AC4 (the version discussed-AC30 Pre + AC4 Power) - Which Primary OT impedance ?

[kagliostro]

I got a broken Fender Frontman 15G and want to convert it in a tube amp

My plan is to fit it with the circuit planned on this thread (AC30pre + AC4 power section)

http://el34world.com/Forum/index.php?topic=19764.msg206086#msg206086

(I've the beautiful layout drawn by Sluckey)

I did a search and after it decided that I'll use a 250v AC PT and Solid State rectify

this will give me a B+ around 350v (the AC30 circuit seems to be around 310-320v so I don't know if I must lower it a bit (in some way))

About OT primary impedance I wasn't able to find info (someone say 5K other 7K other 8K)

so I would like to know which will be the best match for such B+ on that tube (on datasheet the max SE voltage is 250v DC)

(I think I'll use one 6P14P-EV instead of EL84, those russian tubes can afford such a voltage and has a plate dissipation of 14W)

Thanks

Franco

[Tony Bones]

The short answer (and probably the most correct) is that anything from 5k to 10k will work.

Engineers looking for maximum power output will tell you that as a rough rule of thumb, as plate voltage is increased the load should be increased as well. But who cares about a watt or two between friends? 

Then there's Fender that built the blackface Deluxe with plate volts way over the datasheet max and used a load LESS than the datasheets might suggest (though the datasheets weren't suggesting anything at 400V+.) Of course the BFDR is PP, so what's it have to do with a jacked up AC4 anyway?

You could look at graphs like this one that plot different stuff vs load.



Or you could study graphs like these that plot other stuff vs output power. Each graph is for a different plate load.





Or you could go back to my first answer, which was that almost anything will work.

[DummyLoad]

use fender champ OT.


--pete

[drgonzonm]

A quick search of the web on 6P14P tube, indicates it is  equivalent to the EL84 tube.  So use EL 84 data. 

The EL84 sheet I have, uses 8k as plate to plate at 300v (pp) and 4K5 to 5k at 250v (SE).  I would also look at some those OT impedance matching tables for various power tubes.  Additionally I would look at common amplifiers, that use el84's and then search the Hammond Transformer site, for the specifications on replacement transformers. 

Since you are in Europe, consider an isolation transformer, and a filament transformer.  This should drop the B+ to where you want to be maybe lower. 

I have not reviewed the AC30 circuit, so I do not know if it has any chokes.  Lets say it doesn't,  the PI filter circuit will maximize the B+ voltage if necessary.

What you save in PT iron by using an isolation PT, you can spend on a choke 3-4 Henries, and 80 ma? 

Good Luck

Consider a split load pp amp, gain a couple of watts, and simplify your OT selections to PP type. 

[PRR]

> B+ around 350v

The EL84 and similar are 12W-14W Pdiss tubes.

You will lose 10% in OT DCR and cathode bias, 315V.

12W/315V is 0.038A. That's your plate current.

315V/0.038A is 8,269 Ohms. That is a first-guess load impedance for this tube and voltage. Wide variation is acceptable. 5K may bring-out the compressed plate curves under the knee. 10K may be cleaner up to a point and then clip abruptly.

> use fender champ OT.

Often the best answer.

Especially since the Champs tend to run very similar B+ with a same-size tube (6V6). And a "Champ OT" will be scaled for *guitar* (not 25Hz hi-fi, not 300Hz intercom).

[kagliostro]

Many thanks to all you friends

Seems that a commercial Champ OT (7K version, there are also 5K), will be a good choiche for this tube at this voltage

---

PRR, sometime you talk about curves under the knee, this thing is not so clear to me, can you give a brief explanation ?

Thanks

Franco

[PRR]

.

[shooter]

Now that's a graph

[kagliostro]

OK, nice pair of knees, however the difference on being under the knee, instead of being on the top side, escapes me 

(Insofar as I had always been convinced that the better part was over the knees and not below    )

Franco

[shooter]

the better part was over the knees

My understanding is you want to operate on the "long", basically horizontal portion rather than the sharp vertical portion.

[kagliostro]

 

[shooter]

I lifted this from Merlin's page, I sorta "visualize" a sin wave with zero crossing at the bias point, then see how much it can swing till it hits the "stops", where it begins to scrunch up.

I know, not as original as PRR

EDIT:
graph "borrowed" from;
http://www.valvewizard.co.uk/Common_Gain_Stage.pdf

[sluckey]

the better part was over the knees

If you continue to trace two curves above the knees you will come to a junction. That's the sweet spot!

[kagliostro]

Thanks for the explanations Shooter & Steve
 


Franco

[Tony Bones]

I lifted this from Merlin's page, I sorta "visualize" a sin wave with zero crossing at the bias point, then see how much it can swing till it hits the "stops", where it begins to scrunch up.

I know, not as original as PRR

Those are triode curves. Triodes curves generally don't have knees. Look again at the graphic that prr posted above. The one with the lower half of a woman. Look past the female legs at the curves behind them. The curves take a turn at low voltage. That curve is the knee that prr was originally referring too.

[shooter]

triode curves

I was lazy and didn't want to superimpose a sine wave, That graph is also about harmonic distortion, but it "illustrates" how a sine wave "operates" in relation to the linearish portion of the graph, (above the knee).

[PRR]

> nice pair of knees