SP14 looking for some help

I've been enjoying my SP14 for a couple days without any issue.  Tonight the sound started fading on both channels and began to hum.  I took a look at the preamp and the power led was flickering.  After powering the unit off & on it works fine for a few minutes then the same thing happens.  Any advice on what to test?  I tested the rectifier tube and it appears ok.  Any help is appreciated.  Thanks!

did you build this yourself from a kit, reason I am asking is that if you did, check all connections and solder joints to begin with

MontanaWay wrote:did you build this yourself from a kit, reason I am asking is that if you did, check all connections and solder joints to begin with

Yes, this was a kit I built.  Initially I had a bad solder joint which caused no volume from the left channel.  After going over many of the solder joints that troubled cleared up.  I'll take your advice and re-flow the solder joints.

Does the problem I describe sound like a cold solder joint?  Seeing the LED flickering had me thinking it was a power issue of some type.  Thanks for your input!

-Jeff

Bob Latino wrote:If the sound fades on BOTH channels then you look for a problem that affects both channels like the main power supply. Probably a cold solder connection on the power transformer or the main power supply capacitors. Try resoldering all power transformer connections and main power supply cap connections ..

Bob

Bob, good suggestion.  With both channels fading & the LED flickering I was thinking power supply also.  As I mentioned, I had one channel acting up previously so I was able to concentrate on that side of the board.  I'll do some work on the preamp this weekend.  Thanks for the advice!

-Jeff

The LED is powered by the filament power supply.  Loss of filament power would explain both channels fading out together.  If it is not a bad connection somewhere in the filament power supply, it is possible, but probably unlikely, that the filament voltage regulator is bad.  Also check the filament power supply output to make sure the voltage is correct and ripple is low (should only be a few mV).

If you feel safe working inside powered-on equipment you might consider what has worked for me. I have a wooden (birch?) tongue depressor with one end cut to a point. I use this as a probe to locate cold or fractured solder joints.

j beede wrote:If you feel safe working inside powered-on equipment you might consider what has worked for me. I have a wooden (birch?) tongue depressor with one end cut to a point. I use this as a probe to locate cold or fractured solder joints.

Thanks for the tip!  I used that method once on a phono preamp & it worked.  I did poke around the power section on my SP14 with no luck.  

This morning I resoldered the parts that might affect both channels.  I thought it may have worked but after 50 minutes both channels faded away and the hum returned.  This time the LED went dark and I noticed the tubes were cool to the touch.  

I just finished taking the board out and re-flowing all the joints.  Right now it sounds great, listening to Nick Drake.  My fingers are crossed.  I'll update later tonight.

Thanks everyone!

Since reflowing all the solder joints the SP14 has been working like a champ. Sounds great! Thanks to all for their help.

Jeff V. wrote:Since reflowing all the solder joints the SP14 has been working like a champ.  Sounds great!  Thanks to all for their help.

DAMN!  Literally after posting the preamp was working the hum started again and the volume faded.  It played perfectly for over 8 hrs after working on it.

Not sure if one of my components could be overheating or is just bad.  Not sure how to proceed.  Any thoughts how I may be able to isolate the trouble further?  Thanks!

-Jeff

turbotoy wrote:The LED is powered by the filament power supply.  Loss of filament power would explain both channels fading out together.  If it is not a bad connection somewhere in the filament power supply, it is possible, but probably unlikely, that the filament voltage regulator is bad.  Also check the filament power supply output to make sure the voltage is correct and ripple is low (should only be a few mV).

Turbotoy, I just saw your post.  No clue how I missed it until now.  Thanks for the info.  I need to look into this further.  To check to filament PS voltage do I check where it terminates on the board?  Also how is ripple measured?  All I have is a DVM.  The only connection I haven't checked is the splice I made on the transformer mains.  Everything keeps pointing to the power section.

Actaully my previous reply was a little in error, as the SP14 has independent filament power supplies for each channel (you tapped off one to power the LED).  As you said both channels faded out at eactly the same time, then the source of the issue must be upstream of both filament power supplies, which is really only the filament transformer and any wiring before that.  Some transformers have built in thermal resets that trip off at high temperature, and are self resetting.  Is the filament transformer getting hot?  If so, the likely cause would be excess filament current draw.  Check for solder bridges to ground anywhere in both filament circuits.  You could also measure the current flowing through the filament transformer to make sure it is as expected.  Also, check your filament voltage on each channel, both using the DC setting on your DMM, and also using the AC setting.  Looking for ripple using the AC setting of a multimeter is rather crude compared to a scope, but it can sometimes be informative on a relative basis.  If one filament supply circuit has greater ripple (i.e. greater measured AC voltage), the problem is likely in that filament power supply.  Also see if one filament voltage regulator heat sink is hotter than the other, which may be the case if it were flowing more current.

Thanks, very insightful Turboboy!  So this is what I'v found.  Checking the right channel side  filament voltage is low at 3.5V and the other side reads 5.5V.  This is when the preamp is working.  Once it stops working properly the right reads 1.2V and the left jumps to 7.5V.    
If I power the preamp down for 5 seconds it seems to reset and starts working again.  Also, it may be a coincidence, but when I have the cover off it will run trouble free longer than with the cover on.  Or at least that has happened a few times.  Does this sound like a transformer issue?  I'm using a Antek toroidal.  

-Jeff

It doesn't sound like a transformer issue since you are still measuring some voltage after the issue occurs. The first thing I would do is figure out why the right channel filament voltage is low at 3.5V; something is wrong there. Make sure all the components in the right side filament power supply are correct, including the voltage divider resistors to set the output voltage from the regulator (R20/R21). What is the voltage at the input to the regulator (at C36)? If the set resistor(s) are wrong, you may be dropping too much voltage across the regulator causing it to overheat. Compare all voltages to the left side filament power supply and try to track down where the differences are. Also inspect all the right filament connections downstream of the power supply as well for anything that may be causing excessive current draw.

turbotoy wrote:It doesn't sound like a transformer issue since you are still measuring some voltage after the issue occurs.  The first thing I would do is figure out why the right channel filament voltage is low at 3.5V; something is wrong there.  Make sure all the components in the right side filament power supply are correct, including the voltage divider resistors to set the output voltage from the regulator (R20/R21).  What is the voltage at the input to the regulator (at C36)?   If the set resistor(s) are wrong, you may be dropping too much voltage across the regulator causing it to overheat.  Compare all voltages to the left side filament power supply and try to track down where the differences are.  Also inspect all the right filament connections downstream of the power supply as well for anything that may be causing excessive current draw.    

Again, very good info.  Strangely, both R20 & R21 on both sides are measuring 95ohms.  Way out of spec.  I'll remeasure in the morning & check the bottom of the board and make sure there is no bridged joints.  Not sure if these resistors are connected in the circuit.  Thanks, very much appreciated!

Reading R20 & R21 I get 95.5ohms.  Being that the values of the resistors are 470, 120 the readings are showing the resistors in parallel.  Should I be seeing this?
When I read filament voltage I'm testing where the transformer connects to the board.  The connections should be 6.5VAC.  Since I'm testing the connection on the board, will the power supply components affect the voltage I see?
I hope to test more when I get home form work.  Thanks!

turbotoy wrote:It doesn't sound like a transformer issue since you are still measuring some voltage after the issue occurs.  The first thing I would do is figure out why the right channel filament voltage is low at 3.5V; something is wrong there.  Make sure all the components in the right side filament power supply are correct, including the voltage divider resistors to set the output voltage from the regulator (R20/R21).  What is the voltage at the input to the regulator (at C36)?   If the set resistor(s) are wrong, you may be dropping too much voltage across the regulator causing it to overheat.  Compare all voltages to the left side filament power supply and try to track down where the differences are.  Also inspect all the right filament connections downstream of the power supply as well for anything that may be causing excessive current draw.    

Just tested a bit and found, while the preamp is on, the left heat-sink is blazing hot and the right is cool to the touch.
Voltage at C36 is 71VDC at Neg & 85VDC at Pos Left Side.  Right Side 74VDC Neg, 81VDC Pos.  Not sure what would cause the difference with the heat-sinks?  The filament transformer did not feel hot.  Kind of lost at this point.

-Jeff

I'm away on vacation for a week with very limited web access, but it definitely sounds like a case of the filament regulator(s)
overheating. Each one should have 9vac applied to it, giving around 11vdc before the regulator knocks it down to 6.3vdc.
The regulators will shut down when overheated. Make sure they are not getting too much voltage, and that the heatsinks are getting plenty of cooling. This hasn't been a problem before but something I have anticipated for those without proper ventilation.
Keep the cover off especially if running more than an hour, and when I return from vacation I'll get back with you on some ways around this problem.

Thanks Roy! I'll send you an email. Enjoy your vacation!

-Jeff

The SP14 is singing again. It was a mistake I made that caused all the trouble. Once the problem was isolated to the filament voltage I took a better look at my work. I found the filament transformer wires weren't paired properly. I swapped the position of the blue leads and now the SP14 is working like a champ! Thanks to all who helped!

-Jeff

I was excitedly following, happy that you solved the problem and started to enjoy your job..