5 watt build

[Matty_V]

I think its time to spring my 5 watter into action. I posted the schematic, just threw some stuff down. At the time there were others coming up with 5w ideas so I borrowed bits and pieces here and there. I'd appreciate any advice, error corrections, neat ideas to try. Also what voltage ranges I should be targeting for so I know what b+rail dropping resistors to use. I haven't gotten to the level of using tube charts to design stages yet. Thanks!

[HotBluePlates]

Well, if you're going to use Doug's 125P1B, then according to his transformer information, the output voltage is 640v CT, which means 320-0-320v. That implies 320v * 1.414 = 452v peak as a possible no-load voltage, even more if your wall voltage is high. I'd like to see 500v or more for a filter cap rating (or at least I wouldn't choose to use a 450v cap). I know Fender got away with 450v caps, but my preference is conservative rating.

You can probably be more lax with nodes C and D, but that's your choice.

I can tell you that you will probably not want to take the 6V6 plate feed from node A if you have a speaker with decent bass response (i.e., something bigger than an 8" speaker). In a SE amp, the output tube plate does not get the benefit of hum cancellation at the OT, so if your speaker will reach down to 120Hz, you will want an extra stage of filtering prior to that plate feed.

According to some info I found on the net, the 125C3A is rated for 50mA. This is marginal for a choke that is also to carrying the plate current. Therefore, you may want to use the 125C1A (rated 90mA), which Doug sells at the same price. Or, you could leave your supply as-is, and build out to the "left" of node A with, say, a 100 ohm 3w resistor (or bigger in a 5w rating).

So either way, you will want 5 stages of filtering in your amp if you have a speaker with bass, or you will be asking us how to kill hum (which will only be done when you add that 5th stage of filtering). You will then take the second filter cap as your new "node A." Since you probably don't want to buy 2 chokes, a resistor is more easily added between the plate and screen node, and plan on ~1k, and move the choke earlier to carry the plate current. The reason is you'd like the series filter components that carry big current (those feeding the plate) to have as low impedance as possible above the frequency you're trying to filter out.

The chokes quoted are both 4H, so they look like ~3k to 120Hz. A 1k resistor between plate and screen will likely gain you enough filtering while not knocking down your B+ too much. It's fine for the screen to be roughly equal to the plate voltage.

The 68k grid stopper on the 6V6 seems excessive.

My gut reaction is that the preamp design is not a good idea. Specifically, I'd question the layout implied by running the signal through 1/2 a 12AX7, out to a pentode, then back through the other half of the original 12AX7. That seems like a sure recipe for oscillation, because input and output wiring will almost certainly cross over each other, with lots of gain in between. Was this a proven method from someone else's amp, or a means of chopping down a bigger design (with an extra channel or effects) into a 2-preamp tube package?

I wouldn't want to speculate on good B+ values for this circuit, given that likely problem. If you just have to have triode-pentode-triode, I'd suggest accepting 3 preamp tube sockets and use a 7-pin (maybe 6AT6) for the cathode follower, or use a 9-pin and omit a triode. Or maybe you use that extra triode for something like trem. The input tube position could stay a 9-pin, possibly with a paralleled triode.

[Matty_V]

Wow. Ok, I'll need to read that over a few more times to soak all that in, but thanks for taking the time to help me out. I can't remember whose circuits I was looking at, but they had a similar triode-pentode-follower setup. Some of the values I straight up robbed. My other option, like you mentioned, would be to parallel the first stage and lose the follower. Would that give me enough bark (subjective I know) with the tone stack losses? I believe thats the 5f2 stack, or maybe I'm thinking of something else.

I'll prototype it both ways when the time comes. Again thanks for all the sweet info.

[Matty_V]

And by lax on nodes c & d...do you mean I don't need a 500v rating or I don't need 20uF of filtering?

[tubenit]

My gut reaction is that the preamp design is not a good idea. Specifically, I'd question the layout implied by running the signal through 1/2 a 12AX7, out to a pentode, then back through the other half of the original 12AX7. That seems like a sure recipe for oscillation, because input and output wiring will almost certainly cross over each other, with lots of gain in between. Was this a proven method from someone else's amp,

I was afraid of it doing the same thing. However, when I tried it on my Tweed Overdrive 2CF ....... there were NO issues at all. No oscillation and the amp is dead quiet at idle. However I did this on the overdrive section and not specifically the V1 tube.  I actually mounted a terminal strip on top of a layout board at first and played it that way for a month and then redid the layout board with the V2b being the cathode follower for the 5879.  Both versions worked fine.

If you look at how I drew up the layout, there really isn't much in the way of criss crossing wires.

Matty_V,  

Only thing I can say other than following HBP's guidelines is to consider a .01 coming off the V1-1 plate instead of .02.

With respect, Tubenit

[Boots Deville]

The switchable bypass capacitor values on V1A will give you 4.7uF, 6.9uF, or 9.4uF.  I don't think you'll hear a big difference between those three.  I would try something like maybe .22uf, 1uF and (4.7uF, 10uF or 22uF).  Tubenit does this in a lot of his designs. Look at those for some ideas.

[Matty_V]

Dang it, I was using the capacitors in series formula (c1*c2)/(c1+c2) for that switch, good catch Boots! I'll change up those values.

Thanks Tube, I want to use the follower. I'll try to lay mine out in a similar fashion, if this is the road I end up going down.

[HotBluePlates]

Hey Tubenit!

Well, you have a proven build, so that might nullify my comments. Just don't build it in a production amp, because Murphy will ensure to have to tinker every other amp. It can work of the socket placement is pretty tight and there is care in running the wiring; however, it wouldn't be a good approach to take often.

Matty:
Were you looking to take the overdirve section of Tubenit's Tweed Overdrive 2CF amp as your amp's 1 channel? If so, you're missing the portion that is the clean channel, which then cascades into the OD section.

And by lax on nodes C and D, I meant you could use a 450v rating. Yes, if you yank all the tubes, the B+ will sit slightly above the rating. However, the current will also be limited somewhat by the dropping resistors in the B+ supply.

[Matty_V]

I came up with the basic skeleton myself, I just used tubenits and someone elses schematics to get an idea of what components to use around the 5879 and some other things. I've never done a pentode in the preamp stage before.

I'm wanting to try out the morph control. Sounds neat. Plus the word morph is intriguing.

[tubeswell]

I'm wanting to try out the morph control. Sounds neat. Plus the word morph is intriguing.

I used this on a couple of amps now and you can get some cool and useful tonal effects depending on how you configure the pentode screen voltage and plate load.

If you are using a DC-coupled CF stage, you can take the morph control pot's input-side cap* directly to the CF cathode. This reduces the AC load on the pentode's plate, and you still get the benefits of the morph control only it is now being driven by the CF with resulting increased BW.

I have found by trial and error, that one thing you need to watch if you use the DC-coupled follower with the pentode (when you run the pentode the way that Merlin does in his book), is that the pentode's screen resistor might need to be higher resistance than you think, because the DC-coupled CF steals more current through the pentode's plate resistor, which drags the pentode's plate voltage down close to the screen voltage, which can cause undesirable noises. I found this created an issue when I first used the DC-coupled follower with the EF86 stage values in Merlin's book, and I managed to stop it by upping the screen resistor a bit to further decrease the screen voltage/reduce screen current. JM2CW

* the one that goes to the pentode plate in Merlin's circuit

[Matty_V]

Tubeswell, it was your schematic I "borrowed" the morph control from . So its now drawn up just like yours. I'll keep the screen resistor in mind if I hear any undesirable noises. According to my girlfriend I should be checking my screen resistor every time I plug in and jam. 

[tubeswell]

FWIW schematic attached for how I implemented the morph control from the CF cathode (what I was talking about in my earlier post). This has the adjusted EF86 screen resistor value and voltages (but you'd want to do a horses for courses thing with that other pentode you've got in your schematic).

[Matty_V]

Tubeswell, would you recommend the 12dw7? Never tried one. What difference in tone does it give you over a 12ax7?

[tubeswell]

A 12DW7 is simply a 12AX7 triode on one side, and a 12AU7 triode on the other side. I used the 12AU7 triode for the CF with a (2W) 47k Rk  (but you could also use a 22k Rk) and got good bandwidth at a lowish cathode voltage, and still plenty of current to drive the TS. I used the 12AX7 triode part of it as a low-gain tone stack recovery stage analogous to the 6G3 - probably could've just used a 12AU7 tube for the whole thing, but the 12DW7 gives a bit of flexibility in that the 12AX7 1/2 of it can be tweaked a bit in terms for more gain from the recovery stage - not that this amp really needs it.

[sluckey]

You used the higher gain half of the 12DW7 for the TS post amp, but split the plate load and are only sending roughly 15% of the plate signal to the PI. I bet using a 12AU7 would give very similar results if you sent all the plate signal to the PI. I don't know if the tone would be any different though.

[Matty_V]

I've been busy with other things but I'm back to the 5 watter again. I revised the schem the best I could according to the suggestions made and drew up a layout. I'd appreciate any further comments or suggestions. Thanks guys.

[jojokeo]

Matty,
Instead of possible problematic issues discussed earlier and using a separate pentode and separate triode you could use something like a 6AW8a (sharp cutoff pentode/high mu triode) or a 6BL8/ECF80 (sharp cutoff pentode/med mu triode) where they're both in the same bottle. You'll have an extra side of your 12ax7 too if you decide to use it, w/ the same amount of tubes also.

[tubeswell]

Hi Matty

In that last schematic you posted, you have the 470k screen supply resistor (for the 5879) taken from the plate (when you should instead take it from the HT). If you take the screen supply from the plate, it will be in 'triode mode' the whole time. 2CW

[Matty_V]

That tube sounds neat. I'd consider using it if I hadn't already picked up some 5879s for this project. Hmmm. I may ponder this more. I could just lose the follower section and perhaps have V1 paralleled on a switch. Would this work well without a follower?

That screen resistor is a drawing error. Thanks tubeswell.

[jojokeo]

That tube sounds neat. I'd consider using it if I hadn't already picked up some 5879s for this project. Hmmm. I may ponder this more. I could just lose the follower section and perhaps have V1 paralleled on a switch. Would this work well without a follower?

Yes, but it's all up to what you desire regarding gain amounts wanted. You'd be surprised what you can get w/ only a couple gain stages set up right and also where the placement & type of tone stack that is used. In a bandmate's BF Deluxe Reverb, the non-trem side has one triode less gain. He also rarely if ever used that side. While I had the amp opened up (installing a switchable presence/NFB control in place of the ground switch) I simply re-wired that channel so that the two triodes cascaded w/ the tone stack placed afterwards (and some better biasing & voicing adjustments too). The gain & overdrive change made a huge difference. That side is "hot wired" now and he uses it for solos or in songs that apply for overdrive. I even see him using whenever he's practicing at home and we're going over songs in lower vol settings. Those old BDDR's sound incredible when driven and as crazy as it seems to have a Vintage 30 speaker in there - it's one of the best amps I've ever heard.
Another case in point - I have an ef86 > CF > el84 amp where I've made the CF stage bypassable. I rather like playing the amp that way much of the time. It's fun getting the overdrive from the CF stage but when it's not used, I get a more-pure tone with still a lot of drive w/ my other settings and controls. You never quite know though when you're playing/making from scratch so then switches and gain pots are your friend.

[DummyLoad]

6AW8a

gm seems too high to be of any use in audio.

--DL

[jojokeo]

The triode section is basically the same as a 12at7 but w/ a gain of 70 (like a 5751) instead of 60 and that's of "audio use"

[DummyLoad]

sorry, i should have been more explicit - i was referring to the pentode.

--DL

[Matty_V]

Ok I came up with another option. I moved the follower section after the first gain stage, then into the amz presence tone stack. I had this setup for a little while on my 18w but decided to go with a tmb stack instead.

I marked up the .47uF cap for the morph control in red. In the previous version I had it going to the cathode of the follower section. Now that I've moved it to the other side of the 5879, can it still attach there? Not sure exactly how the thing works.

Also Jojo, you said you had made your CF section bypassable. Do you have a schematic of that? Just toying with the idea.

[jojokeo]

This is one way to do it but there are many ways. I didn't design it to flip this switch back and forth while playing, just to have as an option. You don't need the voltage divider either but I had a lot of gain to dump at the time.

[jojokeo]

sorry, i should have been more explicit - i was referring to the pentode.

Here's the tube application from a site: "Miniature type used in television receiver applications. Pentode unit is used as video amplifier, if amplifier, agc amplifier or reactance tube. Triode unit is used in low-frequency oscillator, sync-separator, sync-clipper, and phase-splitter circuits."

If it's suggested that it could be used as an if amplifier I guess I don't understand what you're saying & why regarding Gm? I think I got this tube from a suggested use from Merlin's book a while back.