Barry Porter "Net EQ"

[petermontg]

Thanks Gus, I figured it was a switching problem last night checked the board and everything is fine.
The resistor chain checks out on the MM going to pull it out of the EQ tonight and check it thoroughly

Really nice eq of what I heard so far. I think I might need a second :

Edit: So it turned out the switch was missing the internal part of the wiper. Guess that one got through QC. Luckily there are some spare switches laying around.

 

[frazzman]

Just when I was about to pack it in and check myself into the mental institution...

I got to the bottom of my tricky problem. It's kind of embarrassing...

It was the damn toroidal transformer, or more so it's internal location in the enclosure. If I moved the toroid about 12" from the unit the noise disappeared.

After months of chasing my tail I'm very relieved.

I think I will build a separate enclosure for the psu board and toroid.

Gustav - I seem to recall you housed your psu seperately, any tips?

Thank you to all who helped me!

 

[Gustav]

Just when I was about to pack it in and check myself into the mental institution...

I got to the bottom of my tricky problem. It's kind of embarrassing...

It was the damn toroidal transformer, or more so it's internal location in the enclosure. If I moved the toroid about 12" from the unit the noise disappeared.

After months of chasing my tail I'm very relieved.

I think I will build a separate enclosure for the psu board and toroid.

Gustav - I seem to recall you housed your psu seperately, any tips?

Thank you to all who helped me!

Cant be me...

https://pcbgrinder.com/index.php?route=product/product&path=59_70&product_id=52 (check the internal picture, toroid on the back of the unit)

Worked like a charm here.

Gustav

 

[frazzman]

Hi again guys,

Not sure what I was smoking when I thought I'd found my problem. Turns out I was wrong about the toroid.

To minimise anymore guesswork, I got my hands on an oscilloscope. The results are pretty telling but I wouldn't mind some insight as this is my first Forray into the scope world.

So far what the scope tells me is that my problem is not in the input stage as I had initially thought. I have injected a 1khz sine wave and recorded the following results. All feedback welcomed. Am I on the right track ?

See attached pic for probes shots at:
- Input XLR
- Output XLR:
- C7 (end of input stage?)
- R24 (start of output stage?)

Sorry for bombarding the thread and thanks for all your help.

 

[biigniick]

Yes, you're on the right track.  Input looks good.  Something before the output looks messed up. And the output looks even more messed up.  Try measuring the input and output of each band too. Maybe opamp related or power supply related? Looks like it may be crossover distortion? 

- nick

 

[biigniick]

hi all,
i'm having trouble finding a 30uH common mode choke that fits the board in the input section.  what have you guys found in the past?  this is the closest i've found, but it's over 4x too big.

http://www.mouser.com/ds/2/445/744841330-86680.pdf

thanks,
  - nick

 

[druu]

I think most had trouble sourcing this part so a lot of us jumpered instead without any issues.

 

[frazzman]

Yes, you're on the right track.  Input looks good.  Something before the output looks messed up. And the output looks even more messed up.  Try measuring the input and output of each band too. Maybe opamp related or power supply related? Looks like it may be crossover distortion? 

- nick

Hi nick/all:

I followed Nick's suggestion of tracing the signal through each of the bands. Given I am an oscilloscope newbie, I think I made some oversights in my previous measurements. This time I've tried to be more accurate in following the signal path as per the schematic.

I used the scope to measure the sine wave at the beginning of each band @ R99/R100/R101/R102.
- What I found was that the sine wave was already looking messed up at the beginning of the band(s) circuit path. Given this, I had to concur that the problem must be in the input stage.

It wasn't until I changed the VOLTS/DIV on the scope that I was able to see that at C7 in the input stage, the sine wave formation is already showing distortion.

I then went through tracing the signal path through the input circuit from the XLR input through to C7.

I checked off each component on the schematic as I went and noted any points where the sine wave form was incorrect. (Apologies for the crude image - my checklist + scope measurement)

The exact point where the distortion is introduced is at R32. (see attached).
I then measure the components immediately before R32:
- At U1-A pin 1 (opamp output) - sine wave form looks ok
- At R6 (10R resistor) - sine wave form looks ok
- At C12 (33pF cap) - sine wave form looks ok

Given that, I don't know what to make of the obvious problem introduced at R32.

Logically, the problem at R32 flows through the remaining components of the input circuit.
- The waveform is messed up at U5 pin 2(-) but not pin 3(+)
- The waveform is messed up at C33 (33pF cap)
- The waveform is messed up at R33 (3k3 resistor)

It is then most certainly messed up at C7 (end of input circuit).

Any thoughts? I just want to make sure I'm not misinterpreting what I am seeing on the scope and in turn possibly going down another rabbit hole...

 

[biigniick]

So, your problem is most likely U5 or C33, I think. The inverting input of U5, the negative feedback loop of C33 and R32 are all wired together. The components most likely to fail are IC's. Then caps. Resistors are unlikely to do this, I think. Try swapping U5 and see what happens.

- nick

 

[frazzman]

So, your problem is most likely U5 or C33, I think. The inverting input of U5, the negative feedback loop of C33 and R32 are all wired together. The components most likely to fail are IC's. Then caps. Resistors are unlikely to do this, I think. Try swapping U5 and see what happens.

- nick

Hey Nick,
Thanks for your reply. I had a chance to try your suggestions but no luck

- replaced u1 and u5
- replaced c33 - had a film cap there, put in a ceramic just incase my batch of 33pf film caps was bad
- checked values (visually) of r32 and r33

I was hopeful that the c33 replacement was it.

Here's an exhaustive list of what I've tried (unsuccessfully):
- replaced opamps
- substituted "known working" psu board and toroid from my nite eq
- xlr rewire
- test with oscilloscope to narrow down problem area - problem appears to be localised to the input circuit

Running of ideas now... I'm stumped

 

[Gold]

Running of ideas now... I'm stumped

Look at the problem area carefully under a magnifying glass. Look for cracks, solder bridges and general nastiness. Clean the flux from the board with alcohol and a toothbrush. Reflow all the solder joints and check again under a magnifier. Bang on the problem area and components with a non conductive object like a pencil eraser. See if anything strange happens.

 

[Harpo]

Here's an exhaustive list of what I've tried (unsuccessfully):
- replaced opamps
- substituted "known working" psu board and toroid from my nite eq
- xlr rewire
- test with oscilloscope to narrow down problem area - problem appears to be localised to the input circuit

Running of ideas now... I'm stumped

Just so I get it right, from all previous posts you reveiled the parts values of R6, C12, C33 and R33, you used a psu board and toroid from your nite eq, you tested the NetEQ at a frequency of 1kHz and you exchanged some parts (caps, opamps) with other kept a secret parts, except C33 that is now a 33pF ceramic cap.

It might be helpful to at least post your DC supply voltage numbers, the type of opamps used, the level of your 1kHz test frequency and if you fitted pots or substituted these pots with switched resistors because length and position of wires to these switches might make a difference, ... Some pics of your build might help as well.

 

[frazzman]

Hi all,

My net eq saga is finally over. The problem is finally resolved... For real. After many months and desperate posts.

Harpo - what you said to me in the beginning has once again proven to be true. It was the opamps...

With regards to testing the opamps - I (wrongly) assumed in the past that I was dealing with the possibility of only 1 faulty opamp. My approach then involved substituting  each opamp 1 by 1 (with a known "working" opamp). If the problem wasn't resolved, I changed the opamp back and moved onto the next.

Now that I have an oscilloscope I tried a more strategic approach. This time with a sine wave present, I measured sine wave at pin 1 of each opamp. If it looked distorted I replaced it.

On both boards I systematically replaced the opamps until the frequency response was perfect. I had to replace U1,U2,U4,U5 and U6 on both boards.

Attached is the frequency response now (you will recall earlier how much of a ripple there was) THD was nearly 0.7 compared to 0.0015 now.

More likely than not, the opamps were fried when I had the neg and pos leads the wrong way around (misread schematic).

Thanks to everyone who helped me and took interest in my issue. I made some significant oversights here but I am glad for what I have learnt along the way and I now understand what an invaluable tool an oscilloscope is.

Thanks again

 

[Gustav]

Hi all,

My net eq saga is finally over. The problem is finally resolved... For real. After many months and desperate posts.

Harpo - what you said to me in the beginning has once again proven to be true. It was the opamps...

With regards to testing the opamps - I (wrongly) assumed in the past that I was dealing with the possibility of only 1 faulty opamp. My approach then involved substituting  each opamp 1 by 1 (with a known "working" opamp). If the problem wasn't resolved, I changed the opamp back and moved onto the next.

Now that I have an oscilloscope I tried a more strategic approach. This time with a sine wave present, I measured sine wave at pin 1 of each opamp. If it looked distorted I replaced it.

On both boards I systematically replaced the opamps until the frequency response was perfect. I had to replace U1,U2,U4,U5 and U6 on both boards.

Attached is the frequency response now (you will recall earlier how much of a ripple there was) THD was nearly 0.7 compared to 0.0015 now.

More likely than not, the opamps were fried when I had the neg and pos leads the wrong way around (misread schematic).

Thanks to everyone who helped me and took interest in my issue. I made some significant oversights here but I am glad for what I have learnt along the way and I now understand what an invaluable tool an oscilloscope is.

Thanks again

Excellent news!

Gustav

 

[petermontg]

Congrats frazzman ;D

 

[frazzman]

Thanks dudes. I'm really stoked. A few days ago I was pretty much ready to put it on the black market as I was so sick of looking at it.

Also thanks to my buddy Druu who helped me over many many dozens of emails.

Now back to making some music!

 

[petermontg]

Thanks dudes. I'm really stoked. A few days ago I was pretty much ready to put it on the black market as I was so sick of looking at it.

Also thanks to my buddy Druu who helped me over many many dozens of emails.

Now back to making some music!

Never give up the DIY spirit.

 

[therecordingart]

Does anyone have these Gerbers? All of the links I've found are dead.

 

[ruairioflaherty]

Gustav sells finished PCBs, why bother with Gerbers?

 

[therecordingart]

Gustav sells finished PCBs, why bother with Gerbers?

Because I fabricate boards for a living and it's easier for me to just make them.