Building EL-84 push/pull amp out of Zenith Radio Parts

[DRH1958]

Yes you are correct. The lower the value of the bypass cap, as you say .68uF is lower than 2.2uF, the less bass response the stage will have. The formula is 1 divided by the quantity 6.28 X C X R where C is in Farads and R is in ohms. So in the fromula 1uF is entered as .000001. Since the cap value is in the denominator of an inverse function, the larger cap will have a lower frequency cutoff point. Or if you want to do it he easy way, use this: Cathode Bypass Capacitor Calculator
Yes, using that .047 cap before the tone stack is just a personal preference. It's just the way I got used to doing it. Not really necessary. I had a bunch of 200 v Black Beauties laying around and figured if I block the B+ from the stage before the stack, I can use these.

[DrGonz78]

Okay I have done two different schematic edits: one with cathode follower stage and the other with parallel gain stage. I have even begun learning DIY Layout Creator to map out the physical layout of the components and that program is a very handy tool, thanks to Bane on that one.

I have updated some supply voltages as I have decoupled the gain stages better(first 12ax7). I have a question regarding the 1meg bleeder caps I have drawn on the power stage for each filter cap. Do you think that is a good idea or what benefits am I to get including that in my design? Also, what wattage should those bleeder resistors be if I include them on the build?

Also, I have a question about the wattage rating of the dropping resistors on the power supply. I have updated my schematic to show 2 watt dropping resistors to the preamp stages, will that be sufficient since they do not draw such high currents?

I have moved the tone stack to just before the PI as recommended and added a bright cap switch to the gain stage. What sort differences will a cathode follower verses a parallel gain stage present? Just some food for thought thanks. Oh I added a couple of pics to show you what I am working on.

Edit: Also, note that I have an extra 6.3v power supply. Now I have thought about using two separate heater supplies, one for preamp and the other for power tubes. I have also thought about being able to create a voltage doubler and use this extra tap to make a fixed bias supply. I have also just thought about not using this tap at all. Any ideas there too?

[Chuck H]

The first thing I noticed was that the PI is missing a decoupling cap. An LTP needs a cap to separate any load, other than the grid load to the tail, from it's grid.

You aren't using any totem arrangements for decoupling/filter caps so the bleeder value is somewhat unimportant. But you really don't need to bleed every cap. A bleeder off the main cap is fine. The others will discharge through this same bleeder provided the standby switch is in "play" mode. If you really want to be sure about voltage on the caps being drained under ANY circumstances you could include a second bleeder on the second filter. 1M is pretty large compared to what is typically used. It'll get the job done more slowly than, say, 220k, which is a common value for this circuit. No appreciable current will pass through a 220k value here. So, maybe a 220k on the main filter and one on the filter on the opposite side of the standby switch. That'll do the job faster with fewer parts.

Simulating the tone stack values for the parallel arrangement, with the 100k slope resistor, looks pretty poor. Maybe change the slope value to 47k. Or even all the way down to the 33k value shown for the cathode follower schem. I don't think this is the best place for a parallel arrangement anyway. The first stage is where you would want it. Tonally there isn't much difference between a single or double triode so there isn't much reason to place it in the middle of the circuit. But in the first gain stage the parallel arrangement is said to reduce noise. I might just go with the cathode follower anyway.

JM2C

EDIT: It also occurred to me that your zeners are working very hard. I have a string of three 13V zeners in a similar amp right now and their doing fine, but they do get hot. I don't remember my numbers but I remember thinking they would be ok. My numbers may be off but I think with the 15V zeners you'll be right at 5W dissipation for each 5W zener. You should probably go to the 10V units and string five or six instead of the four 15V units. Or put the string on the HV wind CT so you can use the resistor shunt to give the zeners relief. That trick still works with a 0V reference even though things got funny on me in line with the rail.

[Dave H]

That's a lot of gain you've got there Dr. I make it 50 x 50 x 50 x 0.1 x 20 = 250,000 to the 8 ohm tap. The 8 ohm tap will clip at about 12V rms so that's an input sensitivity of about 50uV rms. I don't think I could make an amp like that without excessive noise and oscillation. I know nothing about high gain amps but don't they usually have interstage dividers to keep the gain build up under control?

[Chuck H]

+1. Lotsa gain. You'll probably want to at least include a spot for a divider between stages two and three. You don't have to implement it in the first go, just make sure there's accommodations for one so you won't need to cobble it in on a board with no spot for it. IMHE if the layout is right you won't need to do much attenuating, but it's very likely you will need to do some.

And I just noticed that you're also missing the grid load for the second 12ax7 stage.

[eschertron]

+1 to all above. I like the looks of the first schem, more "Marshally", and as Chuck says, the CF will help the tone stack out.

The comment I'd like to make is I've been playing with a 2xEL84 amp design, started with a cathodyne PI. The cathodyne was 'just right' following a high gain distortion-inducing preamp, allowing the EL84s to add their color to the sound (but not too much). I redesigned with a LTP (because I added a tone stack in front), and now I have way too much gain on tap. I'm rethinking the design (again!). I mention this simply because you may run into the same thing.

Adding the grid load in front of the second[sic] third stage presents an opportunity to control gain or EQ at that point.

[olddawg]

Great advice, this is how I do it, zig-zag and leave an extra set of lugs in case I want an extra diode. In addition to elevating them is also leaving the leads of the diode long as this also adds a small amount of heat dissipation.
In response to whether there is much sound difference, I DO know there is quite a difference between SS, 5AR4, 5U4, and 5Y3 in voltage sag. This is not debatable. It's simple ohm's law. More resistance, more voltage drop. Whether the sound changes I suppose is in some debate. It seems unavoidable to some extent since the voltage change on the plates gives tubes a different sound. I suppose it comes down to how fast the rectifier can recover, so filter cap values will play a role too.

EDIT: I suppose I should have said "how fast the power supply can recover" above. Very interesting thread. Near and dear to me because EL84s are my favorite tube and even though sometimes reasoning fails us, it still is thought provoking and that is always good!

Well... It's debatable if you use a sag resistor.

[DRH1958]

Well... It's debatable if you use a sag resistor.

How is using a sag resistor different? It's still series resistance with the B+ rail. Either series resistance from the tube or a resistor, shouldn't make any difference. I guess I should ask what is debatable, the voltage drop or the sound change as the result of the voltage drop? The drop is not debatable, the change in sound is but I already acknowledged that in the previous post. In fact if I use a 5Y3 I get 25 volts B+ less than a 5U4.

[DrGonz78]

So I added grid load between 2nd and 3rd stage and put a 470k in place. I think that is what you all were saying? Also, I put the coupling cap between the master vol. and the PI, pretty sure that is what you meant too? I was not totally set on the 15v zener's and I have the 10v ready to go. Should I drill some holes in copper(older pennies) for some sort of heat sink ability? I am not going to populate this thing until I have really looked all the factors. You guy's are pretty much my filter check of experience here and I can't thank ya'll enough!

Possibly I should go with the Cathodyne PI instead? Either way I will draw that up to better understand all the various ways this circuit could be built. Thanks for that tip Eschertron!

Oh I just found this schematic last night too. G40v pignose where it uses a pair of 6L6's. However, it did help me implement the cathode follower and amazingly it is very similar to the layout of my circuit. Some of it I am now using as some form of reference and after reading all your comments. I now see direct relationships between the pignose and my amp. I don't wish to just copy the thing really, but it can't hurt to see a somewhat similar circuit for reference.

[eschertron]

Oh I just found this schematic last night too. G40v pignose where it uses a pair of 6L6's. However, it did help me implement the cathode follower and amazingly it is very similar to the layout of my circuit. Some of it I am now using as some form of reference and after reading all your comments. I now see direct relationships between the pignose and my amp. I don't wish to just copy the thing really, but it can't hurt to see a somewhat similar circuit for reference.

Good one! Seeing another amp that has functional engineering similarities is a good sign! Keep in mind that 6L6s need a whole lot more drive than EL84's, so you'll be topped out (from the PI) long before the 6L6s in the pignose would be starting to distort. And notice the pignose drops about 6dB (max signal) with the divider between stage one and two. That's just a caution that you might discover some excess gain has to be gotten rid of during the testing phase.

[DRH1958]

Cathodyne PIs work very well with smaller input signal types like EL84s and 6V6s. I re-build Hammond organ and old PAs and all of them have the cathodyne PI and sound great and have excellent overdrive sounds. One of the more well known amps, the Peavey Classic 30 uses one. The thing about the cathodyne is that you only use one triode and it operates at a slight loss, but have the other triode to put in front of it for a driver. If you have too much gain, just re-do the driver stage for less gain. Adjust to taste.

[Chuck H]

Is that extra dot left of the third stage grid load just a blob (?) or place to add a resistor to make a divider should you choose to? Because that would be exactly what I was referring to above.

Looks good to me. The only thing I'll add is that if you look at the G40V schem you'll see that the tone stack is voiced hard into the midrange. I expect that is to get more useful grind under overdrive conditions. Your design is voiced broader and may sound more "Fendery" when overdriven. But it will certainly do a better clean jazz tone. Small taters if you ever wanted to revoice it.

Don't sweat having a design similar to others. Especially when you "discovered" the design post facto. My signature design is somewhat (to a lot) similar a TrainWreck circuit. I designed it about five years before there was any TW circuit information on line. After seeing the TW design I decided to try the cold clipping third stage. I liked it. I'm using a different value, but the inclusion of that circuit obviously made my design THAT much closer to the TW topology. I have no doubts that anyone familiar with the TW circuits would assume that this is "my take" on it. IT'S MY ORIGINAL DESIGN!.. But I digress... My point is... There's not much new under the sun WRT tube amps. How many times has the same song been written, the same story told or the same fashion created? Fuggit about et.

[Chuck H]

Cathodyne PIs work very well with smaller input signal types like EL84s and 6V6s. I re-build Hammond organ and old PAs and all of them have the cathodyne PI and sound great and have excellent overdrive sounds. One of the more well known amps, the Peavey Classic 30 uses one. The thing about the cathodyne is that you only use one triode and it operates at a slight loss, but have the other triode to put in front of it for a driver. If you have too much gain, just re-do the driver stage for less gain. Adjust to taste.

I use an LTP with my own fav el84 design, but I'm working on a new thing with a cathodyne (should "cathodyne" be capitalized since it's non descriptive designator and a moniker?). Some amp designs do very well driving the piss outta el84's My new el84 design is going to be more of an uber gain preamp with very little over driving of the power tubes so I figure I can spare a gain stage at the PI. More in line with your ideology maybe. With this topology and intention it's even possible to do some tone shaping with the stage feeding the cathodyne and still have enough drive to the power tubes I love el84's. A very elastic tube to design with.

[The Dude]

I use a VKC with my favorite el84 design.

[Chuck H]

I use a VKC with my favorite el84 design.

Very Kool Cathodyne? Variable Kathode Current? Vermicious Knid Critter? (Willy Wonka anyone?)

Now you're just messin' with us