Single ended EL34 amp design

[bepone]

OT taken is too large, no need so big, and price is following 10-15W SE is more than enough..10w will also pass even if output power is 12, 13, 14W this is all the same, will saturate sooner from bottom part (distortion on bass side)

You are probably right, I like this transformer because it looks well built and has a good bandwidth compared to similar Hammond transformers for example. I'm not even sure I will stay with this though, there are possibilities for extra gain using different primary impedances.

What do you think of the power transformer though?

In chosing the SE OT's, DC current is No1 factor! And must be optimised, exactly for this design! Why?

Because in SE OPT you have DC and AC magnetisation in the core, DC magnetisatiion is more important, to "bias" the core exactly in the linear part of the HB curve! Somewhere in the half (this is manufacturer problem if is not like this, and is difficult to measure at home).

We dont want to take 200mA SE trafo if we need only 150! If do, then with 150mA we will bias DC magnetic flux through the core in the sub optimum area, and saturation and distortion of the transformer will limit the output power! We want flux to be in the half of linear area..

The same for output power, no need to take 20W if we need 10W, we will just have lower inductance L, because with bigger core we take lower number of the turns..
Bad for our bass side. Smaller core with higher number of the turns is better for guitar, and tightness on the bass side..

SE OT still must be bigger than PP transformer for the same power because is wasting half of the flux for nothing, only to bias output tube.

So some medium transformer will pass nicely, not too small, not too big

[Stephen1966]

Rg1 as small as possible, so 220-330k

Thank you bepone. As I understand it, the grid balance resistor also has something in relation to the NFB you mention below.

So that I may understand it better, could you explain how and why you arrive at 220-330k for Rg1?

thx,
In case of too big grid leak, out. tube will be less controlled, you can have thermal runnaway, and current in the output tube will raise more and more and more and more until melting.. or some fuse activating.. it is rare the case but is happening, if you want to be sure, just put less than half recommended, (0.7 Mohms in EL34 datasheet)..

This will load previous stage, but this is also ok because load line 330k in parallel with anode resistor in previous stage *ECC83, will generete more 2. harmonic what is also good for us!

Tube less well controlled... thermal runaway... that makes sense. I think I understand the process. The proximity of the grid to the cathode means that heat is conducted and the grid develops a small positive charge. A larger resistor makes this worse. The reverse grid current generates more heat on the grid which in turn attracts more electrons, which increases the heat and so on... thermal runaway and red plating. I'm paraphrasing from Merlin's preamp book, when I say the cathode biased circuit is less susceptible to this because as anode current increases so does the voltage across the bias resistor. And in there he concludes the limit is more an 'advised limit than absolute' and unlikely to cause problems 'provided the Vhk(max) is also not exceeded.' According to the Mullard datasheet:

Rg1-k max (cathode bias) = 700k
Vh-k max = 100V

So I think we are within the acceptable limits with Rg1 = 470k but I take your advice seriously because after all, heat management is an issue when a substantial part of the tube's output power is dissipated as heat.

When you say 'this will load the previous stage' is this the effect known as bootstrapping? I've read about that with regard to cathode followers but you seem to be talking about the same thing here.

[Stephen1966]

OT taken is too large, no need so big, and price is following 10-15W SE is more than enough..10w will also pass even if output power is 12, 13, 14W this is all the same, will saturate sooner from bottom part (distortion on bass side)

You are probably right, I like this transformer because it looks well built and has a good bandwidth compared to similar Hammond transformers for example. I'm not even sure I will stay with this though, there are possibilities for extra gain using different primary impedances.

What do you think of the power transformer though?

In chosing the SE OT's, DC current is No1 factor! And must be optimised, exactly for this design! Why?

Because in SE OPT you have DC and AC magnetisation in the core, DC magnetisatiion is more important, to "bias" the core exactly in the linear part of the HB curve! Somewhere in the half (this is manufacturer problem if is not like this, and is difficult to measure at home).

We dont want to take 200mA SE trafo if we need only 150! If do, then with 150mA we will bias DC magnetic flux through the core in the sub optimum area, and saturation and distortion of the transformer will limit the output power! We want flux to be in the half of linear area..

The same for output power, no need to take 20W if we need 10W, we will just have lower inductance L, because with bigger core we take lower number of the turns..
Bad for our bass side. Smaller core with higher number of the turns is better for guitar, and tightness on the bass side..

SE OT still must be bigger than PP transformer for the same power because is wasting half of the flux for nothing, only to bias output tube.

So some medium transformer will pass nicely, not too small, not too big

This OPT https://primarywindings.com/product/500 ... ansformer/ has 100mA DC current at the primary - any good? Quiescent Ia = 63mA and the theoretical peak - point (A) on the chart is 133mA. I would expect some heat to be generated but the tube is unlikely to ever reach that peak. It would take some really heavy sustained overdrive to get it there!

The 20W output is overkill but regarding that bandwidth, if the manufacturers claims are true it has a much better low end response than the Hammond. A 15Hz roll-off versus a 100Hz roll-off if I remember it well.

[bepone]

When you say 'this will load the previous stage' is this the effect known as bootstrapping? I've read about that with regard to cathode followers but you seem to be talking about the same thing here.

no, it is load, load line for previous stage, this resistor 470k (or 330k or 220k) is in parallel with the anode resistor of previous stage for AC signal analysis, so if is 100k in anode of ECC83 before EL34 via coupling cap, load for ECC83 is 470k//100k=82k, or 330k//100k=76k

[bepone]

This OPT https://primarywindings.com/product/500 ... ansformer/ has 100mA DC current at the primary - any good? Quiescent Ia = 63mA and the theoretical peak - point (A) on the chart is 133mA. I would expect some heat to be generated but the tube is unlikely to ever reach that peak. It would take some really heavy sustained overdrive to get it there!

probably is good, but 100mA is more than you need, if you can find for 70mA /10W even better

[bepone]

if i need to choose from that manufacturer i will choose that one, stand up with 4,8,16 ohm :
https://primarywindings.com/product/kt88_10w/

3k5 will be, 90mA of max current, 10W, M6 laminations, and a lot of inductance..

use of this OT immediatelly will generate another working graph for EL34 (Ub=350V, Uak cca 320, Ug2k cca 300V)
i need graph for Ug2=300V, but i have used Ug2=250V... so those numbers are not valid, only load line is.
it is necessary to spend some time, but i have no time, and is not even important.. amp can work in that working point and bias (cathode resistor) can be determined after first startup, fastest way

3k5.png

[Stephen1966]

if i need to choose from that manufacturer i will choose that one, stand up with 4,8,16 ohm :
https://primarywindings.com/product/kt88_10w/

3k5 will be, 90mA of max current, 10W, M6 laminations, and a lot of inductance..

use of this OT immediatelly will generate another working graph for EL34 (Ub=350V, Uak cca 320, Ug2k cca 300V)
i need graph for Ug2=300V, but i have used Ug2=250V... so those numbers are not valid, only load line is.
it is necessary to spend some time, but i have no time, and is not even important.. amp can work in that working point and bias (cathode resistor) can be determined after first startup, fastest way
3k5.png

Thank you, that's great. I want to say again, how much I value the time you are spending with me on this. As you can clearly see, I didn't begin this with a firm grasp of the essentials but I am learning so much from your patience, understanding and guidance. Not everyone has the time to give as much as you have already given, so yes, grateful is an understatement.

Thank you as well for the clarification on loading. I have the time and there is no rush for me at the moment. When the build happens it will take only a day or two and we might find there that all the charts were off by as much as +/-20%. I agree, they aren't that important. But for understanding - they are invaluable.

[bepone]

no problem, enjoy in the designing of your own amp

[Stephen1966]

In this iteration I have played with a Zpri of 3k5 on the output transformer following the suggestion of bepone.

3500 Ohm Single ended 10W Output transformer (KT88 or similar)
https://primarywindings.com/product/kt88_10w/

I also want to use the 10H choke:

PWCH01: 10 Henry 100mA DC Choke
https://primarywindings.com/product/10- ... dc-choke/

With a Va = 350V and a negligible voltage drop across the choke = 2V I have gone the easy route of finding the value of the screen-grid resistor, marked by point (A) on the mutual characteristics chart. This screen voltage drop would provide maximum power with a minimum of distortion but these values seem somewhat arbitrary and so though the voltage drop indicates Rg2 = 675Ω, I've gone for a 1k value. This reduces the gain, cools the bias voltage and reduces the THD though it does tend towards a slight increase in 3H distortion - mostly inaudible - well below 5%. By choosing a higher value Rg2 grid stopper, we are increasing the compression, touch sensitivity and linearity (hence the 3H increase). We are also moving away from the problem of over-dissipation of the screen and thermal failure.

EL34 - Va 350 - Zpri 3k5-01.jpg

After calculating the sum of quiescent choke and screen currents this provides a Vg-k = -24V; point (B) on the mutual characteristics chart. Rk and Ck are as before.

I've also reduced the value of Rg1 - the control-grid leak resistor but this is a nominal value along with the control grid stopper until I have all my ducks in a row with the preamp design.

SE amp - 02.jpg

Power output is increased over the previous design. About 12.6W(peak) before any losses but along with any estimates of THD these are just working figures before the circuit is constructed and "real" measurements can be taken.

[Stephen1966]

I'm starting to think about the preamp design now and I've been looking at the Champ Revisited block diagram.

Figure2_TheRemakingofaChamp.jpg

I've also been looking at the Bassman tone circuit. I'm thinking I would like to make use of the spare triode in the first stage for a second channel input for a harp. To give two inputs basically: instrument and microphone. I'm starting to look around at various Blues Harp designs and the Bassman comes out on top as the amp par excellence but I'm not yet convinced that I can inject a third stage CF into the tone stack and then straight into a SE, class A pentode. The topology of the Champ Rev. seems a somewhat better way to go. Reading Kuehler's analysis, it definitely seems that the tone of the Bassman is hugely dependent on the PI/PP arrangement. This current design, is going to be something else, but I'm not quite sure what yet. The Marshall JTM 45 seems quite the sonic flavour I am heading towards. And so, a higher mu first stage with a 12AX7 type tube. I'm learning LTSpice as a consequence. Once I get the hang of it, the simulation powers will be useful.

I'm also still thinking about a DC rectified circuit for the heaters. Depending on power transformer options, I might just offer the regulated supply to the preamp sections and let the pentode run on AC - that would simplify all kinds of things.

With a 3A filament winding as standard, there are a number of ways I could cut it. The deal might also include a tube FX buffer (like the Dumbleator) since it only requires an extra tube and this seems within budget at the outset. Not given up entirely on the reverb pan I was given, but a buffered loop can do much much more.

What do you think?

[bepone]

How i have in plan in the future to assembly one S.E. 10W amp , very similar to this one from topic, i can share what i prepared for it, trafos bought from Italy, Piemme, chassis from Tube town.

Calc. and draw load lines 5K:

10W SE constr..png

Mech constr:

10W SE mech.jpg

Chassis: https://www.tube-town.net/ttstore/tt-ch ... ossen.html

Choke for power supply, CLC: DC-EI60013F

Transformers: PS-EI84023CV-N, SE-EI68079CV-N R02

Preamp can be any, cathode bias output...

[Stephen1966]

Yeah! That's kinda what I'm thinking. Interesting PT choice - 100mA off the HT - is that adequate enough to drive the plates under peak conditions? I'm still not decided between:

Amp Maker PWAM01 275V-190V@ 160mA 6.3V@3A https://primarywindings.com/product/amp ... a-6-3v3a/ and the
Hammond 290XEX for Fender Power 049967 / Blues Junior https://www.tube-town.net/ttstore/hammo ... unior.html

I'm thinking the first is going to do the job.

Your 3H choke compared to my 10H choke - do you think I am going overboard on the filtering? What's interesting is that on your Piemme choke they have a polarity marked in the schematic. Just yesterday, I reversed the leads on my choke in the 183 and I hear a distinct difference. It will be interesting to hear what you make of this if anything.

Your preamp looks pretty straightforward compared to what I'm concocting. For the V1 stage, I want the Lo and Hi inputs and I'm looking at installing a 12DW7 there. With the AX system processing the guitar signal and the low gain AU side processing a signal for Blues Harp. It might be cool to put the tone stack at the end of preamp chain, just before the power stage, and to look at driving it with a CF and a 12AT7 is something I keep meaning to try in that role.

[Stephen1966]

Other similarities...

A gain boost, yes. switching out the tone stack.

Once I get my head around Spice I want to try a couple of different tone stacks. The Champ Revisited circuit appears to be something like a Baxandall at least in the first part of the circuit but the Bassman tone stack has nicer mid scoop. All these voltage dividers though - it starts to look very lossy as it gets more complicated so the eventual design might involve just a Treble and Bass, or even just a Tone control - similar to the 5E3.

On the heater side, we are both looking at a 3A supply. Not enough to feed a regulated supply to all the tubes perhaps (the EL34 is hungry) but possibly enough to provide a regulated supply for the preamp tubes and AC for the power tube.

I would like to ask you - how do you implement a Presence control in a SE cathode biased amp? And, do you need a NFB net for this to work?

[bepone]

Yeah! That's kinda what I'm thinking. Interesting PT choice - 100mA off the HT - is that adequate enough to drive the plates under peak conditions? I'm still not decided between:

yes ma this 100mA is working only half cycle, so at least double current is possible to extract (and this is not needed ) it is A class, so current will be around 80mA permanent draw

[bepone]

I would like to ask you - how do you implement a Presence control in a SE cathode biased amp? And, do you need a NFB net for this to work?

when you are playing with negative feedback resistor, this is already some kind of presence because is affecting the highs, but i saw somewhere schematic with this function on SE amp..
but when the amp is setup good, it is not needed