Single tube SE with EML-20B

[tinitus]

Hey, you probably think now I must be completely daft to even try

I hope you can help me to at least get a bit better understanding of how it works

I have never done anything like this before, at all

EML-20B is a driver tube, and a power tube, if Im correct
This means that it can work on its own, without driver tube

Well, I found a 300B schematic, and changed a bit here and there
Maybe not the best start really
Some details havent been changed yet
and some changes I have made may be wrong
something I just did in a hurry, without having much clew about it

Im not expecting anything
and I know I should better let it go
but it tickles my curiosity

 

[tinitus]

a bit more work on schematic, bias curcuit

 

[tinitus]

more schematic work
still in doubt how it works

 

[kevinkr]

I previously replied, but deleted my post as a result of doing a little research on this tube..

I see the EML-30B has a mu of 20, rp 3.3K, recommended Ip of 60mA and a pd of 25W. This means that an SE output transformer with a primary > 10K is required - this is actually not a trivial endeavor, and a suitable quality OPT will be very expensive.

The current proposal requires the use of both an input transformer and an auto-transformer which will severely compromise both the LF performance and the HF performance. A single transformer with a ratio of 1:10 would provide sufficient gain, but also a rather low input impedance for good performance - the impedance transformation ratio would be 100:1 implying a 60K secondary with a 600 ohm input Z, possible, but is it wise if any significant current ever needs to be delivered to the grid? (And it will result in problems with the high frequency response due to interaction with the miller capacitance of the 20B)

Seems like a low mu triode like a 6J5 would do a much better job driving the tube, providing a low driving impedance to the grid of the 20B and a high input impedance and sufficient voltage gain to assure it works with real sources.

I'd go with fixed bias in the output stage were it me, this would allow easy tuning of the operating point if desired and eliminate a fairly lousy capacitor in the output stage cathode circuit.

The killer IMHO is the output transformer requirements, at these impedances I would definitely suggest a c-core as the most viable choice if you want decent HF performance. Maybe AE, Bartolucci or Lundahl have something suitable?

 

[tinitus]

yeah, you are pointing at exact some of the worries I have had
and explaining why, thanks

please note that grid choke is a Lundahl, supposed to give a 1:2 gain, configured like this

I also thought using input trafo for gain, maybe takes care of DC, thus no cap needed, but I dont know

And I have considered a need for use of tube preamp, maybe

 

[Jaime]

Gordon Rankin do it:
Mercury Series, Single Stage, Single Tube Amplifiers
Wavelength Audio Single Ended Mercury Series
I believe AVVT 20SL (VAIC) >>> EML-20B (Emisión Labs)🙂

The Mercury skema is not published but Gordon post this in other forum
MagneQuest/Peerless Forum: Simple 84 by Gordon Rankin

 

[tinitus]

hey, that single tube EL84 SE schematic might be useful
great

regarding AVVT vs EML
yes, same people
but I think designs have improved
Jac(Germany) is now the owner
seems to be in safe hands
ofcourse, using such a tube always presents certain risks
how long will they be around etc

btw, I also chose 20B because its relatively affordable
and I found it interesting
but hey, this is so far just a learning exercise
and sorry to say, it probably stays that
in the end I will probably choose a design I know works
but I learn a bit here and there, small stuff

hey, I made my first schematic, stealing bits and pieces
does it work ?
I have no idea, but I somehow doubt it

Im open to all suggestions

 

[nicoch58]

Bob's EML 20B One Tube Single Ended Amp 😉

 

[tinitus]

Hey, I know Min´s site, its very nice
Bob´s amp ?
Wonder who designed the schematic
I could try and ask Min
its great, thanks

 

[tinitus]

Gordon Rankin do it:

The Mercury skema is not published but Gordon post this in other forum
MagneQuest/Peerless Forum: Simple 84 by Gordon Rankin

grid choke with gain, at least I got that part right, well almost

the one with plate choke, I have no idea
it looks kind of more real to me, dont know why
connecting B+ power directly to OPT always puzzled me, thou I know its the usual way it works

hey, I know where grid and plate is now, thats worth something 🙂

 

[tinitus]

hmm, even simpler, with a high gain preamp

 

[tinitus]

moving into dangerous territory 😱 I must be out of my mind 😀

 

[tinitus]

maybe a bit better now
if only those tubes werent so expencive

 

[Bigun]

What a great looking tube, that's reason enough in my books to build an amp using it.

Do you really need parallel tubes (last posted schematic) ? since a single tube already produces a lot of power and my own tests with much less power suggests you may not need it.

I prefer integrated amps, no need for all those boxes, cables and connectors etc. Would you consider putting the pre-amp in the same chasis ?

My (limited) experience is that using a good 6SN7 would make a nice preamp, one half per channel. You might then not need the autoformer ?

 

[kevinkr]

What a great looking tube, that's reason enough in my books to build an amp using it.

Do you really need parallel tubes (last posted schematic) ? since a single tube already produces a lot of power and my own tests with much less power suggests you may not need it.

I prefer integrated amps, no need for all those boxes, cables and connectors etc. Would you consider putting the pre-amp in the same chasis ?

My (limited) experience is that using a good 6SN7 would make a nice preamp, one half per channel. You might then not need the autoformer ?

Echoing my earlier suggestions.. 😀 I imagine just about anything tube connected to that autoformer is going to have a really hard time driving it, and the miller capacitance will very effectively kill the HF response even with a single tube...

 

[tinitus]

hey Bigun, nice to see you around

autoformer
if its the last schematic, its a special grid choke
from what I understand it does good, other than being voltage multiplier

yeah, paralel SE was just a last crazy idea
I dont know
its a power tube with high gain
maybe theres risk of noise issues, or other tricky stuff

I expect the paralel tubes would make the load on OPT easier, or better, maybe
and it ought to give a bit more output, or headroom
not pushing it too hard should result in longer tube life

no pain, no gain, I suppose

Im not doing this to build myself a better amp
I just want to build a nice tube amp
and I want to build it, and know it inside
or I would rather go buy a Supratek

and to make it complete, it really must have a tube pre

I may choose to build something completely different
But 300B doesnt speak to me, for whatever reason
2A3 ? that one speaks to me, but very low power
same with 6V6 and 6L6
6V6 speaks to me
I have never heard any of them, and thus completely foolish
a funny thing with tubes

 

[tinitus]

hmm, I wonder how this thing would look in balanced/PP
more complications I guess

 

[Bigun]

hey Bigun, nice to see you around

Good to see you too. I've been hiding amongst the tubes for awhile since I finished my SS amp.

autoformer
if its the last schematic, its a special grid choke
from what I understand it does good, other than being voltage multiplier

Well, yes, I believe grid chokes have some advantages, although they seem rather rare in a hobby that is often willing to chase any possible advantage that's available. But grid chokes are usually installed to provide a nice low impedance dc bias for the grid without shorting the signal to gnd. I'm not sure such a thing can play it's best song when used as an input auto-former. But then, the road less travelled offers new discoveries for those brave enough.

yeah, paralel SE was just a last crazy idea
I dont know
its a power tube with high gain
maybe theres risk of noise issues, or other tricky stuff

I expect the paralel tubes would make the load on OPT easier, or better, maybe
and it ought to give a bit more output, or headroom
not pushing it too hard should result in longer tube life

no pain, no gain, I suppose

I'm a newbie here but I would have thought a single tube is a good starting place, unless you really badly need the extra power and your OT can handle it. I would see it adding stress to the OT rather than reducing it. You'll have more dc bias current with two tubes for starters. The two tubes will give you more power, by being able to provide more current. That's only more headroom for a low impedance load and I haven't found low impedance output transformers being so easy to find.

Im not doing this to build myself a better amp
I just want to build a nice tube amp
and I want to build it, and know it inside
or I would rather go buy a Supratek

and to make it complete, it really must have a tube pre

I may choose to build something completely different
But 300B doesnt speak to me, for whatever reason
2A3 ? that one speaks to me, but very low power
same with 6V6 and 6L6
6V6 speaks to me
I have never heard any of them, and thus completely foolish
a funny thing with tubes

I think I would agree with the advice posted already - a two stage amp. Input tube which is your pre-amp and which then drives the grid of your output tube. The fact that your output tube has some gain means you don't need a high gain input tube - which is good because they tend to have high internal resistance which is bad on many fronts.

Based on my limited experience [http://www.diyaudio.com/forums/tubes-valves/167872-my-cellini-triode-amp.html], I'd advise you just go and prototype something. It's a bit of a hassle due to all the chasis work, but it'll settle a lot of questions and you will get to know your amp really well before you finalize it.

Oh, and the first tube amp I heard was the one I built, so you're not alone there !

 

[kevinkr]

hmm, I wonder how this thing would look in balanced/PP
more complications I guess

Here you need a PP input transformer, what you have drawn will not work with unbalanced RCA inputs.

Note that you will have one heck of a time driving the miller capacitance with a transformer.

You would need something like a 1:10+10 or 600:60K+60K to drive this, and this would most likely have to be a custom.

All I can say is buy a couple of these tubes and (carefully) build something!

Were it me I would choose a JJ 300B or a PX25 over this tube for an amp any day, however the mu of these types is too low to use a transformer step up.
This would be great as a driver with a choke, CCS or resistive load with a bigger power tube like the 845 or GM70..

 

[tinitus]

I'd advise you just go and prototype something.

yes

my evaluation
the idea was one tube and one tube only
if a driver tube is needed, I could just as well build with 300B

20B in P-P might make sense too, but gets expencive, and still low power
but you are right about the driver tube
whether its a high gain pre or added driver tube, its the same
could just as well be integrated driver tube


This brings me to another tube, ECC99
I have thought about it a long time, but forgot
maybe now is the time to try it
that too is a triode driver, and strong enough fore a small power amp

all ECC99, either PP or PPP, or even flea power PSE, ecc99 driving ecc99

btw, specced 5watt heat/power disspation of ECC99, is that each half, or complete tube😕