need help: pcb layout for gmeq5

[thinktank]

I would like to design or have help laying out a pcb
for an m-e-q-5 eq. I've been studying the various papers
on the gyraf pultec and noticed that one could use
the same pcb except change the filter pcb. This is where
my knowledge ends. Any ideas on how to go about this?

Also, I'm trying to find the inductor values for the eq portion.

Thanks for any help. Again, I'm really new to this but really
want to learn. Thanks.

Nate

 

[CJ]

Maybe cayocosta could do another drawing for the meq.
I would host it.
it's not a pc board, just the terminal board thingy, which is better for tube diy IMHO.
Are not the inductor values at the Gyraf site?
cj

 

[thinktank]

I really appreciate it. You're right, the schematic has the inductor values.
The translation from the schematic to laying it out is what I need help with.
A drawing would be great. Thanks.

Nate

 

[CJ]

you can have a gander at this. sorry about the load time. lots of pics!
maybe you could even adapt this drawing to the meq if your a wares type of guy.

http://www.vacuumbrain.com/docs/pultec.html

 

[thinktank]

I took a look at the layout and tried to figure out the layout of the meq5,
but couldnt' figure it out. I've got that schematic from the Gyraf website on
the meq5 and also the one for the peq1a and it looked like the layout
was different than the schematic.

Anyway, what would be my next step to try to lay something out?
Thanks for your help.

Nathan

 

[CJ]

You want a pc board for the filter , line amp or both?

 

[thinktank]

You know I've got the pcb from the gyraf peq1a for the power
portion. I guess I need to figure out the filter part in order to
translate that onto a pcb or like you said, point to point. I don't
have a problem doing that. I've done a couple la2a's so I should
be able to wire this baby up. I hope so.

I think the line amp portion is included in one of the gyraf pcb's.

Thanks,
Nate

 

[CJ]

Cool. A turret board would be the way to go. There is a lot of trial and error when finding the right caps to match up with your inductors. A PC board would be a disadvantage if you had to remove and replace caps. You would probably end up lifting a bunch of pads. Sometimes you can just grab a bunch of caps and a turett board and just physically ararange them without soldering.

Or just get a big piece of paper the size of the box you have and do a full scale drawing, then transfer to the real thing.

 

[thinktank]

Sounds good. Trial and error's always a good thing anyway.
I'll lay it out on paper and if you don't mind, I'll email you a
jpeg of my layout. Or just post it here. Not sure how to do
that though. Thanks.

Nate

 

[CJ]

Just draw a bunch of caps on a turret board. Then you draw wires going to the inductors and switches. Then after your done, you can re arrange things to get it a little neater. Remember that some of those big 1.5 uf/100 volt caps are big. You do not want to use electrolytics in the MEQ.

 

[thinktank]

Any idea where I can find turret boards in my area?
I'm right outside of Los Angeles. Or does digikey
or newark have them in stock?

 

[mcs]

[quote author="thinktank"]Any idea where I can find turret boards in my area?[/quote]
You could use the method I used many years ago. Get a piece of wood and some brass nails...

Then you have a "sømplade" :green:

Best regards,

Mikkel C. Simonsen

 

[thinktank]

I'm still trying to lay out the parts that I'll need for my meq5 type eq,
but can't find the correct inductors anywhere. I've got the various
cap and inductor values from the gyraf site, but can't find all of the inductors.

I went ahead and read the post that CJ did on figuring out which cap and inductor sizes are used for which frequency, but I'm trying to figure out what reactance values to insert into the equation. Maybe I'm not reading carefully enough. The equations seem to work for the peq eq frequencies.
But I'm lost a little on how to determine the reactance without having anything to go on at my level of this game. Thanks.

Nate

 

[jhaible]

[quote author="thinktank"]I'm still trying to lay out the parts that I'll need for my meq5 type eq,
but can't find the correct inductors anywhere. I've got the various
cap and inductor values from the gyraf site, but can't find all of the inductors.

I went ahead and read the post that CJ did on figuring out which cap and inductor sizes are used for which frequency, but I'm trying to figure out what reactance values to insert into the equation. Maybe I'm not reading carefully enough. The equations seem to work for the peq eq frequencies.
But I'm lost a little on how to determine the reactance without having anything to go on at my level of this game. Thanks.

Nate[/quote]

Does anybody have the original (not Gyraf) MEQ5 schematics?
On the gyraf version, some component values apparently don't
match the nominal frequencies.

JH.

 

[gyraf]

Shouldn't be so hard to figure out the right inductors with a pocket calculator or a "nomograph":

http://www.testecvw.com/carl/images/ImpedanceNomograph.pdf

Jakob E.

 

[jhaible]

[quote author="gyraf"]Shouldn't be so hard to figure out the right inductors with a pocket calculator or a "nomograph":

http://www.testecvw.com/carl/images/ImpedanceNomograph.pdf

Jakob E.[/quote]

Hi Jakob!

Well yes, and the formula is quite easy, too.
But as you have 2 degrees of freedom, L and C, and no information
about Q, it is important to either know a prooved value of C
or L for a given center frequency. Just choosing an arbitrary L
and calculating the C to match will also change Q.

For instance, if you share one inductor value for 3 bands, this
means the three mid frequencies will come with different Qs.
As the original circuits have a reputation for "just sounding right",
a certain choice of Q for a certain band might be important.
So before I start to assume that L is right in the drawing, and
calculate the Cs accordingly, I'd rather like to know if it is not
the other way round.

JH.

 

[alk509]

[quote author="thinktank"]Any idea where I can find turret boards in my area?[/quote]

I usually buy Keystone turrets from Mouser and a 1/8" thick garolite sheet from McMaster-Carr and roll my own turret board.

If you DIY, be sure to drill the 3/32" turret holes with a drill press, as opposed to using a hand drill or a Dremel.

Peace,
Al.

 

[gyraf]

The MEQ5 capacitor values are known working - they're rev'ed from a Tubetech unit - but If I remember correctly, the inductors are more or less guesswork, as my inductance-tester at the time was lacking to say the least.

And don't worry too much about "Q"'s - small differences in inductor resistance will only show in slightly different "max" settings for the bands.

Jakob E.

 

[jhaible]

[quote author="gyraf"]The MEQ5 capacitor values are known working - they're rev'ed from a Tubetech unit - but If I remember correctly, the inductors are more or less guesswork, as my inductance-tester at the time was lacking to say the least.

And don't worry too much about "Q"'s - small differences in inductor resistance will only show in slightly different "max" settings for the bands.

Jakob E.[/quote]


Thanks, Jakob - so I will start with the C values as given in your drawing.
Just to be sure, the L coming only in two steps for the High Frequency Boost section with 5 frequencies in the MEQ5, is from the Pultec original, too?

I'm asking this because of the Qs. I have once simulated the Presence section of that famous Neve channel, where the lower bands have all their own L and C values, but the higher bands just share one 200mH inductor with only the capacitors altered. Looking at the resonance peaks at fool boost, it's clearly visible how the lower bands have the same bandwidth (in log f scale), while the higher bands get sharper as the center frequency is increased. I have not enough experience to decide why this is musically usefull, or if it is useful at all, but this got me thinking about certain choices in classic circuits that have a reputation for sounding good.

Can't say how imprtant (or not) this is for the pultec. But I have just prototyped that high/low EQ and liked the sound so much that I'm now looking at the midrange EQ as well.
And I've aquired a taste for winding my own coils on RM8 cores of late. (;->)

JH.

 

[jhaible]

Looking at the high frequency boost section of the MEQ5,
as shown in http://www.gyraf.dk/gy_pd/pultec/meq5.gif

So I'm taking the capacitor values in the schematic as given,
and asuming the series resonance of the LC combinations were
aiming to be close to the peak frequencies on the front panel
(with some rounding and tolerance, of course), I get this:
jwL + 1/jwC = 0 @ w=2PI*f_peak

5kHz (given) 1nF (given) 1H (calculated)
4kHz 2n2 800mH
3k 3n9 720mH
2k 8n2 770mH
1.5kHz 16n 700mH

Observations:

a) very different from the L values in the schematics.

b) The 3 lower bands can certainly share one L value.

c) Even with one L value of 800mH for all 5 bands would come
quite close to the nominal frequencies:
1.4kHz
2kHz
2.9kHz
4kHz
5.6kHz

Of course the Cs can be chosen to hit the nominal f even better,
but I'm assuming them as given (easy to read, easy to measure),
and looking for the L values that would most likely have been used
in the original.

This leaves more questions than it answers.
Apparently a considerable Q variation was accepted in order to
be able to use very similar L values for all 5 bands. But then
why bother with 2 inductors at all, and not just use a single value?

JH.