Turbo Gssl help please..

[user 10827]

Hi All,

is there a copy of the silk screen for the pcb, just so i can make sure the resistors are correct, im sure they are but need to eliminate.

i have a slight problem,

the gssl only compresses from the right channel.

when pluging only the left ch in , no compression can be achieved.

I have measured voltages, +-12 all good.

Ratio sw fine and voltages on pin 14 tlo74 behave and are the same as main board pin 14 tlo74.

voltages around the turbo 2180 are all the same as the main board 2180

whats odd is my pin 7 tlo74 is at 0v but so is my main board pin7 tlo74.

every thing looks like it should work, with both channels in i had not noticed this fault, but when you pull the right channel out it no longer compresses, of course audio still happily passes through,

im at a loss now, i have no scope either which is not helping.

one other thing to note when wired all back to normal gssl mode it does not display this fault.

hope some one can help
best
Pete

 

[gyraf]

..make sure that the sidechain board gets signal from both channel's input opamps..

Jakob E.

 

[user 10827]

Hi Jakob,

Thanks for your reply,

not sure I follow, I am under the impression that the purpose of the Turbo board is to individually process one of the input op-amps signal?

Best

Pete

 

[SSLtech]

Hey Pete,

When I get a chance later today I'll have a look at this, but do you have a digital (or analog will do!) voltmeter handy?

We need to send a constant tone to both inputs at the same level (a Y-cable for example) and set the unit for 2:1 compression ratio, then measure voltages at three of the 4 output pins on the TL074. -If you alter the threshold control, you should see the measured DC and AC voltages at a few of the pins (I think it's pins 1, 7 and 14... but don't quote me on that JUST yet!) and they should be the same for each setting...

Read the other Turbo thread from MartyMart for a few pin references, it'll tell you which pin on the TL074 ISN'T the same.

Keith

 

[user 10827]

you sir are super lovely!

yes I have a meter and Lindos Tone Genny that provides L+R tone.

I have indeed read right through martys thread, and most (but not all) voltage seem right.

Sure I understand one of the op-amps is not used in the TLO74, I think its pins 8,9 and 10?

I'm measuring on main board TLO74 pin14 going negative when in compression, correct.

but on the turbo board it only moves by about half a volt, cant remember the specific voltages, (at work right now not in front of unit).

I think here is the problem.

the output (top of my head is it pin 8) of the 2180 you cant seem to detect any DC volts at all on either main or turbo so im assuming its current driven??

when ever you have the time Keith is fine, no rush.

Best
Pete

 

[SSLtech]

Okay, here's a basic troubleshooting run-through:

First print out the Schematic from here, and refer to it as you go through the troubleshooting.

First checks:

When switched to Turbo mode and with SILENCE at both inputs, the turbo TL074 outputs for pins 1 and 7 should all have 0Volts DC and 0 volts AC on them. Pin 14 should have a fixed DC voltage on it which varies as you switch through the three available ratios. This should match the same exact thing happening at pins 1, 7 and 14 respectively on the main board TL074. On the turbo board, pin 8 should have a DC voltage on it which varies as you adjust the "threshold" control.

All good? -Moving on:

Now, leave the compressor switched to 2:1 ratio, with the threshold set to LEAST sensitive... i.e. compressing the least. -Apply a steady tone at a decent signal level (about 0dBm: -In fact anywhere between a 0.5 -to- 1 Volt is perfect) to BOTH inputs. Verify that the outputs of BOTH of the input 5534s on the main boards have a similar AC voltage at pin 6.

Next, leave the meter switched to AC, and measure the voltage at pin 7 of the main board TL074. -You should see that the signal voltage rises and falls as you vary the threshold control. -You see that? -Then make the same measurement at pin 7 of the Turbo board TL074, and you should see the same thing. -In fact you should see basically the SAME voltages for all settings of the threshold control.

All good? -Moving on:

Now meter the signal at pin 1 of the Main board TL074. -It should now have BOTH and AC and a DC component, and how much of each you will measure depends on your meter. -If you have an oscilloscope, you should be able to see a "multiple-camel-hump" sort of a waveform... once more, It will vary in amplitude (and at very low levels, slightly with shape, but don't worry about that...) with the threshold control. -Basicallty, for each setting of the threshold control, BOTH TL074s should have the same sort of shape waveform at pin 1, and whether your AC voltmeter meter is averaging, true RMS, or just making a guess at the waveform, -so long as the readings for each meter setting (AC and DC) are similar for each position that you set the threshold control to, then all is good.

All good? -Moving on:

Now meter the signal at pin 14 of the Main board TL074. -It should now have BOTH and AC and a DC component, and once again it shoudl agree with the same numbered pin on the Turbo board TL074. This time there will be a DC offset (this is the threshold).

(By way of explanation of the SSL circuit here: -the unit starts compressing as the voltage transitions through zero volts, and so the 'DC offset' with silence at the input is the 'threshold' which the rectified signal has to overcome in order to begin compression.)

The threshold control should-once again- have a significant effect once a decently-sized signal is present (and -of course- no effect whatsoever when the input is silent) thus determining how easily the signal 'forces' the voltage across the zero line.

If you get the same signal at pin 14 also, then the only thing before the output of the board reunites with the main board is the 1N4148 diode which goes to point 'C'.

That's all there is to troubleshooting the board in terms of diagnosing and locating the area where things may be going wrong... -Double-check the orientation of the diodes of course.

If you have access to a scope, the shapes of the waveforms should look similar, as well as the amplitudes, and this makes troubleshooting a breeze of course... but nothing should be impossible even with just a modest multimeter.

Pete, let me know how it goes and what you find.

The Turbo is essentially the same as the sidechain section in the GSSL version from the 47k resistors on the left (where the signal splits) to point 'C' (where the signal reunites), and so you should be able to easily see what it should be doing, and you should have a 'working model' on the main board, in order to make comparison readings and measurements.

Report back with your findings. -nothing shall defeat us! :wink:

Keith

 

[user 10827]

Hi Keith,

Thanks for taking the time to write that, its all printed out, I will check tonight and report back tomorrow.

Do you have a copy of the silkscreen to hand? now all the resistors are in its hard to tell what component values are, I doubt I have made a mistake here as it was a kit from Ptown and I have nothing left over!! but you never know..

Its all a bit embarrassing really, I should be able to work this out.

Thank you again,

best

Pete

 

[DrFrankencopter]

Are there schematics for the Turbo? Or is it just a PCB only? I've been away from the board for a while (almost 2 years)...the 'turbo' SSL is new to me. Can someone fill me in?

Cheers

Kris

 

[SSLtech]

As described above:

The Turbo is essentially the same as the sidechain section in the GSSL version from the 47k resistors on the left (where the signal splits) to point 'C' (where the signal reunites)

That's your schematic... Plus a little bit of buffering to reduce threshold voltage load etc, but really that's all there is to it.

Keith

 

[user 10827]

HI Keith, results follow....

First checks:

When switched to Turbo mode and with SILENCE at both inputs, the turbo TL074 outputs for pins 1 and 7 should all have 0Volts DC and 0 volts AC on them. Pin 14 should have a fixed DC voltage on it which varies as you switch through the three available ratios. This should match the same exact thing happening at pins 1, 7 and 14 respectively on the main board TL074. On the turbo board, pin 8 should have a DC voltage on it which varies as you adjust the "threshold" control.

Turbo Board TL074 DC volts
Pin 1 +.04
Pin 7 0v
Pin 14 Varies correctly (2v, 3.2v, 3.8v)
Pin 8 does vary with threshold.

Main Board TL074 DC volts
Pin 1 -.15
Pin 7 0v
Pin 14 Varies correctly (2v, 3.2v, 3.8v)

All good? -Moving on: YES

Now, leave the compressor switched to 2:1 ratio, with the threshold set to LEAST sensitive... i.e. compressing the least. -Apply a steady tone at a decent signal level (about 0dBm: -In fact anywhere between a 0.5 -to- 1 Volt is perfect) to BOTH inputs. Verify that the outputs of BOTH of the input 5534s on the main boards have a similar AC voltage at pin 6.

YES, all good on pin 6.

Next, leave the meter switched to AC, and measure the voltage at pin 7 of the main board VCA. -You should see that the signal voltage rises and falls as you vary the threshold control. -You see that? -Then make the same measurement at pin 7 of the Turbo board, and you should see the same thing. -In fact you should see basically the SAME voltages for all settings of the threshold control.

NO, only DC volts seen on both boards 11.8v. My multi-meter does display an AC voltage for a few seconds but as you probe it it ramps down to 0v, this is the same for both boards, im guessing this is most likely correct.

EDIT: Harpo has PM'd me, suggesting there might be a typo above, did you mean Pin 3? (not pin 7) this would explain the funny results i found here

EDIT: My mistake... I meant pin 7 of the TL074, NOT the VCA... -Keith

Now meter the signal at pin 1 of the Main board TL074. -It should now have BOTH and AC and a DC component, and how much of each you will measure depends on your meter. -If you have an oscilloscope, you should be able to see a "multiple-camel-hump" sort of a waveform... once more, It will vary in amplitude (and at very low levels, slightly with shape, but don't worry about that...) with the threshold control. -Basically, for each setting of the threshold control, BOTH TL074s should have the same sort of shape waveform at pin 1, and whether your AC voltmeter meter is averaging, true RMS, or just making a guess at the waveform, -so long as the readings for each meter setting (AC and DC) are similar for each position that you set the threshold control to, then all is good

results as follows..

Turbo Board Pin 1 (least compression into compression)

AC .07 -- .26
DC .04 -- .03

Main Board Pin 1

AC .23 -- .50
DC 0v -- .12

Now meter the signal at pin 14 of the Main board TL074. -It should now have BOTH and AC and a DC component, and once again it shoudl agree with the same numbered pin on the Turbo board TL074. This time there will be a DC offset (this is the threshold).

Turbo Board Pin 14
AC 0v -- 0v
DC .2 -- .15

Main Borad Pin 14
AC 0v -- .75
DC +.17 -- (neg)1.5v

So, I think this is where the problem lies, Nothing on Pin 14 of the Turbo at all.

I'll let you digest and await your conclusion,

Cheers Keith,

Best
Pete

 

[SSLtech]

Pete,

Looking at the schematic, TL074 pin 7 should have an AC signal and basically NO DC on it on either board... that flummoxes me.

Pin 3 should be connected to 0V ground at all times, so you should read nothing there, or else either your meter's faulty or your black probe isn't on 0Volts! :wink:

Lemme havva fink.

Keith

 

[Harpo]

.., and measure the voltage at pin 7 of the main board VCA...

Pete,
sorry for confusion, obviously I read it like you. Maybe better should read '.., and measure the voltage at pin 7 of the main board TL074.' (pin 7 of the VCA is its +12V supply node)
-Harpo

EDIT:Not your fault... I was in full 'ramble' mode and I wrote VCA instead of TL074. -Original post now corrected.

 

[SSLtech]

Ah... I wrote 'VCA' and I meant 'TL074'

Please accept my humblest apple-hoagies!

Keith

 

[user 10827]

hey no need to apologise!! :thumb:

so we can say with no signal present both TLO74 pin 7's have Zero volts, I shall check tonight what the AC voltage is when signals are present then we will have completed and recorded all voltages from Keiths procedure, hopefully then we should be able to move to the next step..
what ever that might be...

thanks guys!!

Pete

 

[SSLtech]

I edited my earlier troubleshooting direction post to reflect the correct procedure.

So now... referring to the schematic, you should be able to "follow" the left-to-right procedure which I'm describing. -Find where things diverge in terms of dissimilar readings, and you'll have found where we need to look.

Here's an overlay, in case you need to check component placement:

Keith

 

[user 10827]

thanks Keith for the image, just what i needed.

A far as measuring AC movement with the threshold on Pin7 on the Turbo Board TLO74 i get Bugger all, here is the problem.

So ill take a closer look round the TLO, i know this is where the problem is, I know its not the TLO its self because ive swapped it.

but does the the output of the 2180 feed that first opamp(pin7)so maybe its the O/P of the 2180...??

just a thought, double checkin n all on the ceramic caps, the ident 101 is the 100n and 104 is 100pico??

im sure it is but doubting my everything at the mo..

Cheers guys..

Speak soon

Pete

 

[idylldon]

[quote author="sinestaraudio"]just a thought, double checkin n all on the ceramic caps, the ident 101 is the 100n and 104 is 100pico??

im sure it is but doubting my everything at the mo..
[/quote]

You have them reversed. 101 is the 100 pico and 104 is the 100n.

To get mfd: 104 is 10 plus four zeros or 100000 picofarads. Move the decimal point over six places to the left, and you get .1mfd.

Cheers,
--
Don

 

[SSLtech]

The opamp feed pin 6. It's a current-output device so you can't read any signal voltage there. You can test the 2181 by swapping it with the sidechain VCA on the MAIN board, if they're socketed... (I usually socket stuff for this reason)

-If you have so signal at the signal input pin you'd also have an issue with no signal appearing at pin 7...

A suspect VCA (if socketed) can be eliminiated from suspicion by a 'paper-clip' link from pin 1 to pin 8 replacing the VCA... it will then just pass signal with no gain and no attenuation.

Keith

 

[user 10827]

what a plum I am! ok ill swap those Caps round, see where I get then.

Ill get back on it tonight.

On another note, I built Keiths CRCRC board, My noise floor was -75dB, now its -90dB!! now that is cool...

Pete

 

[Harpo]

If you have fitted a 100nF instead of a 100pF in the feedback loop across the 33k resistor, this would set the lpf -3dB cutoff from 48kHz down to 48Hz. Testing this with a f.i. 1kHz signal will result in a ~26dB lower level or voltage gain of 0,05.

Just curious, the schematic, pcb-overlay and Keith's post call for a That2081. Will a fitted 47R matter for a That2080 here (the pin-4 to cut question)?