Gain switch for 9k mic pre?????

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

jeroddumas

Well-known member
Joined
Jun 19, 2004
Messages
264
Location
georgia, USA
Could someone help me with the values for a gain switch for the SSL mic pre. I think I would need a double pole switch, correct? Or can I use another value pot...????
 
> will a 5k log pot work??

A dual 5K audio (preferably reverse-audio) will "work", though not much will happen in the left end of rotation.

> I would need a double pole switch, correct?

Correct.

> the values for a gain switch

Yes, a switch may be easier to get than this oddball pot. But computing gain-set resistors is hard work, and quadruply hard with the 2-stage design.

......off the top of my head, these emitter-emitter values will work:

1st stage: 27, 38, 54, 76, 110, 151, 220, 300, 440, 1K, 2K

2nd stage: 39, 54, 80, 110, 160, 220, 320, 440, 1K, 1.5K, 2K

That should give you 6dB steps, more or less, covering about 72dB to 20dB. Steps will be a little short at the high-gain end, and a bit erratic at the lowest gains, but usable.

Those are the effective emitter-emitter values. The actual resistors will be less, because we want to use a Shorting switch and series the resistors. So for the 1st stage the actual resistors run:

27, (38-27)=11, (54-38)=12, (76-54)=22, (110-76)=34..... (2K-1K)=1K
 
> the values for a gain switch

Actually.... I think the design was compromised to use the dual-pot. Consider: assuming a 10% taper pot, at half-rotation the gain is about (2.25K/0.25K)^2 or 9^2 or 81 or 38dB. A not uncommon setting for a hot mike. But the input noise resistance is over 250Ω, so the noise figure is over 4dB.

If instead of changing both gain-sets the same, we leave the input stage at high gain as long as possible and reduce gain in the output stage, down to about 2 (so the output stage always clips first) we could get a total gain of 81 with gain of 40.5 in the input stage. That gives input gain-set resistor about 51Ω, much lower than 150Ω source, near ideal noise figure at medium gain. Noise figure won't rise much until overall gain of around 30dB.

It also saves a few precision resistors and makes the arithmetic a little easier. {three pencils later} arrrr, not that much easier. I hadda break down and ask a Pentium. Don't trust these answers:

1st stage: 26, 26, 26, 26, 26, 26, 52, 107, 225, 500, 1330, 8000

2nd stage: 39, 80, 170, 370, 900, 3300, 3300, 3300, 3300, 3300, 3300, 3300

Gains:
pos__1st,dB__2nd,dB__total,dB
_12___37.8____34.4____72.2
_11___37.8____28.3____66.1
_10___37.8____22.1____60.0
__9___37.8____16.1____54.0
__8___37.8____10.2____48.0
__7___37.8____4.1_____41.9
__6___31.9____4.1_____36.0
__5___25.9____4.1_____30.0
__4___19.9____4.1_____24.0
__3___14.0____4.1_____18.1
__2___8.0_____4.1_____12.1
__1___1.9_____4.1______6.1
 
Thanks PRR,

Would these be the resistor values I need to use? I'm not that advanced into electronics, so I am assuming that these are the values. Also If I chose to use the 5K pot would I lose any DB of the output?
 
shameless bump. I want to get the ssl 9k running. Waiting on the mat-02 and transformer. So I need to get this answered. Gain switch or 5k pot. If i use the pot will I lose any db?
 
[quote author="PRR"]I hadda break down and ask a Pentium. Don't trust these answers:[/quote]

:green:

Q]How many Intel Pentiums does it take to change a light bulb?

A]1.000000016

Keef
 
working schematic

> ......off the top of my head, these emitter-emitter values will work:
1st stage: 27, 38, 54, 76, 110, 151, 220, 300, 440, 1K, 2K
2nd stage: 39, 54, 80, 110, 160, 220, 320, 440, 1K, 1.5K, 2K
That should give you 6dB steps, more or less, covering about 72dB to 20dB. Steps will be a little short at the high-gain end, and a bit erratic at the lowest gains, but usable.


Can someone please check my math?

> Those are the effective emitter-emitter values. The actual resistors will be less, because we want to use a Shorting switch and series the resistors. So for the 1st stage the actual resistors run: 27, (38-27)=11, (54-38)=12......wrong, (76-54)=22, (110-76)=34..... (2K-1K)=1K

jer> Would these be the resistor values I need to use?

FIRST STAGE
27 on PCB
38-27= 11
54-38= 16
76-54= 22
110-76= 34
151-110= 41
220-151= 69
300-220= 80
440-300= 140
1K-440= 560
2K-1K= 1K

SECOND STAGE
39 on PCB
54-39= 15
80-54= 16
110-80= 30
160-110= 50
220-160= 60
320-220= 100
440-320= 120
1K-440= 560
1.5K-1K= 500
2K-1.5K= 500

In picture form:

Gain-sw-1.gif


Gain-sw-2.gif
 
I found a stereo-pot logarithmic 2,2k on a german electronic shop.
Should this be ok for that use the ssl9k use?
(NB:in my mind, dual pot means stereo: 2x3pins, so if that "conception" is wrong, sure that pot will not fit)
It's not reverse, but i don't know what reverse means for a pot. Is it important?
:?: :?: :?:
i'd be glad if you could answer me

thanksfully
gilles (excuse my frenchglish)
 
[quote author="PRR"]
FIRST STAGE
27 on PCB
38-27= 11
54-38= 16
76-54= 22
110-76= 34
151-110= 41
220-151= 69
300-220= 80
440-300= 140
1K-440= 560
2K-1K= 1K

SECOND STAGE
39 on PCB
54-39= 15
80-54= 16
110-80= 30
160-110= 50
220-160= 60
320-220= 100
440-320= 120
1K-440= 560
1.5K-1K= 500
2K-1.5K= 500
[/quote]

How can i recalculate this for a 24 position switch? I assume one could not just fill it up with the "mean-values" for the in between steps since it's a rev log scale...

Was thinking of trying this one: http://www.uraltone.com/kauppa/product_info.php?products_id=1077
 
[quote author="PRR"]> the values for a gain switch

Actually.... I think the design was compromised to use the dual-pot. Consider: assuming a 10% taper pot, at half-rotation the gain is about (2.25K/0.25K)^2 or 9^2 or 81 or 38dB. A not uncommon setting for a hot mike. But the input noise resistance is over 250Ω, so the noise figure is over 4dB.

If instead of changing both gain-sets the same, we leave the input stage at high gain as long as possible and reduce gain in the output stage, down to about 2 (so the output stage always clips first) we could get a total gain of 81 with gain of 40.5 in the input stage. That gives input gain-set resistor about 51Ω, much lower than 150Ω source, near ideal noise figure at medium gain. Noise figure won't rise much until overall gain of around 30dB.
[/quote]

I don't really get this. This idea of having the two different stages work at different gains is not implemented in these values, right? This is just a normal rev log scale?

[quote author="PRR"]
FIRST STAGE
27 on PCB
38-27= 11
54-38= 16
76-54= 22
110-76= 34
151-110= 41
220-151= 69
300-220= 80
440-300= 140
1K-440= 560
2K-1K= 1K

SECOND STAGE
39 on PCB
54-39= 15
80-54= 16
110-80= 30
160-110= 50
220-160= 60
320-220= 100
440-320= 120
1K-440= 560
1.5K-1K= 500
2K-1.5K= 500
[/quote]

Then what are these values?

[quote author="PRR"]
1st stage: 26, 26, 26, 26, 26, 26, 52, 107, 225, 500, 1330, 8000

2nd stage: 39, 80, 170, 370, 900, 3300, 3300, 3300, 3300, 3300, 3300, 3300
[/quote]

It cant be the total resistance of the switch/pot since it's ending at 8000 and 3300 but the total should be around 2500R...???

I have some nice 2x24 Elmas' that could come to use here if i can manage to find the right resistor values. Anyone have an opinion wether there's anything to gain by this instead of using the dual carbon pots? Is it just a waste of nice switches? :?
 
Back
Top