Somewhat tweaked 5E3 layout - request for comments.

[Wombaticus]

I agree, the layout looks fine and I think Fyl et al. are overthinking this. It's a low-gain amp, so ground loops, heater ripple etc. aren't mega-important. The main thing is where you return the PT CT to, and you got that right.

The other important thing is the input jack ground, but you haven't shown that in your layout at all. Douglas Self, in a hi-fi context, recommends grounding the input jack directly to the chassis and grounding the preamp bus to that. The idea is that if ground loop current, or RF pickup on the guitar lead's screen, comes at you from outside, it can't get inside the chassis.

For the cathode resistor, you really want to keep it away from the electrolytic caps. How about one of those aluminum housed Dale resistors bolted to the chassis?

Dale / Vishay - RH01010R00FE02 - Resistors - Passive Components - Allied Electronics

Also whenever using a MOSFET, RG and I recommend a 12V zener diode between gate and source (diode's cathode to gate). It helps avoid mysterious MOSFET deaths by ensuring the gate voltage can never get excessive. Some power MOSFETs have it built in. The other vital thing is a gate stopper resistor, but you remembered that.

It may be me that's overthinking it, not Fyl, as I'm explicitly trying to make sure that there are NO ground loops. If you see one implied by the layout, kindly point it out. That's what I was after with this thread in the first place.

I didn't draw in the inputs. But I always use insulated jacks, but then connect both the input grounds and the preamp circuit ground close to the input jacks using a very short low impedance run. So I think I'm good there.

I like those aluminum housed resistors! I've used those before in conjunction with can caps with an untapped reservoir cap running to the plate cap, it works out really well. This could certainly be another place to try it.

For the FET I specced the STP7NK80ZFP:

STP7NK80ZFP STMicroelectronics MOSFETs

I think it'll be up to the job. It's a 800V 30W part in a fully insulated TO-220FP package, and includes both gate-source and drain-source protection: http://www.st.com/stonline/books/pdf/docs/8979.pdf. Seems like a nice part, although I normally would have got the STF3NK80Z, which is cheaper, but Mouser has it on backorder.

Thanks again for all your help!

[stokes]

Instead of using a voltage divider to elevate the heaters,just connect the 100 ohm resistors to the cathode of the power tube instead of ground,this will elevate it more than enough to keep the 60 cycle hum out of the signal path.I wouldnt recommend using that extra filter cap for the preamp.I tried it with a couple of 5E3's and wasnt happy with the result.Gave it a less than ideal bass response with some strange ghosting.I ended up removing the extra filter.

[fyl]

Instead of using a voltage divider to elevate the heaters,just connect the 100 ohm resistors to the cathode of the power tube instead of ground,this will elevate it more than enough to keep the 60 cycle hum out of the signal path.

Sure, but the 5E3 uses a cathodyne PI, with a high cathode to ground voltage. Some tubes - mainly Sovtek - really don't like it, and elevating the heaters by more than 30V is IMO a Good Thing™.

I wouldnt recommend using that extra filter cap for the preamp.I tried it with a couple of 5E3's and wasnt happy with the result.Gave it a less than ideal bass response with some strange ghosting.I ended up removing the extra filter.

Provided that the amp is properly defarted the additional cap is another Good Thing™: less noise, better touch control and a smoother response with a cleaner disto (is this an oxymoron or what?).

[tubeswell]

I I'm explicitly trying to make sure that there are NO ground loops.

Its not so much the fact there there may be ground loops that is important IMHO, but rather that there may be different micro-rises in potential in the ground return path(s), which bleed back into the signal path as 'backwobble'. This is where Merlin's article of grounding is useful. The method of keeping relevant parts of the amp grounded with the relevant decoupling/filter caps minimises the interaction between different points in the ground return path. For a 5E3, all that this need entail, is keeping the preamp decoupling/filter cap grounded at one point (the input jack grounds) with all the other pre-amp ground returns going, each by their own separate wire, to that same point - and then having all the other grounds, each going by their own separate wire, to another point - at the other end o the chassis somewhere like a PT bolt. I do all my amps using the principle, and they are quiet as a mouse.

[stokes]

Sure, but the 5E3 uses a cathodyne PI, with a high cathode to ground voltage. Some tubes - mainly Sovtek - really don't like it, and elevating the heaters by more than 30V is IMO a Good Thing™.



Provided that the amp is properly defarted the additional cap is another Good Thing™: less noise, better touch control and a smoother response with a cleaner disto (is this an oxymoron or what?).

Just what does the PI have to do with the elevated heater?20-30v's is more than enough to do the job.Adding extra 60hz hum at that stage where the voltage divider is cant be a good thing,just adds more for your caps to filter.You might think the extra decoupling stage in the preamp is a good thing,but its not in this circuit.Adding extra filtering in the preamp stages is not the same as adding more capacitance in the main supply,after a certain point it will degrade the tone.By the time the B+ gets to the preamp caps it is already ripple free,the purpose of the filters in the preamp are to de-couple the stages more than for filtering.Designers who knew way more than you and I decided that stage didnt need to be de-coupled,why mess with success.I tried it and it isnt beneficial.

[fyl]

Designers who knew way more than you and I decided that stage didnt need to be de-coupled,why mess with success.

They alo used two-pronged mains cables with no chassis ground and a death cap, why mess with success?

[stokes]

Oh yeah,the infamous "death cap".Ever heard of anybody actually dying from this?Or,in fact an actual occurance of anyone being hurt by it?There are actually many amps still working fine,a bit noisier than a 3 prong,but still working fine,I have a couple myself.Truth is,if this "death cap" did actrually short out,it is only putting 120 vac on the chassis,hardly enough to kill anyone but a small child.I'm not saying it isnt prudent to change to a 3 prong set-up,but it hardly calls for the terror of the "death cap" description.Just more internet hoodoo.And exactly what are you basing the "more filtering in the preamp" is a "Good Thing" that you claim a trade mark on?And,again,why do you think the PI has any bearing on elevated heater winding?Elevating the heater by more than 30vdc is unnecessary,just what do you base the "Good Thing" on?Extra components are "Good Thing"?You seem to be quick with the smart assed responses,like the "If you say so" response to Steve,who you should know,is actually an EE,but I havent seen anything of substance in any of your responses.

[fyl]

I havent seen anything of substance in any of your responses.

If you say so.

[stokes]

Like I said,its all you got

[Wombaticus]

Oh yeah,the infamous "death cap".Ever heard of anybody actually dying from this?Or,in fact an actual occurance of anyone being hurt by it?There are actually many amps still working fine,a bit noisier than a 3 prong,but still working fine,I have a couple myself.Truth is,if this "death cap" did actrually short out,it is only putting 120 vac on the chassis,hardly enough to kill anyone but a small child.I'm not saying it isnt prudent to change to a 3 prong set-up,but it hardly calls for the terror of the "death cap" description.Just more internet hoodoo.And exactly what are you basing the "more filtering in the preamp" is a "Good Thing" that you claim a trade mark on?And,again,why do you think the PI has any bearing on elevated heater winding?Elevating the heater by more than 30vdc is unnecessary,just what do you base the "Good Thing" on?Extra components are "Good Thing"?You seem to be quick with the smart assed responses,like the "If you say so" response to Steve,who you should know,is actually an EE,but I havent seen anything of substance in any of your responses.

I think what Fyl is driving at is that some 12AX7s, (particularly the Sovtek ones he mentions) don't have particularly robust cathode-heater insulation. This is, of course, most problematic in a circuit such as a cathode follower (or cathodyne to some extent) where the cathode is necessarily at quite a high potential relative to an unelevated heater. Anyway, it's a well-established cause of premature tube failure.

[Wombaticus]

Its not so much the fact there there may be ground loops that is important IMHO, but rather that there may be different micro-rises in potential in the ground return path(s), which bleed back into the signal path as 'backwobble'. This is where Merlin's article of grounding is useful. The method of keeping relevant parts of the amp grounded with the relevant decoupling/filter caps minimises the interaction between different points in the ground return path. For a 5E3, all that this need entail, is keeping the preamp decoupling/filter cap grounded at one point (the input jack grounds) with all the other pre-amp ground returns going, each by their own separate wire, to that same point - and then having all the other grounds, each going by their own separate wire, to another point - at the other end o the chassis somewhere like a PT bolt. I do all my amps using the principle, and they are quiet as a mouse.

If I've done that layout right (it's possible I haven't), I think you will find that it adheres very closely indeed to Merlin's recommendations. A closer reading of Merlin's book, or especially the excellent new article on his website will show that he specifically disavows methods involving multiple chassis attachment points, except in the instance of complex multi-channel amps where there is little alternative to this second-best method.

[Wombaticus]

..... I wouldnt recommend using that extra filter cap for the preamp.I tried it with a couple of 5E3's and wasnt happy with the result.Gave it a less than ideal bass response with some strange ghosting.I ended up removing the extra filter.

Hmmm, that's interesting. What specifically about the bass response did you find had changed? If anything, adding another filter and more decoupling here should enhance bass response... but maybe because of the silly 100nF coupling caps (which should already make the amp very bassy), the perceived additional bass wasn't welcome?

I can't for the life of me think of a reason, though, that additional filtering and decoupling could actually *cause* ghost notes. Any theories?

Anyway, I did mark them in that layout as "optional". For now I'm going to keep an open mind and find out for myself and A/B them... I'm perfectly prepared to haul them out if it turns out you're right.

[stokes]

Hmmm, that's interesting. What specifically about the bass response did you find had changed? If anything, adding another filter and more decoupling here should enhance bass response... but maybe because of the silly 100nF coupling caps (which should already make the amp very bassy), the perceived additional bass wasn't welcome?

I can't for the life of me think of a reason, though, that additional filtering and decoupling could actually *cause* ghost notes. Any theories?

Anyway, I did mark them in that layout as "optional". For now I'm going to keep an open mind and find out for myself and A/B them... I'm perfectly prepared to haul them out if it turns out you're right.

At the time,it was a good many years ago,I did have a discussion with Gerald Weber about it and he did give me some theory about excessive filtering and the extra decoupling in a preamp,I dont remember exactly what the theory was so I dont want to mis-quote him.What I got when I had the first preamp tube fed by its own supply cap was a lot of static associated with low frequencies as well as ghosting.I discussed troubleshooting this problem with Gerald and when he heard I had the extra stage he suggested removing it and it worked.As I said,he gave me his explanation of why extra filtering in a preamp is a bad thing,some of it had to do with the fact that at that point in the B+ rail your caps arent really filtering any ripple,they are mainly for de-coupling that stage.I was actually building 2 5E3 type amps,one with the normal 2X 6V6 and the other with a 4X 6V6 output and had the same issue with both.It was over ten years ago,so I dont remember all the details in his explanation,but I have always avoided more filtering in the preamp of any amp I've built.You can give it a try,of course,I am just sharing my experience with it.At best,its not really necessary,extra filtering at this stage is not really beneficial,at worst you will end up with the same problem I had.I am a big proponent of extra filtering,but only as far as the first and screen stages,in the pre it could be detrimental.As for the elevated heaters,my suggestion is purely to save you parts and room.The idea of an elevated heater is to have the "ground potential" of the heater "floating" above your signal ground,any more than 30 or so volts wont be any better.I am well aware of the problem some AX7's have in a cathode follower circuit,but it is irrelevant in this instance.I just wanted to hear what FYL was basing this on,but apparently its nothing.

[fyl]

I can't for the life of me think of a reason, though, that additional filtering and decoupling could actually *cause* ghost notes. Any theories?

None scientific whatsoever. But I guess that self-proclaimed amp gurus must have some very colorful explanations in stock.

Fact: ghost notes are caused by (1) intermodulation in inadequate power supplies and (2) compounded by poorly chosen coupling time constants.

(1) is taken care of by using a stiffer power supply, ie. lower impedance/more capacitance. No more 100/120 Hz from the supply modulating the signal, no more intermodulation and no more beat products. But too stiff a supply can lead to a somewhat too linear-sounding amp (good for hifi, not for MI).

(2) can be corrected by using the right cap values. You wrote that Fender used silly 100nF coupling caps and you're 100% right. Use 22n/47n or 22n/22n and the amp won't fart/intermodulate as much - there can still be some blocking in the cathodyne, which can be suppressed by adding a 100K or more stopper resistor between the 1M and V2b's grid.

[tubeswell]

As I said, I've done all my amps using this method and they are quiet as a mouse - even the 5E3s. The proof is in the pudding so to speak. I got no beef with what Merlin says either - his book is compulsory reading. Its just that there's more than one way to skin a cat.