3D "AIR" EQ - "Night EQ" PCB's Complete!

[chrissugar]

The sure way to determine what type of pot is, is to use an ohm meter.
Put the pot in exact mid position and measure the resistance between wiper and the left end then between the wiper and the right end.
1-If values are equal, there is a linear pot.
2-If the left side is 10% of the total value of the pot and the right side is 90% it is a log pot.
3-If the left side is 90% of the total value of the pot and the right side is 10% it is a reverse log pot.

chrissugar

 

[khstudio]

[quote author="peterc"]

Is a CCW Audio pot a Reverse Audio Taper

Probably not, the taper is usually specified after the value. You said the value was 500kA, the A value is log taper usually. O dont know where the CCW comes from. Is it printed on the pot?

Peter[/quote]

I looked up the Bourns part #'s & codes & that is what it said.

It sure sounds like reverse taper...

The part # on the Bourns POTS are:
91A1DC24T23

The part # on the Switchable high shelf is:
91C1DD20T23

Please check it out... they have a PDF with the codes... Mouser, Digikey.

Kevin

 

[MrZpliff]

I looked up the Bourns part #'s & codes & that is what it said.

It sure sounds like reverse taper...

The part # on the Bourns POTS are:
91A1DC24T23

The part # on the Switchable high shelf is:
91C1DD20T23

Please check it out... they have a PDF with the codes... Mouser, Digikey.

Kevin

I agree, as I looked up Bourns-site too, I'm pretty shure a CCW-pot is a reverse-taper pot...
the Gyraf Calrec EQ uses reverse audio pot's.
I haven't checked but it may wor´k in a similar way :?

 

[khstudio]

I hope they're available without too much BS :?

When it comes to the film CAPs... what do you think we should use :?:

I JUST noticed an "AV" on the side if one of the films... Brand maybe?

I wish I knew the real type but I agree with Peter, If we use good caps it will be fine :thumb:

I STILL would like to know what the purpose is of the 10watt 150 ohm resistor from the positive 18v rail to GND??? :?
You can see it on the original PCB file NOT the schematic!

Kevin

 

[chrissugar]

decoding the 91A1DC24T23

91-MODEL-single turn in line PC pins
A-BUSHING-metal plain 3/8"
1-SECTIONS-single
D-ANTIROTATION LUG-no lug
C-SHAFT TYPE-1/4"
24-SHAFT LENGTH-3/8"
T-ELEMENT TYPE TAPER- CCW audio
23-RESISTANCE-500Kohm

So this is a 500K revlog pot

decoding the 91C1DD20T23

91-MODEL-single turn in line PC pins
C-BUSHING-metal plain 1/4"
1-SECTIONS-single
D-ANTIROTATION LUG-no lug
D-SHAFT TYPE-1/8"
20-SHAFT LENGTH-5/8"
T-ELEMENT TYPE TAPER- CCW audio
23-RESISTANCE-500Kohm

this is also a 500K revlog pot but with different mechanical dimensions.

I have the feeling that it will be easyer to make this with a rotary switch (12 position Lorlin) with resistors than finding 500K revlogs.

 

[MrZpliff]

[quote author="chrissugar"]
I have the feeling that it will be easyer to make this with a rotary switch (12 position Lorlin) with resistors than finding 500K revlogs.[/quote]

I think that's a good idea if you want to use it as a stereo-master EQ...

 

[khstudio]

I RE-UPLOADED THE PCB WITH THE CORRECTIONS
(in the above post) HERE's a link also:
http://www.twin-x.com/groupdiy/albums/userpics/3d_air_eq_-_Top_Q_v1_fixed.jpg
http://www.twin-x.com/groupdiy/albums/userpics/3d-air-eq---Top-Q-v1sw.jpg

Yea, the "AIR" pot has a tiny shaft...that's all.

About the stepped gains:
I NEVER went higher than about 2:00 or too much lower than 10:00 on the gain controls.

The air only goes up & I've went as high as HALF... no more.

OK, I'm packin a bowl & headin out in the snow to the shop...I got etching to do :green:

If you guys spot any errors in the Re-Posted PCB (high shelves only vers)
PLEASE let me know SOON before I print it.

Also, I don't have the Inductors for the output... hopfully thy won't matter for testing :?

Thanks,
Kevin

 

[clintrubber]

Keven, I'm curious to which program you are using for the schematic.
Drawing-style of that software looks neat. :thumb:
Are there any direct transfers to a simulation program possible perhaps?


Then this:

Please allow me some thinking out loud w.r.t. the circuit
(v2 of the first page):

When the EQ is engaged, all bands are in parallel and have a
max gain of '20dB minus something' (so that'll result in the
15dB of the max boost). Outside that region each section is unity gain.

So the 'EQ-engaged but flat' setting is realized by all those
500k no-center-detent CCW-pots in their mid-setting, a situation
that likely won't ever exactly be the case.

That'll be fine of course, if you want flat then simply don't insert or
engage the EQ.
And reportedly this is a mighty fine sounding EQ so why worry ?

But it's surprising me a bit for a device that's also intended as a mastering EQ.
I mean, the mastering-version has the switched level-pots so
exact levels are realizable.
But do all those EQ-bands also line up decently even when the level-pots
are optimally adjusted ?

Has anybody done a simulation yet for the complete circuit ?
For a 'flat-but-EQ-engaged' setting I expect there'll still be a slight ripple,
especially in practice when there are unavoidable component tolerances.

As said, it won't matter but it came to mind when trying to understand
what's happening here in this circuit.



Cheers,

Peter

 

[chrissugar]

I had exactly this thought Peter.
I was expecting the pots to be linear and at exact mid position the resitance to be the value for 0dB, but with antilog pots it looks like there is no chance with the eq bands engaged, but at 0B position to have a flat response.
This is one of the reasons why I think about a switch. For mastering with a 12 step Lorlin, 0.5dB/step you can cover 6dB (in mastering normaly there is no need for more) For mixing,with 1dB/step you can cover 12dB (plenty enough)

chrissugar

 

[clintrubber]

[quote author="chrissugar"]I had exactly this thought Peter.
I was expecting the pots to be linear and at exact mid position the resitance to be the value for 0dB, but with antilog pots it looks like there is no chance with the eq bands engaged, but at 0B position to have a flat response.
This is one of the reasons why I think about a switch. For mastering with a 12 step Lorlin, 0.5dB/step you can cover 6dB (in mastering normaly there is no need for more) For mixing,with 1dB/step you can cover 12dB (plenty enough)

chrissugar[/quote]
Hi ,

Chris, those switches look like a nice approach. 'Unfortunately' they need to be 'CCW' indeed, just like the pot in an API 312-type of circuit needs to be for controlling gain.

But that's only one thing - even with optimally aligned controls there still could be that 'ripple' that you see for instance when all bands of a graphic EQ are full-on or full-off: They don't realize a ruler-flat response.
And the EQ of this thread might have it already in its neutral-setting.

Again, FWIW all this - since many ears have already confirmed this is a very nice EQ (so to #&*% with technical objections).
But obviously we're here not only to get some boxes to make music, we're technically interested as well. So because of that I brought this up.

Cheers,

Peter

 

[khstudio]

Good points... I never thought about the taper & the mid position... I've been so wrapped up in just getting the circuit right for us to learn & build.

The more expensive version used stepped gains (EQ3)
Try doing a search for these & there should be some of the original articals & tech stuff somewhere so you can get a better idea what it's all about.

I'm just happy I got to this point. :green:

I finished the photo process on my boards this mornig & I'm heating the tank for etching now so if I have enough energy well see if it works pretty soon.

I don't have the exact pots but a linear should work for testing...
Same thing with resistors... I may be missing some values but I'll just string some together for now if I have to.

Hey, a simple "OFF" switch could shut the bands off too NO :?:

I'll tell ya, when all the pots are straight up & you flip the bypass on & off there isn't much change if at all that's noticable.

Kevin

 

[khstudio]

Keven, I'm curious to which program you are using for the schematic.
Drawing-style of that software looks neat. Heck Yeah!
Are there any direct transfers to a simulation program possible perhaps?

(KEVIN) :wink:

www.expresspcb.com (it's free)

I'm not sure if you guys have the latest files or any at all but go get that software & you can print & etch yourself if you want. You NEED the software to do it. (single sided - green traces are where the jumpers go.)

Peter (Green pre) DID run some simms & said that it all worked.

Kevin

 

[clintrubber]

[quote author="khstudio"] I've been so wrapped up in just getting the circuit right for us to learn & build.[/quote] :thumb:

Try doing a search for these & there should be some of the original articals & tech stuff somewhere so you can get a better idea what it's all about.

Do you know if someone has already scanned some stuff out of the manual ? I saw that manual on eBay coming by, but it has been closed by now.

I'm just happy I got to this point. :green:

All fine ! I mean, you're doing all the hard work and we here sit back, enjoy a cigar and mumble a bit about the circuit :twisted: :green: :thumb:

Hey, a simple "OFF" switch could shut the bands off too NO :?:

You mean disconnecting some sections from the summing point ? No, that will interrupt that frequency region.
As I understand the circuit it NEEDS the 'neutral'-setting of the CCW-pot to have 0dB for a certain frequency range.
Disconnecting a section completely will punch a hole in the freq. response, even a bit more than the max-cut would do.

I'll tell ya, when all the pots are straight up & you flip the bypass on & off there isn't much change if at all that's noticable.

Kevin

'Straight up' = in the neutral position ? Could well imagine that; the ripple might be measurable and noticable in very controlled listening environments, but for about all other cases little influence might be experienced. Which means that the combining of the bands is done well, but I was surprised that they took that approach.

Cheers,

Peter

 

[chrissugar]

[quote author="clintrubber"]
Chris, those switches look like a nice approach. 'Unfortunately' they need to be 'CCW' indeed, just like the pot in an API 312-type of circuit needs to be for controlling gain.
[/quote]

Don't forget that this is not a potentiometric circuit (divider), it is a variable resistor, so there is no need to make a chain of resistors. You can use one separate resistor for each step, which is much easyer. :grin:

chrissugar

 

[clintrubber]

[quote author="chrissugar"][quote author="clintrubber"]
Chris, those switches look like a nice approach. 'Unfortunately' they need to be 'CCW' indeed, just like the pot in an API 312-type of circuit needs to be for controlling gain.
[/quote]

Don't forget that this is not a potentiometric circuit (divider), it is a variable resistor, so there is no need to make a chain of resistors. You can use one separate resistor for each step, which is much easyer. :grin:

chrissugar[/quote]
Indeed ! Either as say 12 separate resistors to a common point or as an incremental chain.
I've briefly been thinking about an easy trick to get 12 boost & 12 cut positions from one cheap Lorlin by means of adding a 'boost/cut'-switch but that won't be too easy I'm afraid because of the 'bent law' required :?

Bye,

Peter

 

[chrissugar]

[quote author="clintrubber"]

Hey, a simple "OFF" switch could shut the bands off too NO :?:

You mean disconnecting some sections from the summing point ? No, that will interrupt that frequency region.
As I understand the circuit it NEEDS the 'neutral'-setting of the CCW-pot to have 0dB for a certain frequency range.
Disconnecting a section completely will punch a hole in the freq. response, even a bit more than the max-cut would do.
[/quote]

And this is the exact reason why you can't build an EQ like this with just one or two bands. You need to have all the bands to leave the rest of audio untouched.

Kevin, you can do a test.
Desolder the resistors in series with the pots of each band except the air band, and listen what it does.
it would be interesting to do this with each band, and listen to it's sound.

chrissugar

 

[khstudio]

x

 

[clintrubber]

[quote author="khstudio"]Also, do you think I/we will have to compensate (impedence/load) when removing some bands like I have :?:

Kevin[/quote]
As it looks like so far they all NEED to be there, so there's no issue of compensating missing bands... as far as changing impedances go...
You could of course 'correct the topology' for missing bands but then better duplicate the whole circuit I'd say.

On the other hand, I can see how the air-band could be used as a stand alone thing when used in say a subgroup insert. You do your main-mix as usual and route some channels to the subgroup as well. The subgroup-faders control the added air.

Or even better yet, use it like you would a reverb, so with FX-sends from the channels. Receive the result on an FX-return.
(Check topology for signal-polarity).

Bye,

Peter

 

[khstudio]

I think you guys are right... & I thought it myself

Please look at the schematic for me & tell me how to make the FIXED 2.5k band FULL band & cover everything under the "AIR" (switchable shelves)

At least I'll have that which is the best part anyway. (this is just for now with the boards I just made.

Kevin

 

[khstudio]

x