elskardio

That 1512 gain adjustment driving me nuts!
« on: July 21, 2020, 11:29:22 AM »
Hi Guys,

I'm building a small utility mic preamp for a local studio. It's a simple That 1512 schematic (in attachment) and the gain adjustment is starting to drive me nuts!

I did a lot of testing with different values and most of the gain control happens between 200R and minimum 3R. Using a C5K potentiometer results in almost 75% of the pot rotation doing nothing.

Looking at That's Gain Table (see in attachment) I should get different results...

Any hint or tips?
Thanks




squarewave

Re: That 1512 gain adjustment driving me nuts!
« Reply #1 on: July 21, 2020, 12:37:12 PM »
Sounds like your pot is broken. Maybe it's not really reverse log? I just put a cheap C5K into a similar circuit (standard API 312 style circuit but that has the same gain equation) and it worked as expected. Yeah, the last 1/4 turn is more dramatic than the first 3/4 but the other positions do control the gain. If "75% of the pot rotation [is] doing nothing" is to be taken literally, then that sounds like your pot is bad. Double check by looking at the output level in a scope or listening to a tone.

Note that reverse log tapers are far from perfect. The slope of the taper changes around mid position. So gain control around that taper transition might be a little awkward in this particular circuit. If you want perfection, get a stepped switch and compute resistors yourself (but IMO that is overkill and a decent C5K is spot on for this circuit).

Another alternative would be to use C1K and then add a -20dB pad to the input.

Michael Tibes

Re: That 1512 gain adjustment driving me nuts!
« Reply #2 on: July 22, 2020, 06:10:26 AM »
Or maybe use a simple rotary switch for the gain setting, at least for evaluation?

Michael

Re: That 1512 gain adjustment driving me nuts!
« Reply #3 on: July 22, 2020, 01:29:34 PM »
I've seen pots that claim to have log law but are not logarithmic at all!, you should try and measure the resistance of your pot at different positions and plot in a spreadsheet to see the real pot law. You can also try tweaking the pot law by adding a fixed resistor between the wiper and one of the other terminals.

I've ploted the Pot Resistance VS Gain of the 1512 in a linear scale simulating a 5K pot in series with a 5ohm fixed resistor, as you can see, almost all of the action is in a very narrow part near the lowest resistance.



If instead, I plot with a log scale in the X axis, simulating a logarithmic pot, things look better, but still far from perfect. Maybe this is what you are experiencing, anything beyond 200ohms seems to make little difference, you won't get a fully linear adjust even with a reverse log. Using a smaller pot and adding a pad, like squarewave mentioned, can be a good solution, the ultimate best is using a rotary switch with fixed resistors.



A 500 ohm pot plus a 20dB pad, seems to be the best compromise, although finding a 500ohm reverse log pot its not that easy.



« Last Edit: July 22, 2020, 02:18:10 PM by Dualflip »

elskardio

Re: That 1512 gain adjustment driving me nuts!
« Reply #4 on: July 22, 2020, 02:18:00 PM »
Ok so I finally found a configuration where I'm happy with the overall gain adjustment.

First I swapped the 1512 for a 1510. Using the same C5K potentiometer, there's already more control of the gain at lower rotation.
Then after reading That's Design Note 138, I added a 5k resistor between the 1510's RG1 & RG2 pins. This lifted the Gain/Rotation curve a few dB higher.

After a lot of testing my conclusion is that the 1510 is a better choice if you're planning to use a potentiometer to control the gain. The 1512 is better suited for a rotary switch with fixed resistors.

Re: That 1512 gain adjustment driving me nuts!
« Reply #5 on: July 22, 2020, 02:21:22 PM »
Ok so I finally found a configuration where I'm happy with the overall gain adjustment.

First I swapped the 1512 for a 1510. Using the same C5K potentiometer, there's already more control of the gain at lower rotation.
Then after reading That's Design Note 138, I added a 5k resistor between the 1510's RG1 & RG2 pins. This lifted the Gain/Rotation curve a few dB higher.

After a lot of testing my conclusion is that the 1510 is a better choice if you're planning to use a potentiometer to control the gain. The 1512 is better suited for a rotary switch with fixed resistors.

The 1510 also has lower THD

squarewave

Re: That 1512 gain adjustment driving me nuts!
« Reply #6 on: July 22, 2020, 03:15:55 PM »
You can also try tweaking the pot law by adding a fixed resistor between the wiper and one of the other terminals.
That trick doesn't work with reverse log.

Re: That 1512 gain adjustment driving me nuts!
« Reply #7 on: July 22, 2020, 03:20:27 PM »
That trick doesn't work with reverse log.

I wasn't suggesting he used a reverse log, a linear will do.

squarewave

Re: That 1512 gain adjustment driving me nuts!
« Reply #8 on: July 22, 2020, 07:51:35 PM »
I wasn't suggesting he used a reverse log, a linear will do.
You can strap / slug a linear pot to make log but that technique cannot be used to make a reverse log from a linear pot.

abbey road d enfer

Re: That 1512 gain adjustment driving me nuts!
« Reply #9 on: July 23, 2020, 03:10:51 AM »
You can strap / slug a linear pot to make log but that technique cannot be used to make a reverse log from a linear pot.
Actually, the true answer is ; you can "slug"  or "strap" a Lin potentiometer to act as a Log or RevLog, but you can't slug a Lin pot to act as a Log or RevLog rheostat.
Strapping/slugging actually works for voltage divider arrangements only. Dun wok for 2-node circuits.
Who's right or wrong is irrelevant. What matters is what's right or wrong.
Star ground is for electricians.


abbey road d enfer

Re: That 1512 gain adjustment driving me nuts!
« Reply #10 on: July 23, 2020, 03:23:36 AM »
Looking at That's Gain Table (see in attachment) I should get different results...
This table is absolutely correct. the problem is the pot. So-called RevLog pots are actually two Lin tracks connected at mid-rotation.
The gain vs Rg formula shows that in order to have a relatively decent gain spread, the value at mid rotation should be about 300 ohms. That would be a C3 taper, which nobody makes in standard. Usually RevLog pots are C20, C15 or C10, the number after C representing the value at mid-rotation as a percentage of nominal value.
Manufacturers have to order custom pots in order to solve the issue. Maybe you can buy a spare part for an existing product...?
This is why many DIYers or boutique mfgrs resort to using a coarse switch and a compementary gain trim.
Who's right or wrong is irrelevant. What matters is what's right or wrong.
Star ground is for electricians.

JohnRoberts

Re: That 1512 gain adjustment driving me nuts!
« Reply #11 on: July 23, 2020, 08:57:38 AM »
+1 last century we tooled up custom pot tapers to smooth gain over the full range. IIRC our custom taper was something like 25k total resistance with reasonable adjustability down at the single digit ohms end.

So, for onsey twosy perhaps try to source a replacement pot from a manufacturer's service department, who has tooled up a similar full custom pot. 

JR

PS: Sorry no I do not remember part numbers but there were several in the Peavey system.
Don't only half-ass tune your drums. Visit https://circularscience.com to hear what properly "cleared" drums sound like.

elskardio

Re: That 1512 gain adjustment driving me nuts!
« Reply #12 on: July 23, 2020, 09:38:06 AM »
The gain vs Rg formula shows that in order to have a relatively decent gain spread, the value at mid rotation should be about 300 ohms. That would be a C3 taper, which nobody makes in standard. Usually RevLog pots are C20, C15 or C10, the number after C representing the value at mid-rotation as a percentage of nominal value.

Great info! thanks

I finished the little guy last night and took a few pictures  :)

It will be tested for the next week in the studio before I build more for their other rooms.
Cheers








Re: That 1512 gain adjustment driving me nuts!
« Reply #13 on: July 23, 2020, 10:16:56 AM »
Very nice work, what did you use for the silk screen on the front panel?

elskardio

Re: That 1512 gain adjustment driving me nuts!
« Reply #14 on: July 23, 2020, 11:13:15 AM »
Very nice work, what did you use for the silk screen on the front panel?

Thanks. Front & Back panels are made with PCBs

squarewave

Re: That 1512 gain adjustment driving me nuts!
« Reply #15 on: July 23, 2020, 12:29:40 PM »
Thanks. Front & Back panels are made with PCBs
That'll work. But you really should have copper pours grounded to the chassis.

elskardio

Re: That 1512 gain adjustment driving me nuts!
« Reply #16 on: July 23, 2020, 01:01:15 PM »
That'll work. But you really should have copper pours grounded to the chassis.
;)


squarewave

Re: That 1512 gain adjustment driving me nuts!
« Reply #17 on: July 23, 2020, 01:19:31 PM »
All good then! Nice work.

elskardio

Re: That 1512 gain adjustment driving me nuts!
« Reply #18 on: July 24, 2020, 08:51:36 AM »
So, for onsey twosy perhaps try to source a replacement pot from a manufacturer's service department, who has tooled up a similar full custom pot. 

Thanks for the tip. I just found 5k reverse pot used in Cameron guitar amps. Apparently they have a tolerance of 2.5%
I’ll report back once tested

abbey road d enfer

Re: That 1512 gain adjustment driving me nuts!
« Reply #19 on: July 24, 2020, 11:18:58 AM »
Thanks for the tip. I just found 5k reverse pot used in Cameron guitar amps. Apparently they have a tolerance of 2.5%
I’ll report back once tested
Tolerance is not the issue. When I mentioned C3, it's not the tolerance, it's the taper. I seriouslky doubt guitar amps use C3 pots.
Who's right or wrong is irrelevant. What matters is what's right or wrong.
Star ground is for electricians.


 

Related Topics

  Subject / Started by Replies Last post
9 Replies
1872 Views
Last post April 10, 2005, 09:34:59 AM
by VacuumVoodoo
3 Replies
2095 Views
Last post July 15, 2008, 08:14:08 PM
by pstamler
0 Replies
1906 Views
Last post January 21, 2009, 09:22:18 AM
by geoff004
5 Replies
653 Views
Last post December 20, 2018, 02:51:15 AM
by ruffrecords