GSSL HELP THREAD!!!

[gyraf]

..that marking is sometimes very weak, and often VERY hard to see if it says 5534 or 5532..

 

[quadwould]

Hello.. great thread of GSSL all knowingness.....

i have a (or many) issue(s)...

first off. in Bypass when sending a -18dBfs tone, i get an output of -12.. i know this is from the +6 there is in the unit. so that does not woryy me.

What does though is this: it compresses like an 11*76 in nuke mode,on steriods. When its engaged, both channels are really quiet, and with different settings, get even quieter. the right channel is higher in level (95%) of the time depending on what settings there are dialed in.
it varies , on average, from +6 higher to +22 over the left channel in level. But even with such differences the overall level is still reallllllly low. Like -36dBfs and lower most of the time. i have created an MS Excel spreadsheet that i can email anyone. it has the dB readings for the left and right channel with the makeup gain all the way down, and all the way up as well for each possible setting combination of attack and release when the unit is at a ratio of 2:1. i didnt measure the levels for when its was at 4 and 10 ratio because that was too tedious and i think the problem is not with the ratio as it seems to work.

another strange thing i sthat the higher the input level, the lower the output level.. it iss like its ducking exponentially.. :?

also, the threshold does not seem to be doing anything

and also the meter is pinned all the way up. its BehR****R meter. i put the 2K in the main pcb but it didnt seem to help. the 330 does not seem to help either.

what is the sensitivity of the BerDinger meter anyway?


any help would be very usefull.

i can email the excel file with the level readings to anyone upon request

also i have tried swaping the vcas and opamps from left to right and back again and nothing happened.

if you need to know voltages are specific places, let me know where

cheers guys

 

[EEMO1]

the 5534 was right. seems like i should get something to pins 2 and 6, but im not... tried to trace the lines to pin 2 and 6 but im not getting any readings.

 

[gyraf]

Inverted inputs are tricky to measure, as they're often "virtual-ground" points if feedback is applied (which is the case here)

 

[tree]

thanks jakob - really appreciated.
will do.

yes - sorry - should have mentioned that those (after input amp, before sidechain) resistor measurments were AC.

ability to measure signal with mult meter AC setting valuable info for the very green - personally speaking, of course! :green:

 

[tree]

jakob, trouble in the hen house!

before doing the VCA bypass test, i took a look at PIN 3 of L side VCA...
though all joints looked good, i followed the circuit off of pin 3 that led to ne5534 and found apparent solder bridge between ne5534 PIN 5 & 6.
took the chip out and cleaned up contacts, put back in... tested with 1k tone... NOTHING on Left side.... looked closer, in my excitment, i had oriented ne5534 wrong way around!
turned off unit.

reoriented ne5534 correctly.
turned on unit.
POOF. :shock:
smoke over ne5534 area.
turned off quick.

WHAT HAVE I DONE?????
WHAT SHOULD I DO????

 

[[michael]]

Deleted, wrong topic I think. :grin:

 

[csonics]

Ok, this question will totally indicate my inexperience in all of this but I'm just finishing up my gssl and I am not quite sure how the wiring goes to the in's and outs. I want to connect the in's to a balanced connection and the outs I want to make balanced as well but with an earth lift if need be.
Does anybody know the wiring scheme for that? Once again I know that this is totally a green question. Sorry for the ignorance.

 

[tree]

jakob....

okay,

i've replaced the ne5534 that i seem to have fried in L side VCA circuit.
no poof.
no smoke.
BUT
now when unit fed 1k test tone, L side output is even hotter than original problem. 15db hotter not "just" 9db.
AND
when compressing signal, thresh and gain make up pots now have only about 3 db effect on L side output, whilst R side seems to respond as it should.

any idea what ELSE this indicates i may have fried here?????

am sooooooo bummed.

where to begin???

 

[quadwould]

Squeak Squeak Squeak Squeak ... (goes the robot with only 1 wheel)

sounds like Tree and I are having similar problems.

read my above post on page 58 for some fun details. and PM or email me for the excel file

cheers



help....?

 

[tree]

Jakob;

BINGO!!! :thumb:

GOT IT!!

A LONG DAY OF TRYING, FRYING AND RETRYING, BUT ANOTHER GSSL IS BORN.

Jakob E. wrote:

Tree,

Your input circuit looks fine from here (If the said voltages are AC! - always tell that explicitly for reference..)

Try removing the two sidechain VCA's, and connect the input to the output on the - now empty - socket with a small piece of wire. That would be pin1 to pin8 - the two outer pins.

This rules out the VCA gains - now check your AC levels at the output - measure AC level referred to ground at both + and - out individually. All four outputs should be close to the same level.

If levels are still different, then check component values, shorts, and breaks at the output stages - that would now be the area where you have the problem.

If levels are now the same, then the small DC voltage at the control pin of one of the VCA's is the troublemaker. Check that area (including the related 5534) for shorts and component values

Jakob E.

because of your valuable suggestions (100X :thumb: s, Jakob!) followed PIN 3 trace of L side vca circuit, found a tiny solderbridge at the ne5534 socket.
BUT THEN...
wrongly reinstalled the ne5534 in my haste to test repaired socket - after the smoke cleared, i just replaced any suspicious burnt smelling bits with ones that were identical, but smelled better :wink:
seems that after my careless backwards chip reinstall poofed :green: , the that vca got ugly too, Left output got even uglier than original prob, but as soon as i swapped in a new 2180 and ne5534 etc....
we was rockin again!

Your GSSL is a beautiful thing -

THANKS AGAIN.

 

[gyraf]

:grin: Heey..! Glad that we got that one sorted, Tree..!


--


[quote author="csonics"]I am not quite sure how the wiring goes to the in's and outs. [/quote]

Csonics,

Look at the schematic. And look at the component overlay in the pcb-artwork .pdf-file. This is the only way to get a grip ow what the unit does.

And that should make it quite clear how to hook it up: In and out's hot and colds are marked + and - , 0/Ground is marked 0 (on the component overlay and in the schematic)

input xlr's female, output xlr's male

0 to pin1
+ to pin2
- to pin3

Jakob E

 

[EEMO1]

im getting really pissed. im going to desolder everything and start over soon.

can't find any shorts, wrong resistors or anything. im starting to hate to even look at the damn thing. :evil:

 

[quadwould]

Jakob!!!

Great Sounding Unit!!!!

my issues were all down to 2 tiny shorts with hairline solder bridges. ARGARGAHHHH.. Silly me for not looking closer....


But Damn.. what a great sounding piece of Kit


Now.. on to tame the threshold sensitivity.

and probably tame the output too.

 

[FrankSL]

Hi,
I've just built a mono version of the compressor, by simply leaving out all right channel components.

The unit seems to work fine except for a strange ratio behaviour: the scope trace of the output signal (with a 1V sine on input) with compression enabled show a weaker signal with 1:2 ratio than with 1:4, which is also weaker than 1:10.

Looks something is reversed... Am I missing anything?

EDIT: I've just found a thread where Jakob explains that this behaviour may be correct with a constant sine signal because the ratio setting changes both ratio and threshold, is this correct?

Thaks,
Frank

 

[gyraf]

Hi Frank,

Yes, that's right..

:thumb:

Jakob E.

 

[FrankSL]

Thanks Jakob, that problem was scaring me... :thumb:
Frank

 

[Luny Tune]

I thought I'd try to avoid double posting already discussed problems by reading through this thread. Alas, I stumbled across a little thing that I'd like to have clarified although I hope I'll never have any use for it. I'm talking about the fuse!

The schematic I've checked (the fuse wasn't included in my kit) says that I need a 315mAT fuse for the clone, but earlier in this thread there's talk about a 1A fuse. Well, I don't need additional headroom for meter light or anything so am I right in assuming that the 315mA fuse will be sufficient?

Somehow it strikes me as being upside down to put in a bigger fuse to be on the safe side!?!?:wink:

 

[gyraf]

Fuse size depends a lot on your power transformer - and as people are using all sorts of sizes here, it's hard to say anything in general.

For the small on-board type transformers, 250 - maybe even 100mA works fine.

For larger, specially toroid types, you may have to go as high as 630mA to prevent the fuse from blowing at power-up..

Jakob E.

 

[Luny Tune]

Well, I think the confusing part for me is I thought the fuse was for protecting everything after that point in the cirquitry and that the design of that cirquitry would lead to a certain fuse size required.

Anyways, I see your point as there's a lot of different configurations and preferences out there, so...

With a Dagnall D2185 tranformer I gather I'm good with a 250mA fuse.