All things LA2a related

[ruairioflaherty]

Hello everyone,

long time listener first time caller. I bought an LA2A from a fellow forum member last week and it arrived today. It's a lovely build on turret board but unfortunately it didn't make the trip unharmed. Basically the signal path is fine but there is no compression. When I turn the gain reduction knob to 10 I get a faint distortion on the audio.

I have devoted all of my learning time here to solid state so I'm a bit a valve idiot but I spent the evening pouring over various schemos and getting a grip on what goes where. I swapped the 12AX7 to no avail.

No for the newb questions. I would like to measure my voltages to compare them to CJ's voltage chart to see where I stand -

1) Is that the best voltage ref chart to compare to?
2) What signal level should I put into the unit for testing? Should I aim for 4V AC at 200Hz on pin 7 of the A-10?
3) When measuring the output level of 0dB is that between pins 2 & 3 on the output xlr or between pin 2 and ground?
4) Finally am I correct that all voltages are measured referenced to the ground/chassis?

I seem to be losing my sidechain signal after C11 but I'm not sure what kind of voltage I should be seeing on pin 5 directly? The chart says 30V after C11 but I'm down in the millivolts...

Once I get the measuring methodology straight I can report back with some hard facts. Do the 6AQ5A tubes break down easily?


Appreciate the help guys.

Thanks,
Ruairi

 

[dripelectronics]

that faint noise when turning the peak reduction could indicate the
stereo adjust pot and the frequency response pot need to \
be turned full clock wise.

if no compression , then turn counter clockwise full counter clock wise,

check your source , make sure its a hot signal going into the compressor.

one of the cells in the t4b may have fallen away from the light panel , you might open it up and inspect.

g.

 

[ruairioflaherty]

Hi Gregory,

thanks for your input. I've popped open the T4B and all seems to be in order. In my tests last night I tried the stereo adjust pot and frequency response pot in all position and while it did affect the tone and character of the distorted audio it did not begin to compress.

I think there is something wrong in the sidechain, even with a healthy signal going in to the machine and the peak reduction turned way up I am getting only millivolts at the input of the T4B, CJ's voltage chart shows 30V AC after C11 which tallys with my own understanding of what it takes to drive the EL panel.

Can someone confirm that I should be measuring all of my AC and DC voltages relative to ground? And also what kind of input signal I should drive the unit with while measuring these voltages - How about 200Hz 0dBat 0.775V? I'll be able to come back with some voltages around the sidechain and perhaps you guys can help me isolate the issue.

Thanks,
Ruairi

 

[ruairioflaherty]

Hi everyone,

I did a little bit more with this tonight. I took out the T4B today and cleaned the pins, I also had a probe around with a chopstick while powered down to make sure all of the solder joints and connections were good (and they were).

Tonight when I powered up the unit was compressing! Unfortunately when the compression kicks in there is no bass. I sent white noise at a healthy level to the unit. I watched on a spectrum analyzer as I turned up the peak reduction and I got 20dB level drop at 500Hz but practically nothing at 16K (gentle roll off in between).

So...the line amp works well but when I compress I lose low end. Any leads? I've been searching all evening but the info is everywhere - tough going indeed.

Cheers,
Ruairi

 

[dripelectronics]

ah at least this is working ,
i was thinking it was the cap that connects the
6aq5a to the t4b , (c11) and was going to reccomend
replacing it ,

swap the 12ax7's around , see if that works .

does metering gain reduction work ?

this is exibiting grounding problems ,
not saying that it is , but if you unground one of the input or
out put transformers , you lose low end.

why it only happens on reduction is interesting.

take a pic , perhaps it needs a little cleaning up .

the circuit is pretty simple , so it can't be too deep .

look over the back of the pot's and their connexions .

solder bridges can do this too.

keep on looking ,

g.

 

[ruairioflaherty]

[quote author="dripelectronics"]ah at least this is working ,
i was thinking it was the cap that connects the
6aq5a to the t4b , (c11) and was going to reccomend
replacing it , [/quote]

Hi Gregory,

I was starting to suspect that cap my self and intended lifting a leg to check it last night but then I had compression!

swap the 12ax7's around , see if that works.

I've done this already - no difference when I do.

does metering gain reduction work ?

Yes, it seems to be working well. If I watch the signal going into an out of the compressor I can see the low end rolling off as I compress but the high end stays untouched.

this is exibiting grounding problems ,
not saying that it is , but if you unground one of the input or
out put transformers , you lose low end.

why it only happens on reduction is interesting.

This is the key to solving my issue I think. The line amp is flat without gain reduction. What can happen during GR to lose low end, it must be in or around the T4B no?

take a pic , perhaps it needs a little cleaning up .

The build is really clean, done by a fellow board member on tag strip. I've been examining all of the connections I can see and they seem to be good. The line amo is clean with very little noise.

the circuit is pretty simple , so it can't be too deep .
look over the back of the pot's and their connexions .
solder bridges can do this too.
keep on looking

Will keep looking. I was trying to take a short cut by buying a ready made unit and the DIY gods have punished me :green: . I'm learning lots in the process though.

g.

 

[ruairioflaherty]

I'm making progress on my issue. As I mentioned above the unit is compressing but is losing low end as compression increases - 16K remains untouched and compression increases all the way to the low frequencies.

So I figured the problem must be around the T4. Last night I pulled out the T4B and replaced the LDR in the circuit with various resistors to simulate compression. Even with heavy gain reduction the signal was flat so the problem is in my T4B.

The T4B is homemade as per CJ's page and is very well done. I've checked all of the connections and everything looks fine. It appears that the LDR is acting as a kind of low pass filter and shunting more lows to ground. Anyone ever seen this?

I might try swapping the meter and signal LDRs tonight and I'll order a new T4B today.

Cheers,
Ruairi

 

[AnalogPackrat]

[quote author="ruairioflaherty"]
So I figured the problem must be around the T4. Last night I pulled out the T4B and replaced the LDR in the circuit with various resistors to simulate compression. Even with heavy gain reduction the signal was flat so the problem is in my T4B.[/quote]

Very interesting.

The T4B is homemade as per CJ's page and is very well done. I've checked all of the connections and everything looks fine. It appears that the LDR is acting as a kind of low pass filter and shunting more lows to ground. Anyone ever seen this?

I would suspect the EL panel before the LDRs. The reason is that the side chain drives the EL panel and if you are seeing any freq dependent change in compression its hard to see how it could be caused by the LDR which is simply changing resistance based on the EL panel brightness. The EL panel looks like a capacitor electronically speaking, which also jibes with the freq dependent nature of your problem.

Do you know if the EL panel is from the "official" source or maybe something pulled out of a night light?

Here's a simple test you can do. You apparently have a signal generator, so set it up to feed the input of the LA-2a. Remove the cover of the T4b and plug it into your unit. Position yourself where you can see the EL panel. Fire everything up and make sure things are happening...then turn off the lights. Now sweep the signal generator frequency from low to high. Does the EL panel light up more or less at different frequencies?

I might try swapping the meter and signal LDRs tonight and I'll order a new T4B today.

Yeah, that might an interesting thing to try as well. Does your meter (set to read GR) reflect what's actually happening or is it indicating compression that isn't occurring?

Good luck with it!

A P

 

[ruairioflaherty]

[quote author="ruairioflaherty"]The T4B is homemade as per CJ's page and is very well done. I've checked all of the connections and everything looks fine. It appears that the LDR is acting as a kind of low pass filter and shunting more lows to ground. Anyone ever seen this?[/quote]

[quote author="AnalogPackrat"]I would suspect the EL panel before the LDRs. The reason is that the side chain drives the EL panel and if you are seeing any freq dependent change in compression its hard to see how it could be caused by the LDR which is simply changing resistance based on the EL panel brightness. The EL panel looks like a capacitor electronically speaking, which also jibes with the freq dependent nature of your problem.[/quote]

I understand what your getting at but I don't think the EL panel is the problem. My problem isn't how the compressor is reacting to signals but actually what happens to the signal during compression. Watching on the analyzer with a steady white noise input I can see the low end roll off as I turn up the gain reduction knob. The unit is only compressing low and mid frequencies.

If my EL panel was flawed in some way (not lighting up for low frequencies for example) it would affect when the compressor kicks in but not what it does during compression, ie. shunt a portion of the full range signal to ground.

Do you know if the EL panel is from the "official" source or maybe something pulled out of a night light?

It's a proper LSI panel and the T4B build checks out perfectly.

Here's a simple test you can do. You apparently have a signal generator, so set it up to feed the input of the LA-2a. Remove the cover of the T4b and plug it into your unit. Position yourself where you can see the EL panel. Fire everything up and make sure things are happening...then turn off the lights. Now sweep the signal generator frequency from low to high. Does the EL panel light up more or less at different frequencies?

I ran your test with the generator and the panel does indeed react more poorly to low frequencies but this is mostly caused by the cap in series with the panel. This cap is a deliberate inclusion in the design to reduce low frequency triggering of compression. Bypassing this cap with a jumper yields a more even (but not perfect) light response from the panel. Overall the panel reacted pretty much as I would have expected.


[quote author="ruairioflaherty"]I might try swapping the meter and signal LDRs tonight and I'll order a new T4B today.[/quote]

Yeah, that might an interesting thing to try as well. Does your meter (set to read GR) reflect what's actually happening or is it indicating compression that isn't occurring?

I tried swapping the LDRs around in the unlikely event that the signal LDR had developed a crazy fault. This didn't change anything. I've checked all grounds around the machine and particularly around the start of the circuit. I've traced the circuit up to and around the first calve stage and down to the sidechain and everything seems to be in order.

The key test for me is that when I replace the T4B LDR with a fixed resistance like 10K I get perfectly flat gain reduction. If the resistance in the LDR was interacting with something else in the circuit (a hidden inductor for example :grin: ) why would that effect/interaction disappear when I use a fixed resistor to simulate the on state?

BTW The meter seems to be reading accurately enough.

Good luck with it!

Thank you sir, appreciate the help and I'm learning plenty about these beasties. :thumb:

Cheers,
Ruairi

 

[3nity]

This is weird i had the same problem with a Forsell opto compressor.
this was only compressing kicks in some monos mixes...
Why? i dont know maybe some of you knows..

 

[AnalogPackrat]

Well, if you can pull one of the CdS cells out of the T4b, try measuring its resistance and capacitance (in the dark) while shining a flashlight (a.k.a torch) on it. The resistance should go below 100ohms (maybe as low as a few ohms) under bright illumination and should get into the 100k+ range within a couple of seconds in "pretty dark" conditions. Capacitance should be low and I wouldn't expect it to change much. I've got an assortment of CdS cells at home--will try to remember to measure capacitance on a few when I get home tonight for comparison purposes.

<Edit> Think-o. So the CdS is shunting the signal which would mean if it were the cause of your problem it would have to have series inductance, not capacitance. Duh.
<End Edit>

Another somewhat more painful test you can try is to disconnect one leg of the EL panel, plug the T4b in sans cover. Run it in the dark and modulate the compression manually with your trusty flashlight (torch). Do you still see the strange frequency dependence?

CdS has a wavelength dependent resistance, but its response is pretty broad band only falling off below 500nm and above 650nm or so. The EL panel also has an excitation frequency dependent emission wavelength. But these effects are pretty mild and probably can't explain what you're seeing.

I don't suppose you've got any other CdS cells laying around just to experiment?

<Edit>
Also, you used a fixed 10k resistor in your experiment. That's pretty high. I wonder if you would see anything different (with regard to freq response) with 1k or even 100ohms. This is more like what the CdS would be under moderate to heavy compression. It might help narrow down the source of the odd behavior. I still have a hard time seeing how a CdS cell could cause what you're seeing.
<End Edit>

A P

 

[marshman]

Hi guys,

I could use a little help. I have a pair of la2a's built the same time using drips v2 boards. Both have been working for over a year, no problems.

One of them has developed and oscillation problem (sudden onset), that is only present when the t4 is in. It passes signal fine with no t4.

I am seeing an intermittent 100+vac on pin 3 of the t4. I do not see this on it's twin.

If I am reading the schemo correctly, this is from pin 5 of the 6aq5a?

I have swapped tubes, t4's and chopsticked. All the voltages on the other tubes seem fine.

I am kind of at a loss as to what could cause this ac in this part of the circuit? Any ideas?

Thanks!

mm

 

[marshman]

Ok, I think I see where the ac is coming from, but is 100+ too high?

 

[marshman]

I found the problem. Intermittent ground on the gain pot!!

mm

 

[idylldon]

I built two of Greg's LA2A, one with the original PCB and one with the 2.0 version. Both have worked flawlessly until today, when one of them quit compressing. The gain worked, but I got no compression at all. I traced the problem to a burned-up R34, which was a 2-watt metal film. I checked my other unit, and it's obvious that R34 is running hot in that one as well.

My questions:

1.) Since two units are experiencing the same thing, and I've checked both 6AQ5 tubes to make sure they're OK, my guess is that metal films aren't quite up to the power handling task here even though they're of the correct wattage. Seems they should be, but in practice it doesn't appear to be the case. Are metal films not known to be good in power-rail applications?

2.) I have some 10K metal oxide (they're at least 3-watt) resistors and I was going to use two in series because I don't have any 22K metal oxide power resistors and I need to use both LA2As for mixes this weekend. Any problems using 20K here instead of 22K?

Thanks,
--
Don

 

[compproject]

hi guys, i am new to the forum.

i have read about the bloo kit and some other
based on some PCB.
is any of them still available? i would prefer
complete kits with transformers VU and the case
and front plate so that i can start working as soon
as i receive the kit.

thank you for any info on that respect.
btw, i would be interested in a 1176 kit, too

best

comp

 

[emrr]

to my knowledge there are no complete kits available, only various parts such as circuit boards and cases. 

 

[RedNoise]

for 1176 see mnats 's sticky ! in the black market . ;D

 

[compproject]

thanks guys,

so, there are no bloo kits available any more?

yeah, i have seen beautiful cases, do i guess
a complete electronics part kit would do, if that exists?
what about the optocouplers? i thought these were
difficult to source...

 

[JCT Audio Designs]

Hey guys, I just found this area of the forum and should have been posting here the entire time, but I need some help. The link to my previous posts is below. Here is my problems.

Hey all, I need a little bit of help. I am building an LA2A clone. I have double checked everything and it all looks good, but I am having a couple of problems.


#1 the zero control barely does anything. When I am in gain reduction mode, with the zero control all the way over, the meter will only go up to -10dB.

#2 When I turn the peak reduction control, the signal is gets distorted, otherwise the sound coming through the compressor sounds great.

#3 if I remove the t4b the meter reading on gain reduction does not change. I pulled out each tube one by one, then all the tubes and none of the tubes effects the gain reduction metering. I have tested all the tubes and they all test ok.

With a 1k tone going through it, when I measure the AC voltage at the junction of the 2-68k resistors at the output of the input transformer, I get 9v. If I turn up the gain reduction to 5 as suggested by the CJ layout, nothing happens. When, I did the same on a real la2a, the voltage dropped to what CJ said it would. So clearly, my gain reduction circuit is not working correctly. Can someone explain it, I am having trouble understanding the signal flow. What actually powers the T4B?

Nothing seems to fix my problem. I am hoping someone has had similar problems. Thanks

http://www.groupdiy.com/index.php?topic=31132.0