Badcaps.net Forum
Go Back   Badcaps Forums > Troubleshooting Hardware & Devices and Electronics Theory > Troubleshooting Audio Equipment
Register FAQ Calendar Search Today's Posts Mark Forums Read

 
Thread Tools Display Modes
Old 12-15-2018, 11:32 PM   #1
momaka
master hoarder
 
momaka's Avatar
 
Join Date: May 2008
City & State: VA (NoVA)
My Country: U.S.A.
Line Voltage: 120 VAC, 60 Hz
I'm a: Hobbyist Tech
Posts: 10,534
Post JBL S412PII Speakers – Distorted Bass Repair and Updated Schematics

I had a pair of these JBL S412PII speakers for repair last year. Finally got to typing up the post for this, though (warning: long posts ahead, split into 5 parts!)

A little info about the speakers…
These speaker towers have four speaker drivers each: treble, mid-range, mid-bass, and woofer. First three are connected together to the speaker terminals (through a cross-over network for each driver). The woofer, however, has its own amplifier inside the cabinet (often referred to as the “plate amp” online), which receives its signal either from the speaker terminals above or via a single RCA input, labeled “LFE/subwoofer”.
The owner said he bought the speakers brand new way back in the early 2000’s.
They have been in service pretty much 24/7 since then.
Issue/problem appeared after they lost power in the house one day

The problem with the set I got is that one of the speaker towers had its woofer produce garbled and distorted bass. As I suspected, the issue was with the woofer amplifier, which I verified by swapping woofer drivers from the “good” tower to the one with the distorted bass (neither drivers showed distorted bass in the “good” tower.) Furthermore, the distorted bass didn’t quite sound like a blown woofer, as the owner thought. Rather, there was a 120 Hz buzz heard over the driver when it was touched or pushed, confirming an issue with the built-in woofer amp.

Once I opened the amplifier box to take a quick look, it didn’t take much time to notice how darkened the PCB was from heat. I also noted that there was our beloved tan conductive glue on some spots (and fully baked on one). To top it all off, the board was populated exclusively by Su’scon and CapXon 85°C electrolytic caps, some of which darkened by the heat too. Hmmm, what could possibly be wrong with this amp?

It turns out, however, that the caps and glue weren’t the only issues. Closer look later on revealed poor thermal design on the PCB in general (“thermal design”, what’s that? ), along with a few mismatches with the service manual. So for this reason, I am going to split this repair thread into several parts. Feel free to skip to whichever part is of interest to you, as I tend to be a bit long-winded.

For those who have this amp and are looking to get a quick fix for the same issue (without reading this entire thread): just make sure to do the following items.
- Change all electrolytic capacitors - big and small. (In my case, it was one of the big filter caps that had gone completely open-circuit and responsible for the distortion.)
- Remove all tan glue that has turned black/brown (it goes conductive over time!)
- Reflow all solder joints (particularly on all transistors and large parts.)

The items above should take care of the distorted bass and possibly other issues too. In addition to that, it may also help if transistors Q131 and Q134 are heat-sinked (these generate the ±15V rails and run very hot.) For more details on these speakers, read on.
momaka is offline   Reply With Quote
Old 12-15-2018, 11:46 PM   #2
momaka
master hoarder
 
momaka's Avatar
 
Join Date: May 2008
City & State: VA (NoVA)
My Country: U.S.A.
Line Voltage: 120 VAC, 60 Hz
I'm a: Hobbyist Tech
Posts: 10,534
Post JBL S412PII Speakers – Part 1

Part 1 – Disassembly and Pictures of the Build Quality

I’ll start with pictures - the nonworking amplifier in its case once removed from the speaker tower:



We have us a nicely sealed box to make sure all heat stays inside – a great way to cook components.

Next step, opening the case… and a word of caution about that: do NOT remove all screws on the back of the amplifier box. Some of them simply hold the heats-inked parts (these can stay put). And some of them only hold the plastic back of the box. You can see in the first picture above which ones I had to remove to get the amplifier box removed from the speaker towers.

Here’s what inside the amp box:

The amplifier is powered by a classic line transformer. There are two boards inside the woofer amp case. The bottom one is for the input signal filters and On/Off controls. The top (bigger) board contains the amplifier and protections, as well as power supply rectifiers and filter caps.

Let’s have a look at the top board first (i.e. the output amp board).
https://www.badcaps.net/forum/attach...1&d=1544938622
https://www.badcaps.net/forum/attach...1&d=1544938622
https://www.badcaps.net/forum/attach...1&d=1544938622

Seems like it gets toasty in there, doesn’t it? . Well, the heat certainly wasn’t kind to the cheap Phenolic PCB that this amp uses, especially after so many years. If only there was a way to attach a sample of the smell too, I think you would have gotten an even more complete picture. I actually had to move the amp board in the garage between work sessions, as the smell from the PCB was chokingly potent – 10 minutes was enough to stink up my entire work area for the day.

However, the PCB wasn’t the only one that suffered from the heat. Quite a few of the small caps had discolored sleeves too, especially those located in “hot spots”. Speaking of caps, all electrolytic capacitors used in this amp are general purpose 85°C and either CapXon GS series or Su’scon SL series. To make matters worse, you can see there is some tan glue on the board. This tan glue is known to go conductive with time and added heat. It was completely turned black and even gone slightly conductive on a small add-on PCB. Using my multimeter, I could easily read several hundred kOhms across any two points on the glue!

Note: the picture shows over 6 MOhms simply because I couldn’t hold the multimeter probes too well while taking a picture with the camera. Nonetheless, even 6 MOhms is not good!

Speaking of the small PCB which had the black glue on it… it was placed on a spot marked for transistor Q125. But instead of a simple transistor, the board had this thing:
https://www.badcaps.net/forum/attach...1&d=1544938622
https://www.badcaps.net/forum/attach...1&d=1544938622
Interesting factory “fix”, I must say.

Moving on, here is a picture of the solder side of the main amp board:
https://www.badcaps.net/forum/attach...1&d=1544938720
If you’re wondering why those joints look so shiny, it’s because this is a pre-RoHS board with nice Pb-based solder. Overall, it looks good. But looking more closely, I found ring cracks around many of the heat sink -attached components. There were also a lot of under-soldered joints. Unfortunately, neither of these were the issue behind the amp distortion or the amp not turning On, as I found out later during testing. Now, you might have noticed that the solder joints for C119 through C122 look disturbed – that’s because I took this picture after I did a bit of troubleshooting on the amp already (to be discussed in section 3.)

And here is the input signal / control board:
https://www.badcaps.net/forum/attach...1&d=1544938720
https://www.badcaps.net/forum/attach...1&d=1544938720
The two TL074 quad-amplifier ICs are used for the input signal filtering and amplification as well as handling On/Off logic and clipping protection. Overall, this board looks clean without any overheated components or messy factory “workarounds”. Only noticeable thing is the jumper at switch SW201, labeled “AC-3”. I’m not entirely sure what this does, but on the schematic, it looks like a CR high-pass filter. My guess would be this circuit was meant to filter out very low sub frequencies. Either way, this frees up one of the amplifier blocks in U201 (in case anyone wants to do any mods with it.)

Now, if there is one thing I really don’t like about this amplifier, it is the linear regulator circuit that generates the ±15V rails. It dissipates huge amounts of heat, because it uses the ±VL rails as the source. These are about ±30V each. So with that kind of voltage drop, even 100 mA of current draw will dissipate 1.5 Watts on a non-heat sunk TO-220 component. That is *not* a good design, folks! On top of that, the linear regulators are always running, because they supply power to the On/Off logic. This means constant heat to surrounding components on the PCB. No wonder some caps looked cooked!

Moreover, there is no separate stand-by power supply to feed the ±15V rails when the speakers are “Off” or in Stand-by mode. As such, that big beefy transformer is connected to the line all the time (read: more power wasted.) The worst part about this: there is no circuit or component to disconnect the transformer outputs in case of failure in the amplifier. Therefore, a hard failure will require the transformer to burn out its internal thermal/overload fuse. Seriously, folks, this is very poor design!
Attached Images
File Type: jpg JBL S412PII woofer amplifier (1).jpg (189.3 KB, 96 views)
File Type: jpg JBL S412PII woofer amplifier (2).jpg (79.0 KB, 95 views)
File Type: jpg JBL S412PII woofer amplifier (3).jpg (88.2 KB, 95 views)
File Type: jpg JBL S412PII woofer amplifier (6).jpg (109.9 KB, 93 views)
File Type: jpg JBL S412PII woofer amplifier (4).jpg (289.1 KB, 23 views)
File Type: jpg JBL S412PII woofer amplifier (5).jpg (336.4 KB, 20 views)
File Type: jpg JBL S412PII woofer amplifier (7).jpg (121.7 KB, 20 views)
File Type: jpg JBL S412PII woofer amplifier conductive glue.jpg (122.9 KB, 95 views)
File Type: jpg JBL S412PII woofer amplifier - Q128 assy (1).jpg (110.4 KB, 25 views)
File Type: jpg JBL S412PII woofer amplifier - Q128 assy (2).jpg (83.3 KB, 19 views)
File Type: jpg JBL S412PII woofer amplifier (11).jpg (330.6 KB, 18 views)
File Type: jpg JBL S412PII woofer amplifier (9).jpg (327.4 KB, 20 views)
File Type: jpg JBL S412PII woofer amplifier (10).jpg (298.1 KB, 11 views)
momaka is offline   Reply With Quote
Old 12-15-2018, 11:56 PM   #3
momaka
master hoarder
 
momaka's Avatar
 
Join Date: May 2008
City & State: VA (NoVA)
My Country: U.S.A.
Line Voltage: 120 VAC, 60 Hz
I'm a: Hobbyist Tech
Posts: 10,534
Post Re: JBL S412PII Speakers – Part 2

Part 2 – Service Manual and Updated/Corrected Schematics

As I started troubleshooting the built-in woofer amplifier for these speakers, I ran into a few issues and decided to look up a service manual for this model. This is where I found the service manual didn’t exactly agree with the circuit I had in hand in several places. A more thorough check revealed the service manual schematics actually had a few errors. On top of that, the original schematics do not show where traces cross each other and where they connect (with a dot). This can make reading of the schematics a bit difficult and confusing. Therefore, I decided to correct these discrepancies, including adding the connecting dots. Revised pages of the service manual are attached below. Please don’t assume my revisions are 100% error-free. However, I did check them several times and from what I can see, there aren’t any.
https://www.badcaps.net/forum/attach...1&d=1544939614
https://www.badcaps.net/forum/attach...1&d=1544939614
https://www.badcaps.net/forum/attach...1&d=1544939614

The schematic corrections above address the following:
• [pg. 20 of PDF]: On board connector P101, the PCT line should connect to PROT (pin 3) and *NOT* the MUTE signal (pin 2) as the schematic shows.
• [pg. 19 of PDF]: The designators of caps C127 and C128 should be switched around (as shown with red). C127 filters the -15V rail and C128 filters the +15V rail.
• [pg. 19 of PDF]: Remember the small “factory fix” PCB I talked about above that’s in place of Q125? In the schematic, Q125 is shown as a 2SC1815 (NPN) transistor. As I showed, however, Q125 is not actually a transistor, but a whole add-on PCB with two TO-92 transistors and a wire running to the -15V rail. This is the diagram I was able to trace for it:
https://www.badcaps.net/forum/attach...1&d=1544939614
• [pg. 21 of PDF]: Signal board schematic shows there is a circuit for choosing between “Normal” and “AC-3” sound (which consists of switch SW201 and one of the op-amps of U201). This circuit wasn’t implemented on my signal board. Instead, the signal is by-passed with jumpers directly to R210.
• [pg. 21 of PDF]: Some of the resistor and cap values on my signal board didn’t quite match what the schematic in the service manual shows. I crossed out the SM values and added mine in red. For resistors R249 an R250, I couldn’t even find them on the board. Not sure if these actually exist or not. Could be my mistake, though.
• [pg. 21 of PDF]: U202, block A of TL074 op-amp – polarity of inverting and non-inverting inputs appear to be switched around.

Another interesting thing I noted is that my amplifier board has 25V / 47 uF electrolytic capacitors, but the schematic shows all of these as 50V / 22 uF. Yet, the parts list on page 15 of the SM shows these as 25V, 47 uF as well. So it looks like 25V, 47 uF is the correct part, but perhaps the schematics just weren’t updated? Anyways, that is all I found so far. Original service manual / schematics also attached below.
Attached Images
File Type: png JBL S412PII sm pg19 corrections.png (184.7 KB, 20 views)
File Type: png JBL S412PII sm pg20 corrections.png (168.5 KB, 19 views)
File Type: png JBL S412PII sm pg21 corrections.png (135.1 KB, 20 views)
File Type: png JBL S412PII ''Q125'' circuit.PNG (16.2 KB, 17 views)
Attached Files
File Type: pdf JBL S412P studio series - service manual.pdf (915.0 KB, 13 views)

Last edited by momaka; 12-16-2018 at 12:00 AM..
momaka is offline   Reply With Quote
Old 12-16-2018, 12:10 AM   #4
momaka
master hoarder
 
momaka's Avatar
 
Join Date: May 2008
City & State: VA (NoVA)
My Country: U.S.A.
Line Voltage: 120 VAC, 60 Hz
I'm a: Hobbyist Tech
Posts: 10,534
Post JBL S412PII Speakers – Part 3

Part 3 – Troubleshooting and Repair

Before cracking the woofer amplifier open, I decided to test it one more time, but with my own input signal, just to verify the issue was the amp and not the signal source at the owner’s house (slim chance, but just in case). I used a 3.5 mm to RCA cable to feed audio directly to the LFE/Subwoofer input jack on the amp, similar to what the owner was doing. Then, I applied power and… NOTHING?!? The amp wasn’t turning ON at all. Normally, there is a speaker output relay that would to engage right away (at least it did in the owner’s house). I changed the “On/Off” mode switch between “Manual” and “Auto”, but that did not do anything.

My initial suspicion, besides bad caps, was the badly baked tan glue, which had become black and slightly conductive in some areas. Removing it was the first thing I did. However, the amplifier still refused to turn ON.

As most of the badly burned glue was on the Q125 add-on board, I also decided to pull it out and test its components. But none were bad. It crossed my mind to just install a regular 2SC1815 transistor for Q125 as the schematic showed. However, as I wasn’t exactly sure what this factory “correction” did at the time and since the speakers made it to over 15 years of service life, I decided to leave it alone. Analyzing the service manual schematic later on revealed that Q125 is part of the speaker DC protection circuit (which simply disconnects the speakers in the presence of DC voltage on the output terminals.)

Next, I knew some of the small caps could be dried out - especially those darkened ones in the heat-affected areas. I pulled a few that appeared to be discolored the most (namely C127, C128, C114, C115, C123, and C125), and put them on my ESR meter. The results didn’t surprise me much, though.

As you can see, quite a few of these had very high ESR (one was even completely open). I replaced them with whatever caps I had on hand (that tested good, of course), then connected the whole amplifier setup for another test and… NOTHING AGAIN!

I figured, since the amp was working at the owner’s house when it was warm, then perhaps warming it up again could bring it back to life. It didn’t make much sense really, because at this point, I already had changed or checked all of the small capacitors on the amp board - there was nothing bad left. Nevertheless, I pulled out my heat gun and blasted both boards with a little bit of heat. Surprisingly (or not, depending on how you look at it), the amplifier turned ON instantly. – About time! However, the distorted bass was still there. Oh well, back to the original problem, at least!

I then noticed the bad solder joints on the heat-sunk components and corrected them. Perhaps I should pause for a word of caution here: make sure the big filter caps are discharged! Normally, I always do this on equipment I work on. And in this case I did as well, but forgot about it when I started working on the amp the next day. Those big caps charged up by themselves overnight to several Volts or more (not unusual for big high-voltage caps.) Luckily nothing got damaged when my iron shorted across Collector and Emitter on one of the transistors and I saw a small spark.

Despite the above hiccup… with the solder joints fixed and most of the small caps either replaced or pulled out and tested, the distorted bass problem persisted, along with the cold-start issue if I left the amp to cool down for a few hours. So at this point, all that was left untouched were the four big filter caps and the small caps on the signal board.

I started pulling small caps from the signal board, but they all tested reasonably okay. (Note: up to 5 Ohms ESR is not necessarily out-of-spec for some small 5x11 mm 85°C GP caps, especially after 15 years). I did replace these too, just in case. Yet, nothing changed.

All I could think of at this point was either an issue with the big filter caps having high ESR (as heating the board made the amp turn ON) or perhaps something wrong with one of the voltage rails. I started by checking the voltages of all the DC power supplies: namely, the ±VH, ±VL, and ±15V supplies. The ±15V supplies tested a bit low and slightly unbalanced, starting at about +13.2V and -13.9V each, and then slowly climbing to about ±14V – not great, maybe even suspicious, but not necessarily bad either. So I moved on. The ±VL rails tested fine: about 29-30V DC with 121V AC line. Then finally the ±VH: this is where things got interesting. I couldn’t test the voltages with the amp plugged in, due to not having safe access to test points. However, I noticed that after turning OFF the amp, the caps filtering the ±VH rails had disproportionate voltages: one about 35V DC, and the other just a few mV. I figured at this point it just made sense to pull the big caps out and really see what they measured. Three of them tested okay for ESR and capacitance, so I put them back in. But one of them did not. Can you guess which one it was?

No? How about now?

I know my camera didn’t take really good pictures there, but from the two caps above, the lower cap was leaking liquid from its bung, where its lead enters (and it wasn’t damage from me removing the cap.) This turned out to be C121 (a CapXon GS 63V, 4700 uF): the one filtering the +HV rail that showed only a few mV after unplugging the amp. Surprised? I’m not! My component tester showed that cap was completely open-circuit. Shame on you, CapXon!... or is it? Read along.

To be fair, let’s not forget these speakers are about 15 years old. Moreover, I found out through some measurements with my multimeter that the transformer’s ±HV output rails (without load @ 122V AC line) would output 43 to 44V AC. That translates to 60-62V DC (unloaded). Folks, 60-62V on a 63V-rated cap is NOT a very good idea, to say the least! Granted the voltage does actually drop a little (to about 59-60V DC) once the transformer is connected to the caps and the amplifier turned ON… but that’s still close. US line voltage can vary +/- 5%, which means the line can be as high as 126 Volts. At that input, the voltage across those caps will likely go past 63 Volts. Sure, electrolytic caps typically specify in their datasheets that they can take up a “surge” over-voltage. But I also remember reading that running a cap continuously near its maximum working voltage can lead to sudden failure in some cases. And I think this might have been the case here. Thus, I am actually surprised these 63V CapXon filter caps lasted so long. That’s not to say I give CapXon an excuse for the failure (I don’t!) However, JBL’s poor design / cost-cutting probably contributed to the failure of this big 63V cap a fair amount, if not entirely. Recall I mentioned that the power supply ±VH and ±VL rails are always ON. This means the filter caps were constantly running all these years.

Needless to say, as soon as I found that 63V filter cap was open, I started looking through my junk bin for a replacement. I didn’t have anything with those capacitance and voltage specs, but I found something “reasonably” close. Or was it?

JEE, what a kludge! Clearly a 200V, 1000 uF cap replacing a 63V 4700 uF cap is far from ideal (unless you’re thinking ideal hack job ). But hey, it still beats an open CapXon, right?

With this done, I plugged the amp in the wall (with a series 450 Watt heating element, of course, to limit current in case of failure – similar to using an incandescent bulb, but allows more power.) But you know what’s coming right? – a big NOTHING again. This was getting old now! A little blast of warm rejuvenating air from my heat gun made the amp turn ON once more. This time, however, NO MORE DISTORTION! The amp sounded clean and I could crank it up without issues.

So in short (*looks at all the text typed above* ) neither the small electrolytic capacitors, nor the burned tan glue, nor the bad solder joints were responsible for the sound distortion in my case. Rather it was all due to the failed 63V 4700 uF filter cap. As for the cold-start problem, that’s to be continued in part 5 below.

Folks, this is why I keep crappy caps, PSUs, and all other sorts of electronic “junk” around. A gutless/crappy computer PSU may not have enough components to be worthwhile fixing up, but it could have just enough components to make something fixable. In this case, it was either this JEE cap hack job or ordering new capacitors and hoping the bass distortion issue would get fixed. Considering the moderate costs of the big caps, I didn’t want to run into a situation where I couldn’t get the issue fixed, yet still would have to bill the owner something to make up for the bought parts. If it was just a bunch of small caps, I would have done it right away and not wasted time replacing caps temporarily. But these big caps are usually $3-5 a piece. Thus, before doing any expensive gambling, I always try to find out what the issue is first. I absolutely hate replacing components blindly. It can rack up costs unnecessarily and not always save time.
momaka is offline   Reply With Quote
Old 12-16-2018, 12:21 AM   #5
momaka
master hoarder
 
momaka's Avatar
 
Join Date: May 2008
City & State: VA (NoVA)
My Country: U.S.A.
Line Voltage: 120 VAC, 60 Hz
I'm a: Hobbyist Tech
Posts: 10,534
Post JBL S412PII Speakers – Part 4

Part 4 – Recap Information and Some Improvements

After thoroughly testing the amp and making sure there was no more distortion, the above prompted me to suggest to the owner that it’s probably a good idea to order caps for the other (working) amplifier in his speaker towers, since a similar failure could be expected from it too. The owner agreed, so I purchased replacement caps for both of his speakers.

Here’s a list of the original caps for both the amplifier and the signal-processing board (this is only for *one* amplifier board and *one* signal board).


Because I needed to replace one of those big filter caps in the PSU, I decided to replace all of them. Moreover, I was very concerned about the ±VH rails hitting close to or over 63V. For this reason, I figured it’s best to go with higher voltage –rated caps for C121 and C122. At first, I looked for 100V ones. But all of the Japanese manufacturers’ offerings were two big (either too high or too wide in diameter) to fit. And the non-Japanese brands were more expensive and I didn’t deem them trustworthy. This is when I found that there are 80V-rated caps. So this is what my replacement list ended up to be:

Worth noting here is that both my replacement 50V and 80V large caps had snap-in terminals rather than normal thin leads. As a result, I actually had to drill new holes in the PCB to install them. But at least the snap-in caps were cheaper. I also personally think they hold stronger on the board (i.e. less likely to pull a lead out if the speakers happen to get in an “accident”.)

Before installing the new caps in the board, I finished cleaning the rest of the tan glue.
https://www.badcaps.net/forum/attach...1&d=1544941204

And here is how the new caps looked on the amp board:

Yes, that’s Rubycon and Nichicon goodness for just about every cap. The only exception was a lone Panasonic SU bi-polar cap. Perhaps sharp eyes may have noticed I put a jumper across cap C113 – that’s because C113 and C114 were originally both 25V/47 uF caps installed in series back-to-back (i.e. negative to negative lead), which I presume was JBL’s cheap (ghetto?!) solution of creating a bi-polar cap. Instead of having two 25V/47 uF caps for C113 and C114 connected back to back, I actually bought a real bi-polar cap: the Panasonic SU mentioned above, rated for 35V/ 47uF.

Aside from the recap, I added some basic aluminum DIY heat sinks on transistors Q131 and Q134 of the amp board. These are responsible for generating the ±15V rails. Because they do so from the ±30V (VL) rails, they are by far the hottest-running components on the PCB. Just take note that if you do make heat sinks for yours, beware that there is *not* a whole a lot of vertical space above them (remember, this plate amplifier has a plastic box.) I didn’t realize this (or rather, forgot about it) and ended up with a box that couldn’t be closed. So I had to open everything again, remove my custom heat sinks, cut them shorter, then reinstall again.

Another mod I did was I replaced resistor R148. Original rating is 330 Ohms, 1 Watt. I used the same resistance (330 Ohms), but upped the power rating to 5 Watts. R148 drives the coil for relay RY101 (part number OMI-212L), whose coil is rated for 45 mA @ 12V. Thus, resistor R148 was dropping a whooping 18V across it and dissipating very close to 1 Watt of power (enough to darken the PCB under it.) This mod isn’t necessary. However, it bugs me when I see parts running right at their limits. The resistor was cheap enough that I didn’t think twice about ordering it.

To top it all off, I added some silicone glue around the big caps and my DIY heat sinks. Here is a final look at the amp board with the recap and the mods mentioned above:


Here’s also a picture of the recapped signal processing board:

I used Nichicon FW for many of the 50V/10 uF caps. Nichicon FW is audio-grade. Generally, there wouldn’t be any problem using regular GP caps in such circuit… or low ESR caps for that matter. But the Nichicon FW caps I found were literally just one or two pennies more per cap compared to the regular GP options. And the signal board does not get hot, so there shouldn’t be any problem with 85°C caps.

Finally, I would like to make a note regarding the replacement of small components with bent leads: for anyone working on these speakers or a similarly old single-sided Phenolic PCB that’s gone through some heat abuse, be careful when removing the components with bent leads. It is very easy to rip traces off the board. I tried just about every temperature setting on my iron – from 280C to 330C, but it didn’t matter. I found that I had to straighten out each lead before attempting to desolder the component out. Otherwise, it would sometimes pull the PCB trace with it. I didn’t manage to rip anything off, but I did slightly lift a few end traces before realizing how fragile the PCB is. Luckily, the lifted traces weren’t hard to amend - I just cleared off the solder mask on each affected trace and strengthened it with solder.

That concludes all of the recapping and modding I did on this amplifier.

Last edited by momaka; 12-16-2018 at 12:23 AM..
momaka is offline   Reply With Quote
Old 12-16-2018, 12:32 AM   #6
momaka
master hoarder
 
momaka's Avatar
 
Join Date: May 2008
City & State: VA (NoVA)
My Country: U.S.A.
Line Voltage: 120 VAC, 60 Hz
I'm a: Hobbyist Tech
Posts: 10,534
Post JBL S412PII Speakers – Part 5

Part 5 – Cold Start Issue

Even after a complete recap, the amp still had issues turning ON when cold. So clearly, all that was left to do is figure out what caused this cold start problem. Unfortunately, I couldn’t fully figure it out. By the time I finished troubleshooting the distorted bass issue and doing the recap, the weather became warm again, which made the temperature inside my house warmer as well. This got rid of the cold-start issue most of the time. In particular, I noticed the amplifier would start working almost every time if the room temperature was above approximately 72-74°F (22 to 23.5°C). It would also start working after left plugged in for a while and waited (about 10 minutes). As such, recreating the cold start-issue almost disappeared. As I didn’t want the owner to wait for the repair any more than he should, I reinstalled both amps back in service. Of course, I did notify him of the cold start issue I encountered and told him to contact me if it ever surfaced again.

Nevertheless, the following is what I found while looking into the cold start issue:
- All the voltage rails came up normally
- Relay RY101 did not engage to connect the speaker terminals to the output transistors.
- The stand-by LED remained lit with or without a signal and regardless if the “On/Off” mode switch was toggled to “Manual” or “Auto”.

With that as my starting point, I began looking into what was responsible for driving speaker relay, RY101, to turn ON. Clearly it was transistor Q128, but I needed to find what drove this transistor. It turned out it was several things:
1) Thermal protection, implemented by IC102 (an LM311 op-amp) and components associated with transistors Q129/Q130 (Q130 is thermally-coupled to the heatsink as a thermal diode.)
2) Over-current protection (OCP) implemented by transistors Q114/Q115 and associated components. Q126 is simply a driver transistor that pulls down the base current going to Q128 in case of an OCP fault.
3) Speaker DC voltage protection, implemented by transistor Q124 and Q125 sub-circuit.

Thus, in normally operation, Q124, Q125 sub-circuit, Q126, and IC102 output should all remain OFF. This will allow Q128 to turn ON by means of biasing from resistor R150. And with this, relay RY101 should turn ON almost as soon as the amplifier is plugged into the wall.

Once RY101 is ON, this drives signal PCT/PROT low, which goes back to the signal processing board and tells it that it can un-mute the amplifier. The MUTE signal is generated from one of the TL074 outputs of U202 on the signal-processing board. When the MUTE signal goes “low”, it allows driver/pre-driver transistors Q108 and Q109 to bias the amplifier normally. The MUTE signal also makes the Stand-by LEDs on the front of the speaker turn OFF to indicate that the speaker is ON. (On that note, these speakers were actually designed to have dual-color 2-pin red/green LEDs. When the speakers are ON, the LEDs should turn green. This is achieved by the MUTE signal, which can go down to -15V when it swings to “low” and +15V when it is “high”.)

What I was trying to find out after figuring all of the info above is which protection was “tripping” the speaker relay circuit from starting. I tried pulling the Bases of transistors Q124, Q125, and Q126 to ground with a 10-kOhm resistor to mimic no fault. But that didn’t seem to do anything. After disconnecting the output of IC102, the amp relay turned ON immediately upon plugging in the amplifier. So the thermal protection circuit turned out to be the one that was acting up.

I didn’t find any obviously faulty components in the thermal protection circuit. And unfortunately, that was all the troubleshooting I had time for. If I had to guess, it most likely was either IC102 op-amp that was acting up or transistor Q130 becoming leaky.

Hopefully no one will run into this issue or have to troubleshoot it. But if anyone does, feel free to post what you find. Meanwhile, it’s been a bit over a year since I did the repairs and the owner hasn’t called back. So I hope that’s only because there have been no issues. In any case, the reason I decided to make this super-long thread is because as I started troubleshooting the amp, I took a lot of notes. It eventually became so much that I needed to organize what I had noted and tried, and what I didn't. And from that, I figured I might as well make a post here. I think not only can this thread serve as a troubleshooting guide for this particular set of speakers/amp, but also for general amplifier troubleshooting. Hope you all didn’t mind the long post(s) or maybe even enjoyed them.

Anyways, hasta la vista.
momaka is offline   Reply With Quote
Old 12-16-2018, 12:48 AM   #7
kaboom
"Oh, Grouchy!"
 
kaboom's Avatar
 
Join Date: Jan 2011
City & State: PA
My Country: USA
I'm a: Professional Tech
Posts: 2,288
Default Re: JBL S412PII Speakers – Distorted Bass Repair and Updated Schematics

But the big question is, did you "test" the amplifier(s)? You know, into speakers?



I'll read your posts more thoroughly and check the schematic eventually- it's getting late for me...

Perhaps a thermally-marginal small-sig transistor or drifty resistor is upsetting that relay ckt?

The multi-part posting is OK. Like some of the longer youtube repair videos, it better shows details and progress of the repair.
__________________
"pokemon go... to hell!"

EOL it...
Quote:
Originally Posted by shango066
All style and no substance.
Quote:
Originally Posted by smashstuff30
guilty,guilty,guilty,guilty!
guilty of being cheap-made!

Last edited by kaboom; 12-16-2018 at 12:50 AM..
kaboom is offline   Reply With Quote
Old 12-16-2018, 09:37 AM   #8
R_J
Badcaps Veteran
 
Join Date: Jun 2012
City & State: ☻
My Country: Canada
I'm a: Professional Tech
Posts: 7,314
Default Re: JBL S412PII Speakers – Distorted Bass Repair and Updated Schematics

The schematic values for C113 & C114 are 22µf/50v In series the value would be 10µf/100v. It could be that having 47µf BP is causing the slow or cold start. It takes a longer time to charge the cap
If the original caps were 47µf each in series the value should be 22µf total, JBL may have changed the values from 22 to 47 to increase the charge time and the turn on delay.

Last edited by R_J; 12-16-2018 at 09:46 AM..
R_J is offline   Reply With Quote
Old 12-16-2018, 12:51 PM   #9
momaka
master hoarder
 
momaka's Avatar
 
Join Date: May 2008
City & State: VA (NoVA)
My Country: U.S.A.
Line Voltage: 120 VAC, 60 Hz
I'm a: Hobbyist Tech
Posts: 10,534
Default Re: JBL S412PII Speakers – Distorted Bass Repair and Updated Schematics

Quote:
Originally Posted by R_J View Post
The schematic values for C113 & C114 are 22µf/50v In series the value would be 10µf/100v. It could be that having 47µf BP is causing the slow or cold start. It takes a longer time to charge the cap
If the original caps were 47µf each in series the value should be 22µf total, JBL may have changed the values from 22 to 47 to increase the charge time and the turn on delay.
Yes, I thought about that too. So before replacing C113 and C114 with a single bi-polar cap, I used two Nichicon 50V, 47 uF caps in series for the same series capacitance as the originals. Made no difference, though.

Quote:
Originally Posted by kaboom View Post
But the big question is, did you "test" the amplifier(s)? You know, into speakers?
Yessir!

I suppose I forgot to mention it (in all that text there, lol), but I used my repaired 15" Eminence Kappa Pro-15A. Figured it's a good "sacrificial lamb". After all, the two +/-60V HV rails are no joke if all that makes its way to the speaker outputs in case of failure.

Since I don't have an enclosure for it, I just turned it upside down on my carpeted floor, so I could dump more power into it that way (probably did around 20-40W RMS with the JEE test cap). Funny thing is, if you live in a wooden box house (like most of us do here in the US and Canada), playing some low sub-30 Hz frequencies really makes for a great "mini-earthquake".

After recapping the amp completely and testing it on the above speaker again, I decided to try it on my garbage-pickede Sansui SP-7500X speakers (old 70's stuff IIRC with a 16" woofer), as those can handle quite a bit of power. (Which is ironic, because they also have horrible "thumpy" bass - no low-end whatsoever. The bass rolls-off very quickly below 50-60 Hz. I presume it's because they have a very shallow X-max. Surround is also very stiff... and I don't think it's due to age.) With this JBL amp, I was finally able to get those 16" woofers moving a good amount (and the ports to whisle terribly, thanks to whoever at Sansui thought it was a good idea to put a metal plate with sharp edges over the port holes. ) So I guesstimate I was pushing at least 50W RMS into the woofer. Still got no distortion, though. So the opened CapXon cap was the culprit for sure.

Quote:
Originally Posted by kaboom View Post
The multi-part posting is OK. Like some of the longer youtube repair videos, it better shows details and progress of the repair.
Well, it had to be multi-part. Otherwise, finding any useful information in there would have been impossible.

Last edited by momaka; 12-16-2018 at 12:56 PM..
momaka is offline   Reply With Quote
Old 08-01-2021, 03:25 PM   #10
JuelDK
New Member
 
Join Date: Aug 2021
City & State: DK
My Country: DK
I'm a: Knowledge Seeker
Posts: 1
Default Re: JBL S412PII Speakers – Distorted Bass Repair and Updated Schematics

[QUOTE=momaka;868632]I had a pair of these JBL S412PII speakers for repair last year. Finally got to typing up the post for this, though (warning: long posts ahead, split into 5 parts!)

THANKS!
I have these speakers, not the same fault, but some cutouts, your really good writeup, has helped me
And I like your thoughts concerning this amp, mine has the same "slow start" problem, but they always have power, so its no problem.
But its just a poor amp design, these are warm all the time, and constantly eats power
JuelDK is offline   Reply With Quote
Old 09-13-2021, 08:48 AM   #11
momaka
master hoarder
 
momaka's Avatar
 
Join Date: May 2008
City & State: VA (NoVA)
My Country: U.S.A.
Line Voltage: 120 VAC, 60 Hz
I'm a: Hobbyist Tech
Posts: 10,534
Default Re: JBL S412PII Speakers – Distorted Bass Repair and Updated Schematics

Hi JuelDK,
Glad to hear my thread was of some help to you.
Yeah, I can see these amps having all kinds of problems, given the not so great soldering, combined with heat and cheap 85C caps. I don't remember metering how much power the plate amps use when idle and supposedly in the "off" state. But given the chunky line transformers and linear supplies they use, I imagine they consume at least 5-8 Watts per plate amp... so probably at least 10-15 Watts in "stabdy/off" state, which is quite considerable.
momaka is offline   Reply With Quote
Old 09-14-2021, 06:04 AM   #12
Th3_uN1Qu3
Believe in
 
Join Date: Jul 2010
City & State: Bucharest
My Country: Romania
Line Voltage: 230VAC 50Hz
I'm a: Professional Tech
Posts: 5,563
Default Re: JBL S412PII Speakers – Distorted Bass Repair and Updated Schematics

Ah, the brown glue. I developed a few grey hairs some years ago while troubleshooting a plate Amp that would make random popcorn noises. Recapping didn't fix it. Removing the glue from the preamp section finally did.
__________________
Quote:
Originally Posted by PeteS in CA View Post
Remember that by the time consequences of a short-sighted decision are experienced, the idiot who made the bad decision may have already been promoted or moved on to a better job at another company.
A working TV? How boring!
Th3_uN1Qu3 is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump



Badcaps.net Technical Forums © 2003 - 2021
Powered by vBulletin ®
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.
All times are GMT -6. The time now is 07:09 AM.
Did you find this forum helpful?