Neumann W492 equalizer DIY project (PCB docs added)

[TwentyTrees]

In theory yes, it should be no more complicated than that. In practice, there may be factors at play that I’m not aware of (potential for increased noise?) - maybe spend some time comparing the W492 schematic to that of other 5 band SVF-based equalisers to see if they are doing anything different.

There’s quite a good discussion at https://groupdiy.com/index.php?topic=67088.0 about 5 band SVF EQs, worth having a look through that and reading up elsewhere on the SVF to get a sense of how well it would work.

Or just breadboard it. 

 

[winner]

In theory yes, it should be no more complicated than that. In practice, there may be factors at play that I’m not aware of (potential for increased noise?) - maybe spend some time comparing the W492 schematic to that of other 5 band SVF-based equalisers to see if they are doing anything different.

There’s quite a good discussion at https://groupdiy.com/index.php?topic=67088.0 about 5 band SVF EQs, worth having a look through that and reading up elsewhere on the SVF to get a sense of how well it would work.

Or just breadboard it. 

Thank you very much for your patient reply. Have a nice day

 

[winner]

In theory yes, it should be no more complicated than that. In practice, there may be factors at play that I’m not aware of (potential for increased noise?) - maybe spend some time comparing the W492 schematic to that of other 5 band SVF-based equalisers to see if they are doing anything different.

There’s quite a good discussion at https://groupdiy.com/index.php?topic=67088.0 about 5 band SVF EQs, worth having a look through that and reading up elsewhere on the SVF to get a sense of how well it would work.

Or just breadboard it. 

Hello, I may need to trouble you again. I used f = 1 /(2 * pi * R * C) to calculate the value of specific R, but the result is different from my experimental value. I think it is because the contact resistance is 560 . I hope to use band switch. Can you tell me the calculation method

 

[Palmito]

BanzaiMusic have these cheap Omeg pots
https://www.omeg.co.uk/wp-content/uploads/2018/10/PC16BU.pdf
Are they OK  ? (apart from the space between pins)
They dont have the dual-gang tho...
thanks in advance

 

[ToBSn]

BanzaiMusic have these cheap Omeg pots
https://www.omeg.co.uk/wp-content/uploads/2018/10/PC16BU.pdf
Are they OK  ? (apart from the space between pins)
They dont have the dual-gang tho...
thanks in advance

Yes, these pots will work fine.
But they won't fit in 1HU if I remember correctly.

6 x https://www.omeg.co.uk/wp-content/uploads/2018/10/PC16BU.pdf
2 x https://www.omeg.co.uk/wp-content/uploads/2018/10/PC2G16BU.pdf


ToBSn

 

[Palmito]

Thanks

i found even cheaper at buerklin, and they have the dual-gang

https://www.buerklin.com/en/Products/Passive-Components/Potentiometers-and-trimmers/Potentiometers/Stereo-conductive-plastic-potentiometer%2C-2-x-10-K%2C-0-25%C2%A0W%2C-Solder-pins/p/75E730

 

[erquere]

I found those ones : https://www.rapidonline.com/Catalogue/Product/65-1442

Just need to bend carefully the rear pins with little pliers to make the stereo fit in the pcb holes, i just still have to measure the distance between two pots to check it fits the faceplate I can buy here in Europe

 

[winner]

Excuse me, I want to increase Q value and change bandwidth. What should I do?

 

[erquere]

Question about ground wiring : do I have to leave the pcbs output ground pins unconnected, as usual ? If using shielded wires, do I have to connect thos ground pins to ground pins of the output XLR or - in my case - to the Lundahl transformers ground pins ?

Noob question I know...

 

[TwentyTrees]

Winner - I missed a couple of your messages here:

Hello, I may need to trouble you again. I used f = 1 /(2 * pi * R * C) to calculate the value of specific R, but the result is different from my experimental value. I think it is because the contact resistance is 560 . I hope to use band switch. Can you tell me the calculation method

Yes, I also found that the experimental value was different from my calculations. If you're looking for specific frequency choices, it's probably worth breadboarding and testing things.

Excuse me, I want to increase Q value and change bandwidth. What should I do?

https://groupdiy.com/index.php?topic=28332.msg897659#msg897659 - see the section on Variable Bandwidth. There might be better methods, but this definitely works.

Good luck!

 

[andow]

Just getting started with this and reading schematics I noticed the output stage is different than the original. Two 5532 in the original while this is using only one (and also omits some filtering).
Any insight what might be the reason for this?

 

[lars on]

Hi makers !

I just finished my two channels EQ ! Happy !
Everything seems ok but... on one channel the High Mid band is not working at all and
on the other channel it is the Low Mid band and the High band. Not working at all.
I use Green Omeg pots, didn't tested them yet.
No calibration have been done yet.

Any suggestions is very welcome.

Thank you everyone !

(I'll send pics soon !)

 

[Whoops]

Everything seems ok but... on one channel the High Mid band is not working at all and
on the other channel it is the Low Mid band and the High band. Not working at all.

Any suggestions is very welcome.

Review all the solder joints, look for shorts
Review all Lytic caps orientation
Review all resistor values
check opamp orientation, change opamps
Touch components with chopstick

Build yourself an audio/signal probe, start tracing it

Post pictures here

 

[lars on]

Thank's Whoops,

On Right channel the High Mid OpAmp U6 pin 1 was shorted with Ground !
So, everything is working now on this channel.  Happy !

On the Left Channel the + pin of Cap C6 was cutted. So, Low Mid is alive ! Happy !
High Mid is alive as well, there was a short between R39 and Ground.

And the last High band is in da place, I just forgotten a solder point at R46 (leaving the circuit opened)... hohoho...
:
So, Very very Happy !!

It's the first time I make double layer PCB.
The is so much thin and close tracks. The most of my problems came from here.
Maybe I'll make new PCB with wider tracks... one day.


I'll post pics tomorrow.

 

[Whoops]

I'm Happy you got it sorted

I don't know which soldering iron you're using, for this kinf of work I always use this one, and recommend it:

JBC 14ST

https://www.jbctools.com/2192040-14st-soldering-iron-product-457.html

It has a nice thin tip, and it's 11W, perfect for this kind of work without overheating components

 

[lars on]

Mine is this one https://fhrparts.com/microsoldering/lodning/cgx-936d-soldering-iron.html
It's thin but maybe not as much as yours.

Thank's for the link !

 

[lars on]

About measurement now :

For Filter 1 test, I'm reading maximum 2,35 Volt (+9,6 dBu) on output and not 4,35 Volt (+15 dBu),
as I should read.

Any hint please ?

When all switches off, there is 0,775 Volt on iput and output.

I've seen that  but I'm balanced :

With unbalanced connection (and without the modification) you should do the adjustment so that the maximum gain is +9dB (not +15dB).

 

[andow]

Variable bandwidth
10K pot (RP) with 3k3 series on CCW (RB2) across U3A (low mid) and U5A (high mid), wiper to ground via 330R series (RB3). This involved running wires to one side of each of R16, R17, R26 (low mid) and R31, R32 and R41 (high mid) - I left little loops in the resistor leads to allow this when populating the boards. One caveat - the centre frequency wanders a little as the bandwidth is changed from wide to narrow (e.g. a few Hz at +15dB at 1k), so there may be a better way to implement variable bandwidth on this circuit, but after using it extensively on a variety of material it doesn't feel like a major issue. And ideally I would have preferred rotary switches to pots for stereo matching, but front panel space didn't permit that so I matched the pots instead as best I could.

I tried this and found that it only increases bandwidth, the narrowest setting is identical to the original.
Is there a way to decrease the bandwidth? (I played around with some resistors but didn't come up with something useful.)
Also, did you use a linear pot? There is hardly any change after 10 o'clock, so I guess a log pot would be more suitable here.

Really incredible build, btw! 
And thanks for sharing your findings!

 

[TwentyTrees]

I tried this and found that it only increases bandwidth, the narrowest setting is identical to the original.
Is there a way to decrease the bandwidth? (I played around with some resistors but didn't come up with something useful.)
Also, did you use a linear pot? There is hardly any change after 10 o'clock, so I guess a log pot would be more suitable here.

Hmm, that's really interesting - it's definitely a pronounced change on my build, max low mid / high mid bandwidth increased to 0.45 Q (just under 3 octaves) and minimum decreased to 3.6 Q (1/3 octave). Original Q was around 0.7 or so (more or less 2 octaves).

I spent some time testing various values of RB1-3 and slugging the pot all kinds of ways, measuring the results in REW, and I ultimately preferred how a linear pot worked here. That said, this is a bus / mastering EQ for me so I do tend to use it more with wider Q settings. Having reviewed my notes, if I was looking to get a narrower bandwidth / higher Q I'd start by increasing RB2 in relation to RB1 and seeing what happened there. Let us know what you find!

Really incredible build, btw! 
And thanks for sharing your findings!

Thanks! Still the build I'm most proud of.  ;D I do need to pop her open and tweak the Q values in the HPF and LPF, as they're a bit Chebyshev and I think something more smoother would work better here for bus work.

 

[TwentyTrees]

I've had a chance to re-check the Q range this morning - here's the full range at +15dB 460Hz in the low mid band (ignore the slight wibbliness at top and bottom of the frequency spectrum, that's a problem with the calibration file I'm using in REW - the W492 is flat as a pancake otherwise):

Yes, you're right that the range is a bit cramped at the far end of the rotation, and that would explain why the pot normally ends up in the 6-11 range for me.  ;D  The way I have things wired, inverse log would work better than log to spread things out a bit, but that entirely depends on how you hook things up and label your panel. Of course, tracking between two log pots (inverse or otherwise) is going to be worse than between two linear pots, so stereo work would be less precise - rotaries would have been my preferred choice, but I just couldn't make room in a 2U build!