Jump to content

Sub-10 / Sub-12 / RPW-10 Repair Blog


ngen33r

Recommended Posts

Hello All

I have been repairing Klipsch subwoofer amps for about 15 years as a hobby. I have decided to start blogging my repairs and eventually do videos of each one. This thread is intended to be a blog and a resource for information. I will try to answer repair questions as best I can. Comments and tips are also welcome, If you do not have the experience or the tools, PLEASE do not attempt any of these repairs. You will only end up damaging the board and it will end up costing more for a tech to repair the damage. If you do not have a high quality vacuum desoldering station (Hakko or Weller) and a current limited mains supply, you should not be working on these amps. These subwoofers do not have any user serviceable parts inside. If you open up the sub or attempt any repair you see in this thread, you are doing so at your own risk!!!

 

The PDC board 660038RA (main supply control) is almost always the main failure mode of these amps. The glue that was used becomes conductive and kills the board and the fets (planned obsolescence?). These boards are typically made in China and are super easy to damage. If you don't have a desoldering tool you MUST cut the board off and remove the legs from the thru holes one by one. Once the thru holes are damaged the repair becomes a difficult hack job unless you have a thru hole repair kit.

 

The other most frequent failures are the Bash control board and Q5. If Q5 is SOT-23, it needs to be replaced with a TO-92 2N4401. This mod was done on later revisions most likely due to warranty claims. 

 

I have made a schematic of the PDC board and I also have made replacement boards (I do not sell or supply these boards). Sometimes they are so badly burnt, they have to just be replaced.

 

ALWAYS REMEMBER TO UNPLUG OR SWITCH OFF YOUR AMPS. CAPACITORS HAVE A LIFESPAN AND WHEN POWER IS APPLIED THE CLOCK IS TICKING. THE AUTO ON/OFF FUNCTION DOES NOT PREVENT THIS ON MOST MODELS!!!!!!!

 

660038RA.jpg

PDC.JPG

Edited by ngen33r
Link to comment
Share on other sites

SUB-12 Plate serial number 06310349

 

This amp had no power. The PDC board was shorted, which blew the transformer drive fets and was causing the fuse to blow. The PDC was removed, cleaned and repaired. Q5 was upgraded to a 2N4401 and the amp was fully recapped with Nichicon caps. Gotta love those JUNFU capacitors. It's fun to say, but that't about the only good thing about them. The SUB-12 amp is my favorite BASH (Klipsch) amp, even with all the issues they have. They really put out a good amount of power. It is a good idea to silicone the replacement parts in place. You don't need to go crazy but it helps to have them secured.I use GE silicone 2. Q5 definitely needs to be secured at the head and also over the pads and solder joints. Epoxy over the solder joints would probably be better, but the silicone works just fine.

 

20191124_024744.jpg

Edited by ngen33r
Link to comment
Share on other sites

1 hour ago, carlthess40 said:

Have you ever done or seen the LF-10 sub?
I think Klipsch only made it for one year
One 10” and two 10” pas drivers and a 500 rms amp. Mine is dead , green light comes on but no sound


Sent from my iPhone using Tapatalk

I have not worked on or seen one of those yet. If it is a BASH amp, you can be sure it is a glue or capacitor issue.

  • Like 1
Link to comment
Share on other sites

SUB-10 Plate serial number 06230172

 

This amp also power to the secondary side of the transformer. The +-15V supplies were stable, but the outputs did not have power. The BASH module was not working. At idle the rails should have around 5V across them with the volume at 0. Pin 5 of the bash module should have around 12V between pin 11. This module has the full ~90V rail voltage on it so don't poke around with the power on. Q5 was the issue and replaced with a TO-92 2N4401. From the heat generated the trace lifted of the board but I was able to carefully remove the part with hot air and epoxy the pad back down. This repair also got a blob of epoxy to secure the legs of the part from stressing the traces. A full recap of the power supply was done , also all the bad glue removed and it is good to go.

20191124_135800.jpg

Link to comment
Share on other sites

  • 1 month later...

SYNERGY SUB-12 PLATE SERIAL 06230184

 

This amp was blowing fuses. The PDC control board failed and took out the mosfets. I had to clean and rebuild the PDC, replace the fets, replace Q5 and TH3. TH3 is almost always bad when fuses are blowing. After the board was working, a full recap was done on the amp.The active crossover was especially fun due to all the damn glue used.

20200113_220218.jpg

20200113_220202.jpg

Link to comment
Share on other sites

  • 2 weeks later...

SYNERGY SUB-10 PLATE SERIAL 05420831

 

This one looks a little brown and crispy. The back of the red film capacitors also has some nice black magic smoke on it. Q5 is turning brown from overheating. This will need a full recap, removal of all the glue, Q5 and fets replaced and rebuild of the PDC board.

20200118_011549.jpg

Link to comment
Share on other sites

SYNERGY SUB-12 PLATE SERIAL Y9154604807

 

This is one of the nicest Sub-12 revisions that I have seen. It has all the updates and used a silicone based glue. This one died from a surge or spike. It will get a repair and recap.

20200119_174727.jpg

20200119_174745.jpg

20200121_214212.jpg

Edited by ngen33r
Link to comment
Share on other sites

SYNERGY SUB-10 PLATE SERIAL C707360626

SYNERGY SUB-12 PLATE SERIAL ?????

 

This set was sent in for me to go over. The Sub-12 looked like it was in a war. All of the wires were cut, the fuse holder was stuffed with aluminum foil and plenty of parts were deep fried and that is all after the failure from the glue monster. The Sub-10 was not in terrible shape, but it also good and crispy around the glue and PDC board. The owner couldn't figure out how to remove the speaker wire so why not just rip it out of the terminal. These terminals use locking tabs.

 

These are the before pictures

download.jpg

20200122_232828.jpg

20200123_214920.jpg

Edited by ngen33r
Link to comment
Share on other sites

1 hour ago, ngen33r said:

SYNERGY SUB-12 PLATE SERIAL Y9154604807

 

This is one of the nicest Sub-12 revisions that I have seen. It has all the updates and used a silicone based glue.

I have a working sub12. Would you recommend the revisions you mention as preventative maintenance? It seems to me a good idea would be to remove and replace the glue at the very least. If so, could you make a list of the revisions you recommend?

Link to comment
Share on other sites

21 minutes ago, MechEngVic said:

I have a working sub12. Would you recommend the revisions you mention as preventative maintenance? It seems to me a good idea would be to remove and replace the glue at the very least. If so, could you make a list of the revisions you recommend?

 

This is what I would consider an absolute MUST and in this order.

1 - Have the proper tools to do the job. (Temperature controlled rework station % Vacuum desoldering tool)

2 - Remove any glue that touches the PCB & component legs. 

3 - Replace all the Chinese capacitors with the Tier 1 caps of your choice

4 - Replace Q5 if it is an SOT-23 part with a TO-92 part. 2N4401

 

The glue softens at 155C much over that and traces start delaminating. I use GE silicone II DON'T USE SILICONE I

 

 

 

  • Like 2
Link to comment
Share on other sites

  • 1 month later...

Hello ngen33r, I am posting here in your thread to hopefully get some help.  I have a RPW-10 that worked great for years and then no longer was working...I unfortunately left the switch in the "auto on" position.  I don't use the surround sound system on this TV often, so i do not know how long ago it stopped working...it may have been several months ago!  I found the fuse was blown, replaced it and it blew immediately.  I removed the plate amp, but I did not see anything obvious in regards to blown components in the power supply.  A Google search brought me to this thread and I was on my way!  I followed every suggestion in this thread: 1) rebuilt PDC, 2) replaced mosfets, 3) full recap, 4) replaced TH3, and eventually upgrading Q5 to 2N4401 from SOT 23. 

 

Rebuilding the PDC was done based on your schematic in the first post of this thread.  However, the values of my original resistors were totally different from those in your schematic---although the board number was exactly the same (660038RA).  See pictures below:

 

Original board:

g2fiEMc.jpg?1

 

Rebuilt PDC:

7GOfk1x.jpg?1

 

I also replaced the two IRF730 Mosfets with two IRF740 Mosfet transistors.  I also replaced TH3 (DSP104) even though there was no damage to it and then I replaced all CAPs with Nichon and Kemet (470uf) 105 C caps except I did not change the two 35V 4.7 non-polar caps at C48 and C53 (I ordered regular polar caps by mistake).  In addition to all of the suggestions you made in this thread, I also (6) replaced the two voltage regulators at U4 and U5, (7) replaced the transistor at Q8, and for good measure (9) I changed out all 5 of the output transistors (IRFZ14).  The picture below summarizes all of the changes:

 

Summary of All changes (before changing out Q5):

WH35Ne4.jpg

 

 

Now, I am no longer blowing fuses but the amp is still not working.  After a short period of being on, the Mosfets really start to heat up...I am not noticing any other components getting warm and there is absolutely no speaker noise.  I am now stuck and I am almost tempted to replace the safety capacitors (yellow thingys) and the bridge rectifier.  However, I wanted to reach out to you first for some guidance.  I have no schematic and I haven't got the slightest clue how to trouble shoot without knowing voltage "test points".  Can you please give some advice?  I am tempted to throw in the towel and but a new plate amp replacement, but I feel like I am too far into this to quit now...do you have any advice for a hobbyist?  Thanks in advance!

Edited by wderbi
Link to comment
Share on other sites

 

THE RESISTORS ARE IN PARALLEL ON THE BOARD AND WILL MEASURE THEIR COMBINED PARALLEL VALUE. YOU MUST USE THE VALUES THAT I HAVE LISTED OR THE SUPPLY WILL NOT FUNCTION PROPERLY.

 

The problem is either with the DIAC on the PDC board, Q5 by the BASH board, or the controller on the BASH board. 

 

Where did you get the DB6 diac? NTE is the only place I could find the correct one.

 

Things to check (ONLY CHECK THESE IF THE PDC HAS THE COMPONENTS THAT I LISTED INSTALLED)

 

1 VOLTAGE ACROSS THE 2 REGULATOR HEATSINKS SHOULD BE ~28V

If you have this voltage the primary supply is working. If you don't have this the PDC is still not working.

 

2 VOLTAGE ACROSS BETWEEN THE 2 LARGE RESISTORS BY THE RIBBON CABLE

Once side of the resistors is common make sure you measure the correct side. Should be ~5V with no input and volume at 0. If the voltage is around .5V then the bash board circuits are the problem. 

 

 

Pin 5 on the BASH board should be 12V if the surrounding circuits are working. Be careful this board also sees full rail voltage. If you zap yourself, you are doing this at your own risk. 

 

Cut Q5 off, don't try to desolder it if it tests bad.

 

 

You will have to adjust the bias since you replaced the output transistors. This means the 2 pots have to be replaced and you need an oscilloscope. 

 

The VREGS by the capacitors should be bent away from the caps and they do get pretty hot. That is normal.

 

 

 

 

 

Edited by ngen33r
Link to comment
Share on other sites

Thank you for chiming in! I am just now seeing this after midnite; therefore, I won't be able to check out your queries in detail until later today.  I rebuilt the PDC and replaced ALL components exactly as you specify despite me noticing it was different resistor values from the original PDC in my unit. The diac I purchased was a NTE6412 as suggested by this thread:

I did every step (steps 1-9) in the above post BEFORE replacing Q5 which was originally a SOT23 transistor.  I reassembled everything and connected the plate amp to the speaker.  The fuse did not blow, but the amp didn't work.  II noticed the heatsinks connected to the IRF740s (mosfets) got pretty hot in less than 3 minutes it was connected.  After things did not work, I realized I did not swap out Q5.  I removed the Bash board and used my heat station to remove Q5---I did not test it beforehand, nor did I realize I should have cut it off!  It came off pretty cleanly and I added a bit of solder to each pad.  Unfortunately, I had real hell trying to line up the legs of the 2N4401 in place so that I could solder the transistor in place.  In my attempt to do so, the pad for the collector was pulled off the board by the leg of the 2N4401 that I soldered down to that pad. I removed the 2N4401and instead decided to solder flexible wire to each pad, and solder 2N4401 to the wires instead.   Because the collector pad was messed up, I traced the connection back to the cathode side of diode D7 and soldered a wire for the 2N4401 collector to that connection.  I hot glued the 2N4401 in place and reinstalled the Bash board. The picture below is the new Q5 installed before reattaching the Bash board:

 

Q5 with 2N4401 connected by wires:

CnAmKfq.jpg

 

Unfortunately, after installing the updated Q5, the results were the exact same: Amp not working, IRF740s heating up rapidly.

 

You have given me some test points to look at.  I have removed the safety capacitor at C21 to get the value so that I could order replacements if necessary. I will reinstall it later today and I will check voltages across the heat sinks between U4 and U5  (vREGS) and between R78 and R71 as you outline above in steps 1 and 2, respectively.  I am assuming DC voltages...correct?  Also, am I correct to assume that pin 5 of the BASH board is counting from the side that is closest to Q5?

 

Thanks for the input bro!

Edited by wderbi
  • Like 1
Link to comment
Share on other sites

Hah!  Good to know that you are also OUNVME as well!😃

 

Here is an update:  I reinstalled the safety capacitor and connected the mains and the ribbon cable.  I checked the voltage between the Vregs and it was 27.2V.  The voltage between R71 and R78 was 5.17 when the gain was turned all the way down and no input connected. The voltage between pin #5 and pin #11 on the BASH board was 12.62...The voltage between pin #5 and ground was like 68V.  Based on those numbers, everything seems to check out!

 

However, the amp was not working and the heatsink for the IRF740 mosfets was getting pretty warm so I shut everything down after about 2.5 minutes checking voltages.  I turned off the amp, unplugged it, let everything stay unplugged and in the on position for about 25 minutes to drain caps and let everything cool down.  I next wanted to check the temperature of the heatsink connected to the IRF740 mosfets for 10 minutes after turning the amp on.  Below is a chart showing the heatsink temperature measured by a laser temperature gauge where I checked the temperature every minute:

 

MOSFET heatsink temperature chart:

aseRixB.jpg

 

As shown, the peak temperature of about 150 F was reached after being turned on for 5 minutes.  However, I noticed that the temp started slowly dropping for the next few minutes.😲 Unfortunately, I then noticed that the power light was turned off and that the fuse (2A) was blown. 😞  I am out of 2A fuses so I am ordering new 2AL fuses from Amazon to replace the 2A fuse.  I don't think the 2A fuse blew after 5 minutes because it wasn't a long-delay fuse and I fear that something else is going on.  Any other suggestions???

 

Thanks!!!

Edited by wderbi
Link to comment
Share on other sites

The primary switching transistors should not be getting warm at idle if they are on the heatsink. You have some sort of issue with the output mosfets or the bias adjustment. At this point you have to start probing the power and signal path with a scope.

Link to comment
Share on other sites

Hmmm...that is interesting.  The output mosfets have all been replaced and I recall from your earlier post that the bias was going to have to be adjusted with the scope.  However, would the bias being out of adjustment be the cause of the primary switching transistors to heat up and eventually blow the fuse (after 4-5 minutes) at idle with the gain set at zero? I thought that setting the bias correctly was a way to trim/adjust the gain to prevent clipping and/or speaker damage?  I would not have guessed it would have such a dramatic impact on the primary side of the power supply and/or causing fuses to blow. 

 

Using an O-scope to diagnose issues with electronics is a black box for me and at this point is quite a bit beyond my expertise.  I do know that having the knowledge to scope is an extremely powerful tool to have in the toolbox in the electronics repair world.  I know just enough to be dangerous when it comes to electronics and I have dabbled in and out of it as one of my hobbies over the years; each time learning more and more.  Now that I unexpectedly (and honestly unwillingly) have time on my hands, I was planning to use this amp repair project to further delve into this hobby a little more.  For this project I already invested in a new Hakko FR-301 de-soldering gun (which is great BTW---makes the de-soldering process so much easier).  However, I do not have a scope, but I was thinking about grabbing a LIUMY Professional LED Handheld Oscilloscope Multimeter from Amazon hxxps://www.amazon.com/Oscilloscope-Multimeter-LIUMY-Professional-Scopemeter/dp/B071F1H3PG/ref=sr_1_10?dchild=1&keywords=Oscilloscope&qid=1584052295&s=industrial&sr=1-10 and try learning how to use a scope to diagnose problems by looking through some YouTube tutorials. 

 

However, despite the likelihood that I will learn something valuable, part of me is worried that I have reached the point of diminished returns on this project already and I wonder if its time to ship the board off to the pros...especially if the feedback on that scope from Amazon is that it is not very good for this project. I guess if I had a service guide or schematic I'd feel much more confident.  I don't want to throw in the towel, but scoping to me sounds really intimidating!  As an option, would it be more cost effective to just send it to someone (like you?) and have the troubleshooting done with a scope?

 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...