should change this to some simple DC or even regulated supply, or not worth the fuss?

Try straight off the heater winding first, and see how noisy it is. Are there any hum pots?

jeff
 
Administrator
Joined 2004
Paid Member
This sounds quite similar to some of the Magnavox SE amplifiers over here. (My first tube amplifier as a teenager was one of these.)

Can you trace out the design and post a schematic?

Filamentary hum if an issue can be resolved by raising the filament supply above ground either using a simple resistive divider and cap or the cathode bias from the output stage. (It's free and I've used it successfully in the past - just make sure that there is not much if any audio present on the bypass cap)

Supply filtering is going to be more important, replacing the rectifiers with modern devices and improving the ripple filtering of the supply will help improve the performance.

I suspect there is significant global feedback since the output stage in this amplifier is pentode connected. Yes, in many of these bare bones designs the screen resistor was omitted - supply voltages tended to be low enough that you could get away with this. May help stability to add one individually on each screen.

I'm not sure with the given output transformers whether triode operation is going to be that thrilling or not, but worth a try. You can use a spdt toggle switch rated for 250Vrms to switch between pentode and triode mode. Many will advise not to switch with the amplifier on, and I'd agreed while admitting to have done the same and gotten away with it countless times with no issues whatsoever. (Luck maybe.. caveat emptor.. :D)

Providing individual cathode resistors and capacitors for each EL84 if not present would be a nice enhancement. Double the resistor value and roughly halve the bypass cap value - or keep it the same value if the output transformer is reasonably large and can handle LF OK..

These sorts of amps are a lot of fun to play with, simple but offering surprisingly good results in some instances with relatively minor tweaking. (And the right speakers of course.)
 
Can you trace out the design and post a schematic?

I sure can, here you go. Let me know if it's the one you are thinking.

Take the point about supply filtering and I was initially inclined to use a bridge, but I've never owned or even heard anything with a valve rectifier so I thought I'd try it out for fun...may well change it later on.

Parts have been ordered and are on their way, will keep you posted.
 

Attachments

  • el84.jpg
    el84.jpg
    311.6 KB · Views: 310
Administrator
Joined 2004
Paid Member
Wildly different than the maggot box, err Magnavox amps I have played with.. :D

Something closer to this..
Note that lag compensation on the plate of the 12AX7A/ECC83 will probably be necessary for stability with the feedback loop. I didn't bother to show it as it would need to be empirically determined for the specific OPTs used. Speaker is assumed to be 8 ohm, and an "excellent" transformer was assumed. (3.2K:8 for triode with 250 ohm cathode resistor, 5K primary for pentode) This is intended in the spirit of the Magnavox if not the exact implementation.. :D

R10 can be varied over a range of 470 - 1K, open loop gain should be in the vicinity of 40dB. I would probably choose something between 470 - 680 ohms providing some margin.
 

Attachments

  • 6BQ5_SE_AMP.PNG
    6BQ5_SE_AMP.PNG
    36 KB · Views: 253
Administrator
Joined 2004
Paid Member
Yeah the EL84 symbol in that schematic is wonky, but then so is the actual design.. Grid leak bias is an interesting twist in the driver stage, not good for linearity at the sorts of signal levels encountered in a power amplifier.

Suddenly I have a hankering to design/build a very simple EL84 amp, I guess the GM70 isn't enough.. :p

Mine would have UL and fixed bias.. I'll post something soon..
 
Oh, yes, this is very wrong! Sorry, was late when I was copying this from my sloppy hand-drawn sketch - left some things in and missed some out! :xeye:

Here's the corrected one with a clearer pentode symbol also. The design does have fixed bias - the connection was one I missed when copying!
 

Attachments

  • el84.jpg
    el84.jpg
    315.3 KB · Views: 231
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.