Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[xonlocust]

hi all-

novice question here - i'm troubleshooting a pair of poorly racked (by myself!) pair of neotek channel strips.  in trying to trace the distortion i've fed by oscilloscope a sine wave to one channel, and the strips get the other with my probes "viewing" the other channel.  

my problem is this: when i connect the sine wave input to the neotek, the sine wave gets flattened (grounded out? nulled? flatlined?)  if the sine wave is not connected to ch 1 of the oscillosope, it reads the signal off the probes fine.  i just wanted a reference to compare against as i step through the neotek. is there an obvious answer here i'm missing? something to do with grounding i'm guessing?

thanks much in advance,
nick

[Geoff_T]

Hi

I must admit that I know nothing about Neotek consoles so can't comment on their circuitry other than to warn that some equaliser circuits (like the Neve B284) use an inverting amplifier for the gain. You can't see the signal on the input to this amplifier because it's a current rather than a voltage but you can most certainly see the signal on the output.

I don't know if your oscillator is capable of generating a square wave output but this will tell you a ton more information than just looking at a sine wave. It's very hard to see small distortion artifacts on a sine wave. The rising edge of the square wave is the HF response while the flat part at the top is the LF response, so you test frequency response as well as signal level. Very good at spotting bad caps!

Make sure that you have the grounds of the test equipment connected to the grounds of the module under test or you may provoke ground loops that will spoil your measurements.

I hope this helps!

[xonlocust]

hey geoff-

i didn't know that about putting in a square wave for measurement, thanks for the info! maybe i need to get a legitimate tone generator then, because right now i'm just sending a sine wave from a test tones cd.  as you can see, i'm really a novice tech, just trying to learn the tricks of the trade to troubleshoot/fix my own stuff, and so i can be more conversant with "real techs".

one thing that's suprising though is that i get the sine wave reading off of the 1/4" jack fine (while off the rack), then when i plug it in and test it while seated in the female jack, i get nothing - it's not even anywhere downline in the circuit!

thanks again for your help!

-nick

[Geoff_T]

xonlocust wrote on Tue, 01 June 2004 11:41

hey geoff- i didn't know that about putting in a square wave for measurement, thanks for the info! maybe i need to get a legitimate tone generator then, because right now i'm just sending a sine wave from a test tones cd.  as you can see, i'm really a novice tech, just trying to learn the tricks of the trade to troubleshoot/fix my own stuff, and so i can be more conversant with "real techs". one thing that's suprising though is that i get the sine wave reading off of the 1/4" jack fine (while off the rack), then when i plug it in and test it while seated in the female jack, i get nothing - it's not even anywhere downline in the circuit! thanks again for your help! -nick

Hi Nick

It sounds like the impedance you are plugging the CD output into is too low (or a dead short) and is giving the output stage a hernia. It may be that the CD has a series limiting resistor and the circuit you are testing is forming an attenuator. It's hard to say without seeing the big picture of what's going on.

"Proper" test equipment isn't expensive... you can pick up used HP and Heathkit test gear off EBay or build your own basic test gear for peanuts.

Here's one example...

http://www.qkits.com/serv/qkits/velleman/pages/MK105.asp

PS If you browse through the audio section of that site, there's some cool amplifiers!

[Geoff_T]

Hi

A silly thought struck me...

You mentioned that you are using a CD and you mentioned a quarter inch jack.

The CD output is stereo so, if a minijack to quarter inch adaptor, you have left and right signals on the tip and ring and ground on the sleeve.

If you are then patching that into a balanced input, the two sources are presumably identical and will cancel out in the balanced input stage because it's expecting a difference signal.

You need to make an adaptor lead that takes one side of the CD output to the quarter inch jack plug. The sleeve and ring can both be ground... this isn't ideal but a lot better than what I think you may be doing.

[xonlocust]

hey geoff-

well, i had a *proper* tech friend over to help me out with it and i'm terribly embarrassed to say that i had my line connections switched on the 1/4" ins/out. (i had the + going to ground and vice versa - IDIOT!)  at any rate, that fixed that issue.  upon surveying my handiwork he was like.... what kind of mess have you created?!

the other channel wasnt passing audio past one of the chips, which we then swapped which didn't help, so we're speculating some faulty jfets around the TL074 we're near.    

good idea about the jack, but i was going rca out of the cd player to 1/4" adapter - that one i had actually done properly...

thanks for the help so far,
nick

[xonlocust]

well, one down, one to go... (by the way - we're talking about a neotek series 1e channel strip here)

i got myself a nice software square wave generator and am feeding the tone in.  one behavior that i'm seeing is that when the power is off, i can step through about half of the unit (past the patchbay insert points) tracing the signal, but it gets lost before the direct out.

when i power up though - when i put the probe on the pcb, the line jumps wayyyyy off the screen and very slowly comes back down to where i can see it.  the peak light is constantly lit up, and i'm sure there's some oscillation going on in there. question: is oscillation a symptom only of faulty ICs? ie, if i were to just go ahead and replace ALL of them, conceivably i would fix the problem? i've already tried replacing the ICs (TL074) that seemed suspect (not the EQ chips though) - as well as the FETS that buffer the first IC the signal hits. (as initially suspected by my friend - though it hasn't fixed the problem)

could anyone suggest some high-level troubleshooting techniques here?  i basically get that the process is to stick a tone in - and try and trace it though the schematic until i find where it breaks down, but what am i looking for?  what's the expected behavior after say, a resistor or capacitor?   also, [dummy question stating the obvious] for meaningful results i need power on while doing this - isn't that going to further damage whatever is broken the more i poke around with the power on?

any suggestions for my novice fumbling self would be greatly appreciated.

thanks so much!
-nick

[Geoff_T]

Hi

Well, here's a few general hints.

1. A circuit diagram and component layout drawing can save a lot of time when fault tracing... though aren't indispensable.

2. It's important to have a print out of the IC pin functions and transistor footprints so that you can figure which is an input and which is an output. Generally it's quickest to check what's on the output in case it's an inverting amp and hard to read the signal on the input.

3. Do a voltage test first. Do all the power pins on the IC's have the correct power to them. Neve used to use series resistors in the power rails as part of the decoupling and these resistors could fuse, going from 10 ohms to hundreds of kilohms. If you are missing volts, trace the tracks back to find the broken track or component. Also, outputs are generally in the centre (0v on +/- supplies or half rail on single rail supplies) so check that there isn't a massive dc offset.

4. The purpose of the square wave is to check frequency response at the same time as level. You can move the EQ pots and check that the square wave changes in response to the lift/cut. The vertical portion is HF and horizontal portion LF. This is also a quick check for bad coupling capacitors that might have a good wave on the input but crappy one on the output.

I have a feeling that your problem will be resolved with the voltage test because it sounded like you had some offsets there to kick your scope trace (unless you touched a power rail!)

I hope this helps