Limiters & VU meters

Where we discuss new analog design ideas for Pro Audio and modern spins on vintage ones.
carlmart
Posts: 169
Joined: Mon May 19, 2014 6:11 am

Re: Limiters & VU meters

Post by carlmart »

Has anyone tried or listened to the NJM2762 used as a limiter?

http://www.cn.njr.com/PDF/NJM2762_E.pdf

I wouldn't use it as it's intended to, to control both channels, but just one chip for each channel.
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: Limiters & VU meters

Post by JR. »

I would not be very optimistic about this... looks like a part to be built inside a cheap digital recorder.

The bottom line is that sound quality will be dominated first by the side chain (att/rel/etc) and only secondarily by the gain element.

So any one of several options you have chewed on could be fine, or suck depending on execution.

JR
Cancel the "cancel culture", do not support mob hatred.
ricardo
Posts: 262
Joined: Sun Apr 10, 2011 7:24 am

Re: Limiters & VU meters

Post by ricardo »

carlmart wrote:Has anyone tried or listened to the NJM2762 used as a limiter?

http://www.cn.njr.com/PDF/NJM2762_E.pdf
Well its cheap & simple to use. I'll second JR's caveat about sound quality and I've seen some really awful limiters on cheap recorders.

But ALL solutions need you to work out a GAIN STRUCTURE and do Listening Tests

Why don't you try it and tell us what you find? It might be just what you want.

Only 2 twiddles so you'll quickly get it to be as good as it can be.
carlmart
Posts: 169
Joined: Mon May 19, 2014 6:11 am

Re: Limiters & VU meters

Post by carlmart »

Yes, we plan to do exactly that: to try them and see how they behave, both in audio quality as in limiting effectiveness.

My idea is to design the prototypes with a socket for that stage, so we can plug-in different limiter options and listen to them. The more transparent should win.

I'm working on three of the THAT projects: the one I showed here, the one on DN03 and the one on DN04.

For the chip based type, implementation might be important. I think it's better to use two ICs, even if they are stereo. So the limiting of one channel does not influence the other.

BTW, for the first limiter I showed here, there were some details THAT advised me to correct, so now it looks like this.

There's still one thing they still recommend, which is to choose between the two detections the limiter is using: RMS detection or peak level detection. The former might be better, they say.

So I'm considering ways to eliminate peak detection, which apparently depends on the capacitance you hook onto CT, pin 4 on the 4315. Capacitance shouldn't be small, even if I thought 10uF was small.

But on DN03 (which is the one you think I should use) that capacitor is 220uF.

One important thing on this design THAT had sent to me is because it was designed for single supplies, and all the others on their ANs are using split-supplies.

I'm trying to find the THAT engineer that designed that limiter to see how the threshold, attack and release settings were done, as it's quite likely I would have to adjust them to my application.
Attachments
THAT#1.limiter.jpg
THAT#1.limiter.jpg (263.02 KiB) Viewed 14513 times
emrr
Posts: 570
Joined: Sun Mar 02, 2008 10:27 pm
Location: NC, USA
Contact:

Re: Limiters & VU meters

Post by emrr »

RMS detection has preset attack and release, the notes detail this. With the typical timing cap it's slow, over 20mS attack time. It won't stop transients. The non-linear cap is the best of both worlds, stopping transient peaks much more effectively but acting slowly on non-transient sources, with release more resembling AGC.

There's a visual comparison of RMS and NL over here:

viewtopic.php?f=6&t=281
Best,

Doug Williams
Electromagnetic Radiation Recorders
carlmart
Posts: 169
Joined: Mon May 19, 2014 6:11 am

Re: Limiters & VU meters

Post by carlmart »

emrr wrote:RMS detection has preset attack and release, the notes detail this. With the typical timing cap it's slow, over 20mS attack time. It won't stop transients. The non-linear cap is the best of both worlds, stopping transient peaks much more effectively but acting slowly on non-transient sources, with release more resembling AGC.

There's a visual comparison of RMS and NL over here:

viewtopic.php?f=6&t=281
Which notes are you talking about? And which would be the typical timing cap in what circuit?

Which is the non-linear cap you are talking about?
carlmart
Posts: 169
Joined: Mon May 19, 2014 6:11 am

Re: Limiters & VU meters

Post by carlmart »

OK, I might be getting at what your are suggesting about the non-linear capacitor, replacing the capacitor on CT with the circuit suggested in Figure 6 of DN03. Is that so?

My question is on which circuit I should do that: on the one I enclosed above or on the one DN03, which I'm still not too convinced is the right one for this application.

The problem with that limiter for power amps still is the 4 trimpots I would have to adjust for each channel. Not too practical for a production line.
carlmart
Posts: 169
Joined: Mon May 19, 2014 6:11 am

Re: Limiters & VU meters

Post by carlmart »

On my last chat with THAT support they came with a different suggestion for my limiter, and I'm trying to work on it.

Instead of my first considered design, which I already published here, and the one you suggest, starting from DN03, they recommend to start from circuit on DN115 and work out some changes on it, which might fit my application better.

What they suggest to do on the page 3 circuit on DN115 is get rid of the variable attack/release circuitry, and have the non-linear cap section (U3B and its surrounding R/C/D) replaced by a simple capacitor for RMS-based time constants. Then have the hard-soft switch permanently wired to "hard", and the ratio pot replaced by fixed resistors to generate high above-threshold compression ratio.

I have enclosed the original circuit and a partially modified schematic.
Attachments
THAT#1a_limiter.jpg
THAT#1a_limiter.jpg (58.91 KiB) Viewed 14496 times
THAT#1limiter.jpg
THAT#1limiter.jpg (61.15 KiB) Viewed 14496 times
carlmart
Posts: 169
Joined: Mon May 19, 2014 6:11 am

Re: Limiters & VU meters

Post by carlmart »

Well, I got to a limiter design for my mic preamp, using the affordable and single-supply ready THAT 4315, along with some MC33178 to help out.

I'm not sure yet how you do to set to a fix 20.1 threshold or how to handle the gain trim.

But I think I'm getting closer. Any suggestions or comments?

I wonder why, and I asked so, why THAT does not have just a limiter design, instead of compressors/limiters, on their ANs.
Attachments
THAT#3.limiter.jpg
THAT#3.limiter.jpg (315.65 KiB) Viewed 14470 times
carlmart
Posts: 169
Joined: Mon May 19, 2014 6:11 am

Re: Limiters & VU meters

Post by carlmart »

Probably not up to this forum's standards, but has anyone built and listened to the limiter design on ESP?

http://sound.westhost.com/project67.htm

I wonder if it's better, the same or worst than the NJM limiter chip I mentioned.
Post Reply