Luxembourg Crack failing resistance check after pot upgrade [solved]

[grausch]

Ok, I probably just should have left the Crack alone as it was working perfectly and I had no complaints. Unfortunately the urge to tinker got the better of me and I installed a VA Labs 100k stepped attentuator.

All the resistance checks are working, except for Terminal 10 which is 0. Looking at the wiring, this is the left channel and since it reads 0, I assume I am grounded somewhere where I should not be.

The steps I have done are the following:
1. Marked in / ground / output as per PB's advice. http://bottlehead.com/smf/index.php?topic=5542.0
2. Grounded the 2 channels on the attentuator together. Refer to the first picture for wiring on the top of the attentuator.
3. Removed and redid the braided wires from the RCA jacks, and soldered white and red to the inputs of the attentuator. Ground was attached to the outer ring.
4. Removed and rewired the ground wire from the headphone jack to the attentuator. The second picture shows the wiring on the headphone jack. Everything is per the manual, with the startup voltage mod added. The ground is attached to the top of the outer ring on the attentuator.
5. Wired the black wire from T3L to the bottom outer ring of the attentuator. This can be seen on the third picture.
6. Removed and rewired the red wire from A7 to the output of the attentuator.
7. Removed and rewired the white wire from A2 to the output of the attentuator.

My first though was that the white wire at the RCA jack must be the culprit, as it was difficult to remove all of the solder from there. However, I believe that if the connections was problematic I would not get such a nice 0 ohm resistance. The resistance chekcs at the RCA jacks are both very close to 100kOhm.

Why would only one channel be grounded? I would think that if my grounding was wrong at the attentuator, then both channels should be grounded.

Any troubleshooting advice would be appreciated.

Cheers,

Gunter

(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2F5b7Tglj.jpg&hash=27e703782cafce793bf69a22e520a4b75d7c2df9)
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FU5ANgkV.jpg&hash=4154530975b1ee16f31d671f7828a45408b50e50)
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FmIADzbm.jpg&hash=e5ed0c312e5f191ecd06a87facdac228e5596018)

[Doc B.]

T10 is in the output stage rather than the input/attenuator stage. So my hunch would be that there is maybe a short at the headphone jack. Curiously it looks like the right channel red wire that goes to T6 is shorted to the black wire, but the left channel white wire that goes to T10 looks OK.

You might want to pick the amp up and shake it a little and see if maybe a cutoff bit of wire falls out.

[grausch]

Wow, that is a fast response. Much appreciated Doc.

I desoldered the ground wire leading to the attentuator at the headphone jack, but left everything else as it was. Thus, both the bottom lugs were desoldered. The lug with the two resistors attached was really tricky as there was only a small hole to get the wire through. The two resistors have moved and are touching, but since they connect at ground, I could be wrong, but don't think that is the problem.

The lug with the red wire has always been somewhat recessed. That made soldering on it very difficult and is also why you see a blob of solder on the red wire.

I was pretty careful desoldering and don't believe I melted anything on the headphone jack. Temperature is set to 375 elcius as I was doing PCB soldering for the Stereomour earlier and I find I like that temperature.

Please refer to the attached pic for a better shot of the headphone jack wiring.

Would it be worthwhile to replace the headphone jack, or should I look for a short elsewhere? I should note that the capacitor at T10, has been bent back and forth a few times, but still looks to be in fine working order, and I don't think this will lead to a grounded connection.

Cheers,

Gunter

(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2F2o53HGT.jpg&hash=0e55146477c64a52b061291a85d552408e0c67b7)

[Doc B.]

I think at this point you might want to inspect the connection one at a time. First disconnect the white wire from T10 that runs to the headphone jack. Then measure across the 2.49K resistor  on the headphone jack to make sure it is not shorted. If it is OK, measure the resistance between the loose end of the white wire and the ground tab on the headphone jack.

[grausch]

Ok, I have shaken the amp and I can not hear anything that sounds loose.

I also do not see any obvious shorts. I have traced the signal to follow the resistances at various levels. Black is connected to Terminal 12.

RCA jacks - Both 99.5 kOhms
Input at attentuator - Both 99.4 kOhms
Output at attentuator - Both 0 ohms with attentuator closed and 100kOhms with attentuator open.
Terminal A - A2 for white and A7 for red. Same as output - 0 closed and 100kOhms open.
Terminal strips 6 & 10 - Terminal 6 2.49 kOhms and Terminal 10 0.
Headphone jack - White channel 0, red channel 2.47 kOhms.

Could the problem lie at Terminal B? My soldering & wiring at the Terminal is bad, but no exposed wires touch. I have also poked everything there, and found no difference in resistances.

Doc, just saw your post regarding disconnecting the wires. I will need to do that tomorrow as it is already 0h30 over here. Will post once that is done.

[JamieMcC]

Not sure if this will help you but all three of my grounds connect to the top of the Valab (there is a bridging wire between the grounds like yours its just hidden from view in this shot).

Edit I do notice my inputs and output are orientated on different sides to your attenuator ie right channel in and out are closest to the wood enclosure did you trim off the little plastic locating spigot?

(https://www.head-fi.org/content/type/61/id/1013227/width/900/height/675/flags/LL)

[JamieMcC]

A couple more pics

[Paul Birkeland]

Doc B. didn't notice that you have the shorting mod.  Both T6 and T10 will read 0 Ohms with no headphones plugged in and the shorting mod. 

-PB

[grausch]

@JamieMcC: Thanks for the response.

• Once the grounds are bridged, both outer rings (top and bottom) share a common ground. Thus, it should not make a difference which grounds I used. I did not want to desolder the ground at T3, as access to it was tricky. While contemplating that, I realised that my wire was long enough to reach the bottom so I soldered it there. Tricky, and if I get static it will be one of the first places I reflow the joint.
• My input and outputs should be same as yours. Per PB changing that would lead to a non-working attentuator. We do have left and right channels reversed, but that does not matter. At that point, as long as the same channels are connected back together it should work fine.
• Locating spigot - Not sure what this is. I am assuming it is the little white ?dot? that goes into the aluminium plate to keep the attentuator from spinning. No, I did not need to trim anything of that. I just needed to enlarge the main chassis hole for the attentuator, which I accomplished with a metal file.
• On another note, if you install decide to install this attentuator in your S.E.X., I seem to recall reading on the forums that the grounds should not be bridged. I have briefly read that section on my Stereomour manual and I think it is not bridged also. Difficult to say without having actually gotten to that part of the build yet. I could be mistaken, but I believe the common ground on the attentuator can be done in the Crack because they get grounded together at the headphone jack in any case. If someone from Bottlehead could confirm or correct this, I would appreciate this.

@PB: Well, this is quite a facepalm moment. Thank you for pointing that out and refreshing my memory.

• I had read about it on the forums, and once I did the shorting mod I actually checked for 0Ohm resistance at both of those points to ensure the mod was working correctly. I got 0 Ohms there when I had built my Crack initially. I should have remembered that. Will be updating my manual later today.
• Now, with that out of the way, any potential reason you can think of why the red channel is not grounded (T6 not showing 0)? When I built the Crack, the signal wire attached to the lug, but the lug itself was quite recessed, and I melted some plastic trying to get solder to flow. However, I don't think that will cause it to not ground.
• I am assuming the most likely culprit is the solder joint on the ground lug of that channel. However, with the white channel grounding correctly, I am a little baffled that the red channel does not. I used a single wire through the lugs and soldered it to all 3 of them. It is possible, but doubtful, that I removed some solder from this lug, when I was working on the bottom lug. I will reflow it in any case and report back. I will also plug in some headphones and see what the resistance readings are then.
• Since the problem lies with the startup mod, I believe it should be safe to listen to it like this. I will redo the voltage checks, and unless something is off still listen to the amp. Although, this is something that I would like to correct in the near term future.

@Doc B.: Thank you for the input. You have now idea how much my brain got engaged trying to figure out what went wrong. I have actually started tracing the signal and measuring resistance based on what I believe they should be at certain points in the signal path. Feels like quite an accomplishment. Before I built the Crack and had access to this forum, that would have been unthinkable.

[grausch]

I finally figured out why the resistances don't work on T6. Headphone jack has a recessed lug which disconnects it from the ground.

Refer to the 2nd lug, where it can be seen that no contact is made. Once headphones are plugged in, contact is made, thus no problem.

While on the topic, I believe this is already a Neutrik jack. I prefer not to use a locking jack. I have searched quite a but, but not sure if there are any upgrades available. Does anyone know of an upgrade for the headphone jack that is not locking.

(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2Fi6FdHNR.jpg&hash=f2eba1e26666caece8be8fe8bdf380dfff2ba50d)

[Paul Birkeland]

There is a Switchcraft jack that I would consider an upgrade.  It does require the chassis hole to be drilled out a little bit, and it doesn't have the switching function. 

-PB

[Chris65]

There is a Switchcraft jack that I would consider an upgrade.  It does require the chassis hole to be drilled out a little bit, and it doesn't have the switching function. 

Which one would that be PB?
The enclosed jack series e.g M112BX or the shielded type jack?

[Paul Birkeland]

This guy:

http://www.digikey.com/product-detail/en/152B/SC1035-ND/109342

It's not a very glorious looking jack, but the quality is quite good.

[Chris65]

This guy:
http://www.digikey.com/product-detail/en/152B/SC1035-ND/109342
It's not a very glorious looking jack, but the quality is quite good.

Ok, thanks Paul. Comes in gold or black too.

[grausch]

@PB: Thank you. I had a quick look at them previously, but was looking for something that looked like the Neutrik. Based on your recommendation, I have looked at them again and can source them from Germany. May give it a try. It will be a bitch to solder in the 2.49K resistors though.

With regards to enlarging the hole, I have a metal file that works just fine. Have enlarged holes on my Crack and Stereoumour for the VA LAB attentuator.

With regards to the switching function, does that only affect the shorting mod for the start-up voltages? From my limited understanding, I believe the switch was causing the issue with my resistance checks. Is there any other downside that I am not aware of?