OPT Design Assistante (EL84)

[blindsjc]

Hi friends,
First of all, sorry about my poor english language.
I'm new on the forum and hope to learn a lot here.

I'm starting study the output transformers, have
some experience with solid state amps and effects
but still a newbie with tubes.

I'm using the OPT Design Assistante, great program,
but I have a lot of doubts about the real and correct
way to use it. I plan to start with a EL84 PP amp.

The specs Iam using with opt_da are:

Freq(hz): 40 -> Think its the lowest frequency of transformer
Watts: 20 -> EL84 PP about 18 to 20 watts
SE option not checked -> PP opt
Rp(Ohms): ?? -> not sure about this value
Primary (Z): 8000 -> EL84 PP impedance
Ip (0): 0.140 -> not sure about this value

So, any info about this program and values
will be very important to me. Thanks a lot,
sorry about the idiot questions.

Blindsjc

[BudP]

blindsjc,

Welcome, lot's of people here struggle with English so do not feel embarrassed about that.

The Rp is plate impedance in mhos, often called transconductance it it the actual plate impedance, as opposed to the load impedance. The program is going to add these two numbers Rp & Rl in parallel ie. (1/Rp + 1/Rl) + primary winding DCR. This equals the reactive load resistance of the tubes. This reactive resistance is then divided by the frequency of interest and 2 X pi to find the inductance the full primary of the PP OPT must have, for - 3 dB at the frequency of interest, 40 Hz. Having this value at 1 volt 120 Hz guarantees that you will have at least three times this value of inductance, once power is actually applied. Three times the needed inductance means you will be about -.75 dB at 40 Hz for most listening levels. Very quiet will be about - 2 dB and so will very loud.

The Ip is an indicator of what currents are flowing through the primary wires for ohmic heating, just to make sure that you do not run too hot. This is not usually a factor because you really do not want much more than 100 ohms of DC resistance in each half of the primary winding and here, less is better, all the way down to about 20 ohms per side, beyond that no benefit is derived.

Post the design results when you think you are getting somewhere, there are lots of very competent folks here who will help.

There are NO stupid questions.

Bud

[Yvesm]

Originally posted by blindsjc

. . .
The specs Iam using with opt_da are:

Freq(hz): 40 -> Think its the lowest frequency of transformer
Watts: 20 -> EL84 PP about 18 to 20 watts
SE option not checked -> PP opt
Rp(Ohms): ?? -> not sure about this value
Primary (Z): 8000 -> EL84 PP impedance
Ip (0): 0.140 -> not sure about this value
. . .
Blindsjc

Hi Blindsjc

Thanks for apreciations about the program !

For penthode PP, use Rp = Zp, here 8000.
It will be lower in the real amp if some feedback is used around the outpt stage.
Try a simulation with 2000 to have an idea on how this affects frequency response.

Ip is for each tube, so 0.05 is good for each EL84 (with 250V on plates, Ip is limited to 12 / 250 = 0.048 by the maximum allowed plate dissipation).
This parameter affects the diameter of the primary wire.
Check for TOTAL and AC copper losses to see its effect

Remember to set the gap to 0 mm.

Yves.

[blindsjc]

Thanks a lot friends,
Things are a lot clear now...hahaha...thanks.
and Yves, this programa is really good.

Another question is about the iron cores,
my supplier dont have the model specified
by the program (EI84B), they have a EI85
like this pdf info.

http://www.mgsel.com.br/biblioteca_d...I85_ST_CAN.PDF

There are some place where I can find a
complete reference about this core standards?

I think that this supplier codes haves
nothing to do with the standards which
appears on the program and I will need
to use the mm values from the "core data"
window.

Thnaks again.

Blindsjc

[Yvesm]

Originally posted by blindsjc

. . .
they have a EI85
like this pdf info.

http://www.mgsel.com.br/biblioteca_d...I85_ST_CAN.PDF

They look very similar.
"EI" stands for the mecanical aspect of lams looking as the letters E and I parts that will be packed (stacked) to obtain the core.

The number "84" in european standards is the lenght of the "I" part of the lams. Its width - as long as the width of the lateral legs of the "E" part - is always 1/6th of this value (here 14mm).
The width of the center leg of the "E" part (also named tongue) is always 1/3rd (here 28mm).

Theses relations apply to all "scrapless" laminations.

In your drawing lenght of the "I" part is called "A" (86mm) wich however includes the mounting brackets thickness.
From this value, you may compute all others from relations above.
The "B" value being equal to A * 5 / 6 .
But the height of the stack is missing.

Look at
http://www.elettronicanovarria.com/r...inonuclei.html

(sorry for not having found in english)

The leftmost table is for bobins, the right ones are for lams/core in various grades.

The lams size we need is 84 x 70 mm (note that 70 is 84 * 5 / 6 as stated above)
This means a thongue width (colonna in italiano) of (84/3) = 28mm.
The B suffix stands for a 45 mm stack (paco)

Does that help ?

Yves.

[blindsjc]

It helps a lot,
I was really lost because the internet
resources are so few. Now I see that
things are all related. Hope start to
build something soon.

I'm saving all this info.
Thanks, thanks, thanks.

Blindsjc

[Yvesm]

Great !

Show me what you plan to do when ready.

Over all, your first specifications seem a bit optimistic regarding the max output power from 2 x EL84, probably near to 15 than to 20W.

But you are slighty pessimistic about lowest frequency, I would have targeted 30Hz rather than 40 !

BTW one somewhat compensates the other.

Yves.

[ashok]

Blindsjc ,

What program are you using ?

[JPS]

Hi Ashok

Software is written by Yves.
His website link is.

http://www.dissident-audio.com/OPT_da/Page.html

very good software.

regards

[ashok]

Thanks.
Yes I have seen that before. Never used it.
Cheers.