Need troubleshooting help...with reward!!

[akozols]

HI,

  I'm looking for some help with a Boss processor I borrowed and broke.  I'ts a Boss ME-33.  It takes a 9v ps, but I used a 12v supply I've never had a problem with(till now).  I was playing awhile till I notices the output level dropped and the sound got really buzzy, and smelled that electronic's burning smell.  I opened up the unit and felt a hot component.  I tested my "12" volt supply and it read 16 volts(ouch).  

  I have an electronics background but am a little sketchy with what's inside this unit.  I don't have a schematic so I'm looking for some guesses as to what could be wrong.  I'm going to spend some time poking around to track down the input voltage.  Any suggestions would greatly be appreciated.

  I do have a Dunlop Crybaby shell that I would gladly give to someone that helped me out.  It's not much, but I'm sure you could use it for a project.

thanks,

Al

[freebird1127]

It's the 12V PS that's the problem?  You measured 16V across the output of the supply?  Was it loaded or unloaded?  I know that I've built many +12V supplies and the output has read from 12 to 19V unloaded, but as soon as it's loaded with anything it measures the expected 12V...  I would maybe consider checking whatever's being used as a voltage regulator, the three terminal regulator 7812 (TO-220 package) is popular for 12V supplies.  Easy to swap out and check, and can be ordered for free as samples from several chip manufacturers.  As far as pinpointing the regulator goes, it could very well be attached to a heatsink.

[javacody]

I have no clue, but a stab in the dark would be to look at the electrolytic and tantalum caps. See if any are rated very low (i.e. less than 12 volts).  If they are, maybe replace them?

Sorry I couldn't be of more help.

[Samuel]

replacing electros would be a good place to start - I know alot of BOSS stuff does use 10V caps in 9V supply circuits...is it also possible that the PS was not the same polarity as the device wanted? check, you might have either just blown a polarity reversal protection, or alternately your problems could be much worse.

However, given that this is a digital all-in-one multieffect - it seems likely that the whole thing is going to be surface mount components, so you're probably SOL in general. Sorry!

[akozols]

Thanks for the replies.  After the bad smell, I did find a good 9v supply, but still have the problem.   The original ps was the right polarity and I've used it many times before.  

 Any idea where to get schematics for Boss processors.  Maybe the share simular designs.

thanks,

Al

[NaBo]

oh boy...  sorry to tell ya, but i doubt you'll find a schem for a digital multi-fx processor anywhere online.  did it come with a manual?  i know my digitech processor actually had a schem included in the manual for "certified repair technicians only" or something... lol... riiiight

In general do-it-yourself and repair-it-yourself seems to be greatly limited to analog devices.  Seems to me, that when you get into digital, especially something like a me33, it becomes more like trying to fix your PC than a tube screamer... Check to see that no basic parts have failed, ie: caps, do what the other people suggest... Is there maybe even a fuse in there?!?!?!  lol

I dunno, man, but if none of that works, I'm gonna have to suggest taking it down to a shop or dealing with boss on this one...

:cry: ... (wanted to help and snag that wah shell)  

[stm]

Hi.  If you fell burning smell, the best way to go is open the device and track down the burnt component by:

1) Using your eyes:

The damaged component may have turned a bit darker/brighter, appear more/less shiny than its neighbours, or the solder mask (the green "paint" on the PCB) might appear slightly different around the damaged component.

2) Using your nose (really!):

Use your nose to track down the burnt component.  Open the effect and smell all around the board with your nose just touching the components.  You may be able to find an area where the burnt smell is stronger (it remains for months!).  Also, don't forget that the burnt component may be on the other side of the PCB, or on those stupid add on boards connected with several cables to the main board.

3) Use your brain, Watson:

Most likely the problem is in the +5 voltage regulator. After all, digital logic uses 5 volts, or maybe even 3.3V, so there must be a voltage regulator there.  If you increased the supply voltage from 9 to 16 (or even just 13 volts), you increased power dissipation to more than twice on the 5 volt regulator IC.  Try to identify something like a 3-legged device, or something with the numbers "7805" or whatever makes sense with a voltage regulator IC.  If possible, check the +9V input and +5 output pins with a DVM to see what's going on there.

Also, there might be a zener or transient suppressor at the input that could have been limiting the input voltage to some controlled voltage--a 12V zener makes sense, so if voltage was higher, it might have been damaged after some time of operation.  Usually zeners or diodes turn into a "permanent conduction" state when they overheat, so it may be acting as a crowbar now.  Connect the +9 power adapter and measure with a DVM if it is actually providing +9V WHEN CONNECTED TO THE BOARD AND POWER IS ON.

If there is a problem under this situation, eventually the funny smell may reappear again, making easier to identify the damaged component.

4) Use your tact; put your index finger in...:

Use your fingers to touch every component on the PCB with power applied to search for hot spots.  Sometimes the upper part of the finger can be more sensitive to heat.

It is very likely there should be no more problems than a damaged voltage regulator or diode.  Believe me, I connected a USD 750 digital camera to +28V instead of +12V !!!  The only damaged part was a Maxim Mosfet driver device used a switching power supply.  After replacing the device, the camera went back to life!

Good luck!

STM

P.D. I could make a good use of a wah shell  :lol:

[toneman]

U can get the service manual 4 the unit @

RolandUS

here's their support page.
http://www.rolandus.com/support/updatesdrivers/index.asp

I know there's a phone # somewhere(?) on that page...
i've just phoned them up and ordered service manuals.
U can 2!!
But, if U do *not* understand electronics, I suggest U send it/take it
to an authorized service center.   The support site has addresses &
ph#s of one's in your area sorted by zip.
Service manuals usually run about 20bucks and contail *ALL* the
technical info U need to get it flying again.
Helped me fix my GR30.
I could fix for minimal, if U purchase manual.
Give me a PM if U want.
staytoned
tone

[freebird1127]

the thought just occured to me, check to see if there is a crowbar diode across the power input... although this is normally coupled to a fused system.

[stm]

the thought just occured to me, check to see if there is a crowbar diode across the power input... although this is normally coupled to a fused system.

That's exactly what I mentioned on the second paragraph of point 3) in my previous post.  Usually they may be too cheap to use a real fuse and place a small valued resistor instead.  When overcurrent passes the resistor just burns due to overdissipation, effectively opening the circuit.  If this is the case, you just need to replace a resistor.

[akozols]

HI Guys,

 A quick update for you.  I've been busy and haven't really had time to look into it.  I got the go-ahead from the store I borrowed it from to do what I can to fix it.

 Symptom is now that as soon as I power it on, it will play fine for about 10 seconds before the sound starts to fizzle.  The symptom is the same using wallwart or 6 AA batteries.  With new batteries installed, it's showing that my batteries are low. Looks like it's a power issue dropping battery voltage.

From the looks of it, I think the battery input is after the voltage regulation.  If it is, I'll disconnect and run from batteries to see what happens.

 So far I've tracked down that power comes in and goes into FL1 that goes across to ground.  It then goes into a S07 030J tri leg component.  

I'm offto work, will update later.

Al

[ErikMiller]

What is "FL1?" Fusible link, perhaps? :-)

[akozols]

Ok, tonight I had some time to check out the unit.

   

  9V comes in and drops down to 7.2 when connected.  It then goes thru FL1.  FL1 looks like two cylinders connected to a cap, dipped in some kind of goo.  9v goes thru the two cylinders and are connected by the cap to ground.  

7.2v then goes thru D1.  It's a 3 leg component S07 030J.  It only uses two legs and looks like it's just a forward bias diode.  I get 6.8 coming out the other side.

 It then comes in contact with D2, 1414, which looks like a diode that jumps over to ground.  

 6.8v then goes to the emitter and collector of Q2, C2235.  5.4v comes out at the base.  That's the part that got super hot with the original supply.  It doesn't smell now, but that area of the pcb has a slight odor.  There are a few 16V electrolytic caps and a load of surface mount resistors and caps in that area.  

 With FL1, D1, and D2 removed and running battery power only, the sound still fizzes out.  

 I'm not sure what D1 or D2 is supposed to do.  Do they drop voltage or are they there for protection.  I'm goinlg to try swapping out the 16v caps to see what that does.  

  thanks,
al

[stm]

From what you mention it seems Q2 is performing as a series voltage regulator.

It is possible the tranny got degraded by the severe heat suffered. Make sure voltages on the three pins of Q2 make sense. Try to identify the circuit and surrounding components like resistors and diodes.

On another line of thought, try to identify the +5V supply trace on the PCB and go cheching along it on each visible connection looking for small voltage variations.  This will indicate which component is stealing the power.  Of course the above is only possible if the PCB is single or double sided; multilayer PCBs won't allow this.

So far it seems you are doing good with your troubleshooting!

Good luck,

STM

[akozols]

HI,

  Looks like stm will be getting the wah case.  He could use one and he's been a great help.  I'll pm you.  

  Haven't had much time to play with it, but I'm going to look into getting another c2235 for it.  While playing around with it, I found anolther hot component.  It's not hot enough to burn my finger, but I can't keep in on there too long.  I't a small HC4052 analog multiplexer. Not sure if it's supposed to run hot.  If the C2235 doesn't fix it, I think I'm going to give up.  thanks for all the replies, and 'll post updates.

Alo

[stm]

Good idea to change the transistor.

Also, you should take care of the HC4052.  It is not suppossed tu run hot. If you can't stand your finger on it forever, there is certainly a problem there or downstream.

There is a small chance the problem is not in the IC however !?   This is an analog multiplexer (like an electronic switch), but with an ON or CLOSED CONTACT resistance around 300 ohms when powered by +5V.  In the event one of the outputs is shorted, current flowing through the electronic switch may cause heating due to I^2 * Ron, being I the current through the IC.

If this multieffect was mine, I would replace the 4052 chip anyways due to the stress it has suffered.  Remember it is in the audio path, so you don't want any degradations there!  But don't forget to check what's happening downstream also!

It is a matter of time and persistency until you find and fix the problem.  Also, when you do it, you will feel great having been able to do it yourself (adrenaline rush), and would have gain a lot of experience in debugging.

Good luck  

[akozols]

Hi,

 I promised an update.  I ordered two transistors and two HC4052.  Installed the transistor, I still have the same sound and the 4052 is still hot.  On to the 4052.

The 4052 was a small surface mount component.  I tried to remove by heating up one side of the small legs and lift up on the 4052.  I knew I should have waited to get some braid, but I rushed.  .  I ended up pulling the traces off the board.  I had the data sheet for the 4052 and saw that it was just an audio switch.  I think it toggles the input between the guitar and line input.  There's a function where you can sample an audio input and loop it.  I made a little audio probe that plugged into my amp and poked around the traces of the 4052 and found my signal.  I then used that signal to poke around till I heard an output.  I hardwired it so it will always see only the guitar signal. My initial buzzy staticy problem is gone, but now I have a new problem.


   My new problem is wierd.   If I play softly, the sound will always be there.  If I hit the strings hard, the sound slowly fades in.  After the intial fade, the sound will be there no matter how hard I play.  It's like it can't handle a hot signal right away.

 Here's where it might get tricky to follow.  I get the symptom whether I am processing the guitar or hit the bypass button....BUT...if I hit the sample button, I get a clean guitar sound.  I'm getting the guitar sound since I hardwired the guitar input instead of the audio input.  This tells me that the section that processes the guitar signal is different than what processes the audio signal.  

 So far I've traced the signal using my "amp Probe" and the signal is clean going into a 4552vt. I searched and thats a D\A, A\D converter.  I'm not sure if it's used to to a\d AND d\a.  If I poke around the pins, I get some wierd sounds, but nothing that sounds like my normal/problem output sound.  

 Tonight at work I'm going to start at the ouput and try to trace my way back and mark them.  I'll take note of component numbers and find out what they are to help out.  When I get home, I can check those points to see what they sound like.

  I still have the wah case.  STM gave the most help and I was going to give it to him.  I found that he lives in Chile and it would cost me over $20 to ship.  We agreed that was too much.

thanks,

Al