La2a Response Sweep

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

cayocosta

Well-known member
Joined
Jun 3, 2004
Messages
262
Location
SW Florida USA
Check this out.

la2a_sweep.jpg


It's a sweep of my la2a. The top is without gain reduction, and below are various levels of GR.

Interestingly, I had to open C4 all the way up 380pf to get the response flat without GR. Otherwise, the highs trailed off by about 15k! With and without compression - not good I would think.

With GR there's obviously some dips. 1 to 2 dB.

I traced the GR voltages of everything all the way back to but excluding the 12ax7 in the sidechain circuit and could not find the origin of the dips. So, I suspect it might be worked out somewhere around the 12ax7 setup.

Winston has said that UA sets C4 all the way up by the way.

Kent, you mentioned having scoped your la2a, can you elaborate on what you learned?

The sweep software is pretty cool, and there's a free version that just times out after several runs - but it's still worth getting. I found it through someone's post here - thanks.

It's here:

http://www.sumuller.de/audiotester/
 
There's an R-C network in the sidechain. I don't have the scematic handy, but I recall it looked like a de-Essing network: threshold lower for tones around 6KHz.

We also see the limited bandwidth of the sidechain, letting the ends bump up a bit.

None of this is really relevant to "normal" use with music/speech, except the de-essing tendency will make speech a bit nicer.

In general it is poor form to sweep tones through a limiter. You probably are not measuring what you think you are. Run some big pink noise through and see if it comes out flat.

The 380pF either controls de-essing (which you usually want) or compensates droop in the transformers.
 
Instead of sweeping the device with a rapid tone burst, try playing a constant tone at a constant (known) amplitude, at various settings of the reduction pot. It's tough to get a good frequency sweep when there is gain reduction going on, because the device is now non-linear, and traditional frequency response techniques break down (remember that the amount of gain reduction is a function of the attack timeand the length of time that the signal was over threshold).

I remember a while back that Universal had a little graph on their site that demonstrated this frquency dependance, but I can't seem to find it now. With a known dbu input signal of a constant frequency you should be able to measure the output and derive the gain reduction. Then, plot a graph of gain reduction versus input frequency. that's what I'd do!

Cheers,

Kris
 
BUMP

I just wanted to see if anyone else has any NEW sweeps from an original or reissue to add.

The post from Cayocosta was invalueble to me concerning C4 adjustments & what he found. I just swept mine and also found that C4 needed to be very close to maxed out to get an aprox. 2db roll-off starting from about 2-3k to 20k. If I moved it even a nano-nut, it had a pretty severe roll off. More like 10-15db from 2-3k to 20k. :shock:

What should the factory roll-off look like?

*part 2*
Concerning the detector circuit:
Does the cap (C-14) value determin the freqency or amount of????????
I've reserched this a little & found answers like; I don't use C-14, use 100-150pf or 330pf.

Obviously these values/opinions are all over the place and must have an effect on how the detector works. Im concerned & on a mission to get this right, especially after seeing how C4 effected the overall responce.
I will try and post some sweeps of my LA-2A, showing how the cap effected mine. But I still need to learn how to test and adjust C-14.:evil:

There must be a way to set this without guessing.

HOWWWWWWWWWW?

Thanks, Kevin
 
Anything with an adjustment screw can be tweaked by ear.
Remember that people love the LA2 for vocals.
Add that up and see what you get.
 
Back
Top