LA-4 Help Thread!

[helterbelter]

Thanks...

The Tantalums.... I haven't compared the parts with the schematic. I just went with the BOM that Luny made. But I have a bag with 6.8 uF tantalums, so, maybe I'll replace them later .... or maybe not..... We'll see. It's not finished yet, haha. The threshold pot acts a bit strange, so I've got to have a look into it anyway. 

The meter tracks in the opposite way. Works pretty good actually. The module is going to be placed in my console, so that's the reason why I placed the meter under the knobs. But nevermind the frontpanel anyway, it's just temporary. I only wanted to get it working, and I don't mind the cosmetics yet.

The NSL32... The dot is placed at the cathode pin, so it has to face the markings "opto 1"and "opto 2".  The short legs are the led, the long are the photocell, so, the led legs are put in the inner circle, the long legs outside the outer circle (no polarity, but I think they both have to go the same direction).
http://www.alliedelec.com/Images/Products/Datasheets/BM/SILONEX_INC/Silonex-Inc_Actives-and-Passives_6995014.pdf

 

[geoff004]

I began to finish up my LA4 today - that is to say it's all together but not working as it should.
It sounds good when it's not compressing.  Once the threshold gets turned up the issues begin.
I'm constantly learning about all this stuff and I've got a feeling that a lot of my problems are power related. 
The power transformer is 20VA / 20volts.  Both PSUs are stuffed, with one driving both LA4s and the other driving relays for bypass and LEDs.
When I turn up the threshold the high end drops and the unit distorts.  Also it seems to compress most when the ratio is set to 2:1.  When I set it to anything else it doesn't squash as much - and all the other setting react the same.

Is it just grossly underpowered?

Here are some pictures.
The chassis is Par-metal.  Why ever bother with anything else?
The front was done with the Super77 method.  What's amazing is that any imperfections go away once it's clear-coated.
The transformers are EA2623-1 from ClassicAPI.  Meters are from a forum member - they work beautifully.  Knobs were some of the cheapest that Mouser carries.
The bypass LEDs are multi-color - they light up amber when the unit is overloaded.
The insides will get cleaned up when the issues get taken care of.
I've actually got a twin in the works - just waiting for meters.

 

[Luny Tune]

Look in the manual on page 23 (pdf! The number on the scanned page says 15.) Note that the 2:1 ratio does behave differently.

Also try this basic adjustment of the Threshold and Ratio trimmers: Threshold=9 o'clock, Ratio=10:30. (Remember they're positioned opposite and facing eachother so they're also viewed from opposite angles. )

 

[geoff004]

Look in the manual on page 23 (pdf! The number on the scanned page says 15.) Note that the 2:1 ratio does behave differently.

Also try this basic adjustment of the Threshold and Ratio trimmers: Threshold=9 o'clock, Ratio=10:30. (Remember they're positioned opposite and facing eachother so they're also viewed from opposite angles. )

Yup-
the trimers made a big difference.
Today I'll get a better idea of how it's working - I'll post my results.
thanks!

 

[helterbelter]

I began to finish up my LA4 today -
.....
Here are some pictures.

Hi geoff, your la4 looks great !

 

[Luny Tune]

Oh yeah, Geoff. It looks great!

 

[geoff004]

Well, I set the trimmers to what seems to work for the ratio and threshold.  Mixed with it on piano all day and I have to say I like it more than my 1178.  It still need a little tweaking.  I'm tempted to swap out the LDR - I'm currently using the NSL-32.  I want to see how the 6910+LED work. 
I might try some different op amps too - I just went for what was cheap.

Any hint on setting the 'common mode adjust' trimmer?

 

[okgb]

helter thanks for the opto advise

 

[Luny Tune]

Any hint on setting the 'common mode adjust' trimmer?

Check the manual.

Don't know if I got it wrong actually. I thought that what you should do was to send signal in on both + and - and invert one of them. Check with a meter or scope on the trimmer's wiper and adjust for best possible cancellation.

But the manual says something about measuring noise level @ 60 Hz and 20 kHzwhen shorting + and - compared to a reference signal level at 2 kHz.

Maybe the result is the same... I can't see why my method shouldn't work as intended.

I'm so glad you agree with me that it rocks. ;D

 

[chrissugar]

Looking at your PCB it looks like you have two opamps unused in TL074-2. You left the pins not connected to anything which can be a source for problems (unstable?). A good practice for unused opamps is to connect the non inverting input to ground and connect the inverting in to the output.

chrissugar

 

[Luny Tune]

Thanks for the tip.

 

[geoff004]

Looking at your PCB it looks like you have two opamps unused in TL074-2. You left the pins not connected to anything which can be a source for problems (unstable?). A good practice for unused opamps is to connect the non inverting input to ground and connect the inverting in to the output.

Please correct me if I'm wrong - IC2 pins 5 and 10 go to ground and pins 6 and 9 go to the output?
I'm assuming when you say 'output' you mean the + side out.
Thanks

 

[chrissugar]

No, not to the output of the whole circuit. I'm talking about each opamp's output.
One opamp is pins 5,6,7 and another opamp is pins 8,9,10.
Pin 5 to ground, pin 6 connected to pin 7.
Pin 10 to ground, pin 9 connected to pin 8.

chrissugar

 

[geoff004]

No, not to the output of the whole circuit. I'm talking about each opamp's output.
One opamp is pins 5,6,7 and another opamp is pins 8,9,10.
Pin 5 to ground, pin 6 connected to pin 7.
Pin 10 to ground, pin 9 connected to pin 8.

So basically I'll solder some nice bridges between 6 and 7 and 8 and 9, then jump 5 and 10 to pin 12 (which traces back to ground).
Thanks! (again, please let me know if I'm off the mark)

 

[chrissugar]

So basically I'll solder some nice bridges between 6 and 7 and 8 and 9, then jump 5 and 10 to pin 12 (which traces back to ground).
Thanks! (again, please let me know if I'm off the mark)

Yes, it looks correct.

chrissugar

 

[geoff004]

I finally have a little extra time to try to iron this thing out.
I found that 1 of the LDRs in by build was bad - it was stuck with a resistance of 6K ohms.  With the resistance at 6K one of the meters wouldn't register properly in gain reduction mode.
Here's where it's at now - it sounds pretty good at 2:1 and 4:1 ratios.  When the ratios are set higher the audio becomes distorted as soon as it hits the threshold.  At 20:1 it's pretty ugly and totally unusable.
The threshold and ratio trimmers are trimmed full clockwise - the distortion gets much worse when they're turned at all.
I used almost all metal film 1/4 watt resistors (aside from a few 1/2 watts I had lying around).

I haven't had a real LA4 in my possession in about 6 years so I don't really remember all the characteristics of the unit.  I doubt the original distorts like mine does.

Aside from the op amps mentioned in the BOM (I think they're TL074 CNs, not CPs), all the parts are pretty much as stated (maybe not brands, but certainly values).
Any ideas where to change out components, or maybe what to check?

 

[Luny Tune]

I'm intensely busy right now, so you'll have to find them yourself.

The LA-4 DOES distort at higher ratios and more extreme settings. That's both why so many people love it and why many hate it. What you hear is totally useless is to me absolute MAGIC on bass and drum ambience.

I once said to my dad that it was a little irritating that the steering wheel in my car was shaking when I went over 140 km/h... His response was:"Stay below 140, idiot!" ;D

Anyways, make sure you adjust the units correctly and then compare it to the test files I posted here on the forum somewhere. Search for keywords: Blackface, silverface, 20:1 and so on...

 

[helterbelter]

http://www.groupdiy.com/index.php?topic=30780.0

 

[barclaycon]

Does anybody know what the Vactec device is that JBL now supply as a replacement for the LDR cells in the LA-4 ?
They give the part number as HM13-0499-02.

Is it a VTL5C4-2 ?

Thanks

 

[Luny Tune]

I believe it is. I tried it and it does work.