EL34 schematic confusion

[navin]

I reduced the gain of the amp (it's too high for modern sources) by altering the feedback resistor. To keep feedback constant, I also changed the first tube to 6SL7 (actually, a 6SU7 or 5691). The 6SN7 has matched 33K plate resistors and a cathode CCS. It also has a protection diode from grid to cathode. Input stage and phase splitter have solid state regulation, a variant of the ubiquitous Maida circuit...

SY, we would love to see this new modified ckt. also given today's soruces do we need the AF tube at all? What about using 1/2 ECC88 for the AF tube?

 

[SY]

An ECC88 might work. I haven't tried it. A 12AT7 does a great job when medium voltage swings (30-40V) are needed.

My CCS was posted while you were writing .

 

[JojoD818]

SY,

so i basically replace the 18K cathode resistor of the 6sn7s with your CCS and then I can use matched 33K resistors yes?

can i use 5mm (jumbo) red LEDS? i have a bunch of these.

the hf87 schematic shows 300V on the plates of the 6sn7 so this would be my target voltage when i adjust the trimpot right?

your circuits trully is quite easy to implement. thanks!

JojoD

 

[SY]

SY,

so i basically replace the 18K cathode resistor of the 6sn7s with your CCS and then I can use matched 33K resistors yes?

can i use 5mm (jumbo) red LEDS? i have a bunch of these.

the hf87 schematic shows 300V on the plates of the 6sn7 so this would be my target voltage when i adjust the trimpot right?

Yes.
Yes, but check that they have roughly a 1.7-1.8V drop..
Yes, but don't hesitate to experiment...

 

[JojoD818]

SY,

The 80K is connected to the main B+ meaning the center tap of the opt tranny.

Thanks, when I re-write the schematic that I would use in building this amp I'll include your ccs already if you don't mind. I'm really getting excited on this project.

If I find another quad of EL34s then that's what I'll use but TDSL data on 7027 says it can also be used with my 6.5K opt trannies. yahoo.



Sincere thanks to you for answering a lot of my questions.

JojoD

 

[SY]

Yes, the 80K can go right to the OPT primary center tap or anywhere else you think you've got clean 400-450V. The CCS design is not original to me (it goes back to Ike Eisenson, at least), so it's yours to use, too.

I think the 7027 is a better tube in this circuit.

 

[JojoD818]

Still, thank you for bringing it to my attention. I have the transistors in my spare box so I decided to implement the ccs and learn from it.

I'll have to test the 7027 tubes first since I've never tested nor used them, but it was given to me used but the RCA markings still visible. Hope they are still good though.

 

[SY]

The thing to be careful about is grid current. Some 7027s are pretty poor in this respect, others are just fine. You want to avoid more than 150K or so as grid leak resistor. That means making the coupling caps bigger or compromising bass extension somewhat (I chose the latter, since I biamp).

 

[navin]

The thing to be careful about is grid current. Some 7027s are pretty poor in this respect, others are just fine. You want to avoid more than 150K or so as grid leak resistor. That means making the coupling caps bigger or compromising bass extension somewhat (I chose the latter, since I biamp).

Is the ckt available somewhere?

 

[SY]

The recommendation was sort of general- whichever circuit you end up building, you want a 7027's DC resistance at the grid to be on the low side. In the case of an HF-87 converted to fixed bias, the grid resistors are lifted from ground and returned to a variable negative supply.

In the rebuild I'm doing now, source followers will be added to take me into AB2. This also neatly disposes of the grid current issue. When I'm all done this amp, I'll post the circuits.

 

[navin]

...source followers will be added to take me into AB2. This also neatly disposes of the grid current issue. When I'm all done this amp, I'll post the circuits.

thanks. I guess we'll have to watch this space then

 

[JojoD818]

SY,

I've re-written the CCS you provided...

and the fixed bias HF87...

http://img.photobucket.com/albums/v70/jojod818/Dump/EICOHF87FixedBias.gif

I used 100K as grid leak resistors and a 10K balance pot. Would this be ok?

I'll be working on the psu side when I finish household chores.

 

[jane]

http://img.photobucket.com/albums/v70/jojod818/Dump/EICOHF87FixedBias.gif

I used 100K as grid leak resistors and a 10K balance pot. Would this be ok?

I'll be working on the psu side when I finish household chores.

Is that 10k pot for DC-balance? If so, it's not OK.

 

[SY]

No, that 10K pot will only upset AC balance and do nothing for DC balance. When I'm near a scanner later, I'll post one of my crude sketches of a proper DC balance circuit, unless some kind soul does it first.

 

[JojoD818]

will be waiting for your schematics SY. in the meantime, wouldn't it work the way I drew it?

 

[SY]

No. The grids draw essentially no current under normal use. So what's the voltage drop across a 100K grid resistor? Zero. What's the drop across a 110K grid resistor? Yup, zero again. You haven't changed the bias voltage applied to the grids just by unbalancing the grid resistors. All you do is increase distortion as you move away from balance (remember, at AC, those grid resistors are part of the plate load of the phase splitter).

 

[navin]

will be waiting for your schematics SY. in the meantime, wouldn't it work the way I drew it?

to bias a tube one must have it connected to gnd or +ve via a resistor so that one can control the current to the tube. right?

 

[JojoD818]

No. The grids draw essentially no current under normal use. So what's the voltage drop across a 100K grid resistor? Zero. What's the drop across a 110K grid resistor? Yup, zero again. You haven't changed the bias voltage applied to the grids just by unbalancing the grid resistors. All you do is increase distortion as you move away from balance (remember, at AC, those grid resistors are part of the plate load of the phase splitter).

SY,

I would include the bias adjustment pot in the psu that I would draw, the 10K pot is just there for balance and not for bias adjustment.

It is true regarding the distortion, I tried it this morning on my other amp.

Anyway, if I apply a sufficient negative voltage on the center of the 10K pot, enough to bias the el34 for 50mA each, wouldn't that circuit work?

 

[JojoD818]

navin,

the cathodes of the EL34 in the circuit are tied to ground via 10 ohms resistor for each opt tubes. That way I can measure the voltage drop across the 10 ohms resistor and compute for the plate+screen current.

 

[JojoD818]

the psu and bias supply. i have not done any calculations yet on the bias resistor network and pot but as SY said the target -bias should be adjustable. i plan to have a range of -25V to -40V. so those values at the bias supply are just there for graphical reasons.

http://img.photobucket.com/albums/v70/jojod818/Dump/EICOHF87FixedBiasPSU.gif

i will get another tranny that will supply heaters to the 7025 and 6sn7 and at the same time have a separate winding for the negative bias supply. that way my power tranny will only supply the HT and the heaters for the four power tubes. would there be any problems with such a setup?