SSL 9k5 - Alternative Input Gain Selector - Thoughts/Ideas?

[frejo89]

Hi there,

Iam about to build my second SSL 9k5 Preamp, because Iam very convinced in this thing

But this time Iam thinking of a modded input gain.
more precisely of having a 12-way stepped gain selector, realized with a rotary switch with corresponding resistor values.

Background is that I can't find an alternative for this "nasty" 2k2 pot rev log, except from AML.
They now have a minimum order value and with that I would have to pay almost 75 pounds for the pot and other products to reach the minimum order value, with shipping and customs clearance, and that is ridiculous - but unfortunately there is simply no other supplier for these pots.

Maybe it is even possible to change some other resistor values on the pcb, so that you can have another pot value for the gain - but I dont know..

The considered stepped gain option seems to be a nice alternative for this preamp.

but I'm faced with a big question:
How do I find out the values of the resistors for the 12 positions from 0 Ohms on position 0 to 2,2k at twelfth - especially because the volume should increase audibly consistent with every step.

Logically, I can't simply calculate 2200 ohms divided by 12..

Are there any Ideas for an approach on this? Is this even possible to realize?
Thanks a lot for your help, thoughts and ideas on this!

kind regards

jonas

 

[ccaudle]

My brain is a little too scrambled this week to whip this out right away, but the collective "we" can probably help you work through what is needed.
Do you know how to generate the gain equation for the circuit? If you can do that we can work through the math on how to convert the linear gain values to gain in dB, and make a small script to generate values for steps in various dB values.

 

[ccaudle]

I'm not familiar with the schematic, so while searching I found this old thread which discusses exactly what you are asking:
https://groupdiy.com/threads/gain-switch-for-9k-mic-pre.4731/

 

[GrannyGremlin]

If you can find a nice 2.2k linear pot, you can use a tapering resistor to get a rev log curve. You could even play with the resistor value to tweak the taper to your preference. Worst case Tayda has 2.2k linear Alphas (carbon track pots aren't that bad when they're linear, but it would be out of place in an SSL copy I suppose).

A lot of older consoles used linear faders with tapering resistors to get log response because log faders (especially at the time) didn't track well unit to unit - linear is much more consistent.

 

[frejo89]

I'm not familiar with the schematic, so while searching I found this old thread which discusses exactly what you are asking:
https://groupdiy.com/threads/gain-switch-for-9k-mic-pre.4731/

oh wow, I did not find this thread. Thank you !

 

[frejo89]

If you can find a nice 2.2k linear pot, you can use a tapering resistor to get a rev log curve. You could even play with the resistor value to tweak the taper to your preference. Worst case Tayda has 2.2k linear Alphas (carbon track pots aren't that bad when they're linear, but it would be out of place in an SSL copy I suppose).

A lot of older consoles used linear faders with tapering resistors to get log response because log faders (especially at the time) didn't track well unit to unit - linear is much more consistent.

to find a decent 2,2k in linear is no problem =)

Iam also interested in this! How will this work with a tapering resistor? I never did this before. You have some documentation for this?

Thanks

 

[GrannyGremlin]

to find a decent 2,2k in linear is no problem =)

Iam also interested in this! How will this work with a tapering resistor? I never did this before. You have some documentation for this?

Thanks

http://www.geofex.com/Article_Folders/potsecrets/potscret.htm

the bit about tapering resistors starts about halfway down. The math to get the R value needed is dirt simple.

 

[Rob Flinn]

I think (yeasr ago) I tried using a tapering resistor with a linear pot in this sort of configuration & it didn't work too well. I think this is because it works when you feed the signal into the top of the pot & take the signal off the wiper, but the 9k doesn't use the rev log pot like this it only uses 2 of the connections & therefore you don't get the log law happening. You could just work out the resistors values you need to make a log switched attenuator & then hook them to the switch in reverse.

 

[frejo89]

Ok so I might go into troubles with this?

 

[abbey road d enfer]

If you can find a nice 2.2k linear pot, you can use a tapering resistor to get a rev log curve.

Wrong. Slugging resistors are good only for potentiometers connected as voltage dividers.
In the 9k, the pot is connected as a rheostat (aka variable resistor).

 

[abbey road d enfer]

Logically, I can't simply calculate 2200 ohms divided by 12..

Of course not. You have to use the gain formula.
I don't know what a 9k5 is, but in the 9k, the gain of each stage is 1+2k/Rg, Rg being the pot value+27 ohms.
At max gain, the pots are at zero ohms, so the gain of each stage is 1+2000/27=75 or 37.5dB, and the global gain 75dB.
At min gain, the gain of each stage is 1+2000/2227=1.89 or 5.5dB, for a global gain of 11dB.
If you're satisfied with this range of 64dB, you can split it in 11 steps of a little less than 6dB, or choose another arrangement.

 

[frejo89]

Thank you all for your suggestions!
with 9k5 i meant some kind of short term for the SSL 9k Preamp in 500 Series format.

let me put it this way:

Normally I have absolutely no problem using the original part intended for its purpose.

But it really makes me upset that it seems to be such an unique one and there only seems to be one supplier that offers a 2.2k rev log pot.
And I don't agree paying such a high price for it - due to minimum order values and so on - if there aren't some alternatives.

I even asked OMEG directly to make an offer. There is also a minimum order number there and it is simply not profitable.

There is also practically no provider left where you can order custom pots for maybe a little less money.

So if anyone has an idea for an alternative. Please feel free to share =)

 

[molke]

The link to Harpo’s spreadsheet in post #32 in the above thread still works for me; it probably doesn’t get much easier than this.

Once upon a time, I used the attached version (no guarantees), based on PRR’s idea in the same thread, but quality dual switches certainly haven’t gotten any cheaper since then.

 

[frejo89]

The link to Harpo’s spreadsheet in post #32 in the above thread still works for me; it probably doesn’t get much easier than this.

Once upon a time, I used the attached version (no guarantees), based on PRR’s idea in the same thread, but quality dual switches certainly haven’t gotten any cheaper since then.

looks good. If I get this right in this case I will need a 2x12 rotary switch - whereas lorlin for example just provides 1x12 or 2x6 ones as far as I know.

So you are really right with saying, that its a truly expensive approach as well - in terms of a 2x12 rotary

 

[Dylan W]

looks good. If I get this right in this case I will need a 2x12 rotary switch - whereas lorlin for example just provides 1x12 or 2x6 ones as far as I know.

So you are really right with saying, that its a truly expensive approach as well - in terms of a 2x12 rotary

If you’re comfortable soldering/desoldering SMD, you can go with one of the DACT style stepped attenuators on Ebay. Quite inexpensive and 21 steps. They seem like a great solution for preamp gain controls, in addition to things like stepped frequency pots for EQs. I have another thread going - I’ve been trying to find the supplier/factory for raw parts, maybe there could be a group buy there, but no luck so far.

https://groupdiy.com/threads/dact-style-21-step-smd-attenuators.87926/

But in the meantime, they’re cheap, so you can buy a few, desolder the resistors it comes with, and add the appropriate values for a 2k2 rev log pot.

 

[beatnik]

You could get the same potentiometer sold by AML directly from Omeg UK.

 

[Rob Flinn]

You could get the same potentiometer sold by AML directly from Omeg UK.

That's what I did years ago when the 9k project first came out. I did a group buy at the time.

 

[Rob Flinn]

looks good. If I get this right in this case I will need a 2x12 rotary switch - whereas lorlin for example just provides 1x12 or 2x6 ones as far as I know.

So you are really right with saying, that its a truly expensive approach as well - in terms of a 2x12 rotary

Make sure you use an MMB rotary switch or you might find you get some large clicks when switching gain.

 

[abbey road d enfer]

But it really makes me upset that it seems to be such an unique one and there only seems to be one supplier that offers a 2.2k rev log pot.

Just to make sure we're talking about the same thing, the unobtainium part is a DUAL 2.2k RevLog, isn't it?
Well, I think AML's offer is hard to beat, unless you're prepared to place a MOQ order with Ohmeg.
You may want to try setting up a group buy...
Now you can use on of these. You would have to recalculate the resistors, but it seems like the cheapest source for a 23 position rotary switch;
You may even find a source for the switch alone.
Or your original idea of a 12-position.
https://www.ebay.com/itm/2845286958...44p3HYMUaVpPxcpTLoPuoi0hd7|tkp:Bk9SR6zllbSVZA

 

[amplexus]

Commenting so i remember to come back later- but I am sure I had an online calculator that would tell you the resistor values you needed to calculate a log or rev log taper switch. You enter the number of steps, and the total resistance and it spits out the values.

If I’m remembering it correctly it should work for this configuration. I will do a look today.