Bottlehead Forum
Bottlehead Kits => Reduction => Integration => Topic started by: Karl5150 on March 27, 2014, 03:27:12 PM
-
Ok, the shunt regulator install seemed to go without a hitch (although I didn't try it before starting the C4S install) LEDs and #s all good.
The C4S however is different. The 2 LEDs on the A side of the B board don't light and the numbers are:
2 - 77.3
18 - 104.9
26 - 105
31 - 81.7
Time to replace the transistors on that side of the board..... or do those numbers point to something else?
Thanks for any help on this.
-
How are the numbers from the voltage regulator?
Do the socket LED's light up?
What were the voltages at 18 and 26 before the upgrade?
Do all the socket LED's light?
-
I don't have the pre install #s on the Reduction, the original build #s were all good and it has functioned well since built (noticably lower output than digital sources).
The 4 socket LEDs light (brightest)
The regulator LEDs light (brighter than the C4S board LEDs) The MJ... transistors both read 103V and terminal 56 is 129V
The 6 LEDs that do light on the C4S are dimmer but steady.
Thanks as always
-
LED's not lighting can be caused by not having the bA/bB jumper, or not having the Ib/Ia jumper, or potentially a cold joint on a resistor.
I'd also have a good look at those PCB solder joints, just to be sure you didn't miss one.
-PB
-
Thanks Paul,
The jumpers are there, but as always my soldering skills (or lack of) are always suspect. I will check before pulling and replaceing the transistor.
-
Apparently this time it wasn't my soldering as much as my ability to follow directions:
( ) Attach and solder the white wire
connected to 26U to OA.
It clearly reads 0A not IA.
Now all 16 LEDs light and it functions. I think must have at least screwed up one or both MJE350s cause 18, 26, and 31 all read 102-105V, the volume on that channel is 3-4 dB lower and has a scratchy (not the bad ground type) undercurrent to it.
Once again some guidance would be appreciated.
-
Did you try swapping tubes to see if the voltages follow?
-
Well Doc the voltage, volume and noise (worse) all follow the tube. Installing a SS phono stage clears channel variance issues. I guess my oops on the C4S board wasn't fatal to the board components.
Thanks, I'll start the tube search.
-
Try cooking the tubes in a for a few hours.
-
They were very quiet for the 3-4 months I've been using the Reduction prior to the upgrade. Would running them upside down for the multiple voltage checks cause something that is reversible?
(This is not an expression of doubt, it is just most of my very very limited knowledge is from lurking on the forum and any information helps)
-
It is possible that some junk inside the tube got into a critical area. But if the tube has always had low gain, then you probably need better matched tubes.
One of the disadvantages of zero-feedback designs is that the gain variations between active devices (tubes or transistors) are not equalized. IMHO the cure (feedback) is worse than the disease.
-
First, thanks everyone for responding.
Originally the tubes were reasonably matched as far as I could tell, images were well centered. I have ordered a matched replacement set.
For future reference: do preamp tubes need signal flowing through them for break-in?
Thanks again for the help.
-
Yes, it seems that all components need signal in order to break in in the most expedient manner. There is of course some very tiny noise floor present whenever the circuit is on, but it seems to take much much longer for that to have any effect than a music signal at normal levels
-
The new tubes arrived today and all the LEDs light and the C4S voltages check out (69-70V)
I listened for a couple of hours with only the usual ("that sounds a little off" then a little later "no, its ok") new tube issues. Considerably more gain than before the upgrade.
Then the dreaded scratchy, volume dependent, intermittent noise from only the right channel (not the side I originally miss-wired, post #6)
I swapped tubes, still in the right side. (maybe I wasn't careful handling them and mixed, post 3-28-14)
I swapped one end of the interconnects and it moved to the left, so it's not the amp.
Tried another set of cables, no change.
Swapped the TT interconnects, no change.
Back to the bench, all LEDs light (the 2 on the A side of the right channel board definitely take a little longer to light and subjectively seem not quite as bright) and the voltages are the same.
Looked at the ground solders and one looked a little less shiny so I reflowed both ends of the jumper.
Back inside- no change, after a brief warm-up the right side is still troublesome.
Tried the original tubes, except for volume mismatch, still always the right channel.
I hope this is enough for the bigger brains to point me in the right direction, thanks in advance.
Karl
-
So did the noise not follow the original tubes as you posted earlier? Reading back thru the posts I can't get a grasp on whether the noise follows a tube and/or whether it is in the channel that had the wiring error, or if it is showing up in different channels at different times. Can you now verify that
A) the noise is not associated with a certain tube
B) the noise stays in one channel and that channel is or is not the one that had a wiring error and
C) give us a new set of voltage measurements with the new set of tubes.
I would guess that there may be a damaged transistor after all, but some of the previous info posted was indicating otherwise.
-
Sorry for the confusion
The noise does NOT follow the tubes
The noise stays in the Right channel, NOT the originally miss-wired channel
With the new tubes the #s are:
MJE...transistors -103/102V
2 - 71V
18- 69V
26 - 69V
31 - 70V
Thanks again, I'll try and slow down to be sure what I write is as clear as possible.
-
OK, so we can eliminate the tubes and the repaired left channel miswire as the source of the noise. Since it is in one channel it is not a problem in the power supply. the next step might be to swap the C4S load connections from side to side and see if the noise follows the C4S components. I have gotten far enough away from working on the physical layouts of the kits that I will let PB weigh in on what terminals this swapping involves.
-
I still strongly suspect a cold solder joint in the right channel. The fact that it takes time for this to develop is all the more convincing.
To eliminate it as a source, you can swap the C4S boards between channels, just to be sure.
-PB
-
Thank you guys for your patience with this, I know I haven't been as careful and clear in my reporting of the issues.
1) before the upgrades the S.E.X and the Reduction worked well. Beautifully.
2) after transistor issues the C4S S.E.X. worked well with the stock Reduction.
3) MY miss-wire on the Integration C4S started this thread.
4) After ascertaining the noise that developed was only in the right channel of the Reduction I swapped C4S boards.
Here is where I'm at now:
The C4S board that was in the noisy right channel is now in the left and it works fine - quiet, musical.
There is nothing from the right channel, no hum, hiss, noise....no sound. (Switch RCAs out of the Reduction end and no sound from the left speaker- it is not the amp.)
I checked the wiring (multiple times), swapped tubes, swapped 1 end of in and out RCAs (1 set at a time), changed RCA cables, I even checked resistance between RCA in and out, both channels read the same.... but there is no signal from the right channel of the Reduction.
Both tubes and all 16 LEDs glow. Integration voltages as in previous post.
I ran through the resistance check and the voltage check from the Reduction manual and for now I will only report the abnormalities:
1) Overall the voltages are lower, but in the ballpark.
2) 43, 49 & 54 show OL not 0 ohm in the resistance check and show fluctuating <1V not zero volts in the voltage check.
I hope this is enough for some help, thanks again in advance.
-
OK, nothing from the right channel is very, very helpful!!!!
Download a 60Hz tone and play it into your Reduction at a very low level. Set your meter for AC voltage and connect the black probe to ground.
Measure the AC voltage at terminals:
2
6
10
If you can reach it safely, also check the AC voltage at the RCA jack input.
-
Here goes:
2 - With no tone - 143
With 60Hz tone - 143 to 170+ as input moved from min up
6 - With no tone - 0
With 60Hz tone - 0 to 0.6 as input moved from min up
10 - With no tone - 0
With 60Hz tone - 0 to 30 as input moved from min up
I did not do an input level sweep on the RCA pin because of its location, but a static check matched the signal on the left side.
2, 6 & 10 AC voltage readings mirror the left channel equivalents at slightly higher levels.
Karl
-
Those are the # on the "ACV" scale set at 200 (I did 'em again cause you got me scared :o)
I got a low level "sinus 60Hz -10dB" tone running thru my nuforce microDAC2.
They do vary depending on the volume the computer feeds the DAC, it defaults to 100%
Those # are way different than DC values.
-
Then you might be reading them wrong. 146V AC in the Reduction would be only at a transformer output to the high voltage rectifiers.
I don't have the reduction or the manual. But 143V AC is not going to happen in the signal path. Could it have been 143mV AC?
When you have a meter with different ranges start with the highest range and move downward on the ranges till you have a reading of at least 50% of that range. A reading of 0.6V or even 30V on the 200V range is unreliable.
-
Unfortunately the AC voltage on my meter only reads 200 and 600V scales.
The meter reads 122.4 from the wall outlet and a third check shows the same #s reported, similar on both channels.
DC reads down to mV, maybe I'll try and pick up a meter that reads AC in the mV range too.
Thanks for the heads up
-
Before you condemn your meter, look again. I bet there is a little mv indicator somewhere along the bottom. easy to miss...John
-
Those are the # on the "ACV" scale set at 200 (I did 'em again cause you got me scared :o)
I got a low level "sinus 60Hz -10dB" tone running thru my nuforce microDAC2.
They do vary depending on the volume the computer feeds the DAC, it defaults to 100%
Those # are way different than DC values.
100% out of the computer is too high, try 2%.
What I am looking for is for you to feed a fixed 60Hz tone into the Reduction, then to measure the AC voltage at those points without adjusting the level coming in. If you adjust the level coming in, then you need to start over.
-PB
-
Will do
The DAC also has a pot, should that also be set to the minimum?
I have access to a meter that reads AC in mV.
At 100% the test tone only showed a steady output of about 10% on the level meter built in to the volume display on the computer.
As an aside, I'm sure this is frustrating for all on the help end of the forum, just for different reasons than it is for me on the build end, but for me it is much more of a learning experience than the original builds, which went so well.
Thank you for all your time.
-
You can measure the signal coming out of your digital setup by measuring hot to cold on each RCA plug. Do your best to set those to 2-5mV.
-PB
-
Got 4mV from the DAC measured on the open end of 0.6m interconnects, plugged them into the Reduction and got:
2 - 2.9V AC
6 - 100mV AC
10 - 2.9V AC
I hope these #s make more sense.
Thanks
-
It seems, in my limited experience, that you are putting an AC voltage into a tube, and getting a higher voltage on the other side(it goes from 4mv at the grid of the first triode to 100mv at the anode of the first /grid of second triode, then to 2.9V at the anode of the second triode). In other words: an amplifier...
What is the AC voltage at the center pin of the output jack in this test scenario? If you're losing voltage between 10 and the pin, you've found your bad solder joint.
-
These numbers tell us that the right channel is passing signal.
Did I misread your post about no hum, noise, and sound from the right channel? (I read this as the right channel doesn't work at all)
-
Paul,
Yes it is the right channel. That has been frustrating on my end, seeing resistance and voltages where they should be, on what had been an opperating system. I guess if it passes signal and the AC numbers (once correctly gathered) weren't too far out of line then at least I got the boards right. (small win?)
Josh,
I probed 10 at 10U with access to 10L limited by the cap, stand off and ground terminal. Maybe a cold solder to the RCA out has shown up from all the physical motion of adding the boards and testing.
I have to put aside my preconception that because it worked properly before, the problem can only be from the upgrade point forward.
I will check the AC V and lift the board to check the solder joints to the RCA out.
Thanks again for all your help with this.
-
I have to put aside my preconception that because it worked properly before, the problem can only be from the upgrade point forward.
Those are words of sage wisdom!
-
I had the guy at work who builds/repairs power amps for motion control look at it and he pointed out several solder joints he thinks are marginal. I will spend some time re-flowing and see where that goes.
Again, thanks for all the time and instruction on this.
-
I believe we can marked this one solved.
First, thanks to all, especially PB, for the help. While I am not the most apt pupil, I did learn quite a lot and hope I will remember to apply these lessons in the future.
In the end it wasn't my marginal soldering (although I won't rule it out going forward) that was the source of trouble.
After removing it numerous times from the custom, copper tape lined box it shares with the S.E.X. and reflowing joints, checking the #s, returning it and having it work and then turning it on another time to not work (or vice versa) I realized the common thread:
It always checked-out outside the box. A tight fit, the right C4S board barely touched/barely didn't touch the box. Slightly melt one side of the stand-offs with the iron while applying pressure and the board cants slightly forward. Problem solved. (enter embarrassed emoticon here)
-
Nice piece of diagnostic work! You are to be congratulated. Figuring out a difficult problem often takes out-of-the-box thinking (sic :^)