So this is my first post here about an amp that I intend to fix xP. This EMX640 has been around the block the past decade plus! I mean it was the same PA that I used in many bands all through out 1998-2009. So it has been abused!! lol Lately my old lead singer in one of my older bands had been using it while out at small gigs. They came back from a break turned it on and it just blew a fuse. So, I have put a 6.3amp fuse in it since I have trouble finding a 7A fuse and I have those other fuses around from some tube amps. I mean I have only brought it up on bulb limiter at 40 and 60 watts. The light shined bright indicating a short somewhere... So now time for more info...
First thing I did was opened it up to have a look around and found four inductors on one side of the amp just totally warped. Anyhow I have some pics of these inductors included with this post and they need to be replaced. I am really looking first to test this power transformer to see if it is good still. I mean really how can I test this switching PSU safely without owning an isolation transformer? I am sure there are ways to do this but first would like to ask this community for good advice before I go forward.
Another factor to fixing this PA is the many many transistors attached to the heat sink. I mean these output transistors are totally stuck on this heat sink like glue!! So I busted out the hair dryer to see if they would lift off easily, and no way!! They are on there good and tight!! So it will be a mess just getting these things off to get to the main output board to replace parts(inductors). '
Also, running light bulb limiter, would it be wise to disconnect power from the from preamp to see if the short goes away? Also, I will check the status of the BR and other output transistors(especially those that are in line w/ the ugly inductors). Any thoughts on how to proceed on repair of this old PA would be appreciated! Thanks. Ask questions since I probably have left out missing details. You can see the transistors in the first pic attached and detaching these is looking to be problematic. Also, the main filter caps probably need replacement or at least new soldering. The glue that held them to the board is obviously cracked up bad!
Edit: another thing to note is that one pic shows what those inductors should look like... Also, note it the two 2Amp fuses that lead towards the preamp power supplies did not blow... Also, I was told that at least 3 input channels were not working very well or at all and that probably was related to the side of the amp w/ the deformed inductors.
http://www.fileden.com/files/2012/8/8/3334901/DSCF1026.jpg
http://www.fileden.com/files/2012/8/8/3334901/DSCF1028.jpg
http://www.fileden.com/files/2012/8/8/3334901/DSCF1029.jpg
http://www.fileden.com/files/2012/8/8/3334901/DSCF1030.jpg
http://www.fileden.com/files/2012/8/8/3334901/DSCF1031.jpg
http://elektrotanya.com/yamaha_emx-640.pdf/download.html
Hi Gonz, there are a few familiar faces here.
Isolate the problem. First one I see is that my schematic shows a linear power supply, not a switchmode supply. Your board photos match my drawings, so... Also, my schematic calls for a 10A mains fuse and 2A low voltage fuses. Does the amp have a printed fuse spec on the rear panel. I don;t see 7A.
The power transformer is the last thiing on the list of failures, not the first. Looks like a 6-pin connector connecting the transformer to the board. UNplug it from the circuit and apply the mains power to the unit. The transformer either powers up OK or it doesn;t. Pretty much that is the test.
In any unit that blows fuses, the first thing to look for is shorted output devices. I see six up each side, look for shorted ones. Most likely on the melted inductor side.
Your inductors are probably OK. All I see melted on them is the plastic shrink wrapped around them. For them to be damaged, they would have had to get hot enough for the enamel on their wire to flake off. Not usually the case.
Why would they melt? Too much current through them. And looking at the schematic, where could that current haqve come from? Looks to me like those output transistors. Through one and out another is the only high current routes I spot. SO again, check those outputs for shorts.
A lot of transistors stick to the heatsink with those sheets of insulating silicone, especially if they get real hot. I usually stick a small screwdriver under the legs, like between the backside of the legs and the board, and twist th driver to lever them up a bit. That usually pops them free. Not much different from an amp chassis stuck to the tolex in a cabinet. If the insulator tears, it has to be replaced.
I suppose you can disconnect the preamp, but it won;t be the fuse blower. It runs on low volts, and if it were screwed up, those 2A fuses would go out.
Forget the input channels until the power amps work. And whatever it takes to get to them.
You can check the main bridge and all those large transistors in about 30 seconds, just go down the row with a meter on diode test looking for shorts. Remember to unplug the transformer from the board first, or your bridge will look like it is shorted AC to AC on its terminals - the low resistance of the winding.
The caps are probably OK, but do resolder them tight to the board. If any are tired and dried out, they will reveal themselves later as a hummy amp. Not likely they are blowing fuses, especially with the melted inductors as a major clue.
:tu: ...and also;
Those overheated inductors are in series with the outputs, so;
- Check the associated loudspeakers
- Check all the speaker leads for shorts (we hope), connectors internally in particular
{That didn't happen in storage; it took a few seconds under drive, possibly caused a little smoke, and almost certainly a stink of melting plastic - then the fuse blew. Jes' sayin'}
@Enzo Yes I think my brain was farting when I was writing this yesterday... :duh I really meant that I should disconnect the power supply altogether to test it out of circuit. The end result was positive but I always like to start there to not have a surprise later on.
I was able to jostle those transistors free by way of a screwdriver. I barely tore the insulator but any ideas where to get this material insulator as to replace? I always find that my search for insulators in general is hit or miss at times. A lot of times the different sites are just not stocking the exact ones I want or need.
It says (JU) 7a 125v fuse on the schematic and that is what I found in the amp. So, not too worried about that right at the moment.
@Roly & @Enzo Those inductors really heated up causing the solder on the legs to have loose connections. There are some solder joints on the board that will need to be touched up and I am looking at those transistors. The BR seems fine but a couple transistors are suspect. However, tomorrow I will start removing them from circuit for better testing. I will also look into checking out the speakers but they are not here right at the moment.
Will get back here with some more findings on those transistors tomorrow. I really appreciate all the knowledge that has been passed along on these forums this past year!! Thank you all :dbtu:
Happy New Year Everyone :tu:
So I have found that Q217N, Q218P, & Q223N are shorted and bad. I have done some searches around the net to find replacements. I have found one source that sells them in pairs...
http://www.kpcomponents.ca/product_info.php?manufacturers_id=17&products_id=885&osCsid=ytkymhlvumj
So this might be the way to go, but just looking up I noticed it's in Canada and shipping might prove to be expensive. Anyone have a good idea where to buy B1647 and D2560 pairs? thanks. Also, should I be matching up the pairs on that side of the board to all have new output transistors? Or should I just replace the three blown transistors only...? What do you guys recommend?
Also, I am thinking about replacing those inductors for good measure if my friend wants them changed out. I found perfectly matching parts on Digikey. However, how do I test these parts to see if any of them went bad? As Enzo said it would be a surprise if they were not working, as only the insulation was now melted.
I have bought my Sankens from B&D Enterprises for many years and have not yet gotten any fakes. They list both those types in stock at $4 and change apiece. And they are in the USA.
www.bdent.com
MCM also shows them in stock at about $2 each. I don't usually buy Sanken from them, so I cannot say they have a fake problem or not.
Not fingering any particular company with fakes, but counterfeit transistors is a large problem in the industry these days.
And "pairs?" The transistors in products like this are not generally matched.
Thanks a ton Enzo!! Definitely will try out B&D for this order.
These inductors are stability inductors in series with the main output (one for each pair by the looks). They aren't particularly critical to inductance value, but they do need to be heavy enough to carry their share of the output current, which may be several amps. I think Enzo was talking about the insulating varnish on the wire surviving, rather than the plastic case, which is more important.
You could even wind your own replacements if needs be.
Yes, I meant to say that I think the insulating coating on the actual wire of the inductor was probably fine, The melted stuff was just a thin plastic sleeve that has no impact on the operation of the part.
Does this look like a good replacement for these coils since my friend is wanting to change them just in case.
Replacement part:
http://www.digikey.com/product-detail/en/TSL1315RA-680K3R0-PF/445-3727-1-ND/1913591
Original Part:
http://www.digikey.com/product-detail/en/LHL13TB680K/LHL13TB680K-ND/1171383
Edit: Also, I am thinking the 7A fuse should be fast acting type, but cannot figure this out exactly. TL is on the fuse so it must Time Lag type?? ooops ignore that last one... lol. It was a basic UL on it not TL... :duh
As I remember, according to Doug Self these stability inductors are not at all critical in inductance value, in fact I'd say the higher current rating of the intended replacement would be of more importance.
I'm sure you've seen, as I have, that these are a highly variable species, and I've even used 0.2 ohm wirewound resistors to satisfactory effect.
Thanks Roly just have self doubts sometimes and too many 2nd thoughts. It seems I am on track as my transistors have arrived today and tomorrow the other parts should be here to finish the job. Always appreciate the feedback, thanks. :tu:
Okay I have now installed all the parts... I have a question before I proceed... I have found that Q224
Q223 Q219 Q220, Q218 Q217 when tightened to heat sink do not create continuity as there is thermal installation pad that prevents collectors from touching the heat sink. Same thing can be said about applying Q124 Q123 Q119 Q120, Q118 Q117 on the other side of the board, no beep on meter suggesting the heat sink is grounding out the collectors. I tested this carefully but then once I connect both sides the collectors on Q224 Q223 Q219 Q220 and collectors on Q124 Q123 Q119 Q120 all make ground contact to heat sink. Also, tests ground to bridge rectifier +/- legs in the circuit to the heat sink. The other transistors Q218 Q217 and Q118 Q117 (linear outs on each rail) do not make continuity ground to the heat sink.
So in short, I have ruled out that the thermal insulation is the cause. As testing for grounds to the heat sink on each side do not occur. The heat sink only becomes grounded to those mentioned transistor's collectors when I tighten both sides. Independently tightened one side or the other shows the thermal pad is insulating the collectors. So I just want to be sure this is normal before proceeding.
So is this normal? I am afraid that I am gonna fry these transistors or something else could be bad. I am then thinking in circuit this is normal. What do you guys think? THanks ???
Edit: One last thing to note is that there is resistance from heat sink to the collector ranging and changing between 7-17 ohms. Which really makes me think it is just in circuit, especially after looking over the schematic for a bit.
So anyhow using light bulb limiter I went with a 40 watt bulb and tried to fire up the amp. The result let me know there was still some short on the board. Carefully started looking at other transistors that might be shot and did not find any (I think). Good news is that all the output transistors are still fine and have not been damaged as I used the limiter. I even pulled off Q224 and Q223 from the board, as they need to be removed for testing, and was relieved that they were fine.
I found an ultra-fast-recovery diode that was open when measured on the board. Pulled it off the board and it definitely is part of my problem now as it is shorted out. I have attached the datasheet for this part below in the hopes that someone can recommend a good modern replacement part. I have some rectifier diodes around that might be good replacements, but will have to research the specs of this part more. Can anyone help suggest a good replacement for a diode rl2z???<<< as it reads on the actual diode (on the schematic it is part D222 SF22) Thank you for any input greatly appreciated!! :tu:
Edit: I am really having trouble looking at the data sheets and matching up this part. Reverse voltage, forward voltage drop and recovery time has me confused as I go to Mouser or somewhere to find the part. I am thinking that this part will get me through the night...
http://www.mouser.com/ProductDetail/Fairchild-Semiconductor/EGP20D/?qs=sGAEpiMZZMtvcUztdGSumNDUeNUKcr9kM2qH%2fwpsiBQ%3d
Can someone confirm this as I feel a bit confused, thanks
Reverse voltage
This is the maximum voltage a diode can withstand in the reverse direction. It's what we mean when we say that a diode is a 400 volt, 600 volt, or 1000 volt device. A replacement has to be equal or higher.
Forward voltage drop
This is the voltage across the diode when carrying a specified current in forward conduction, typically 0.7 to 1 volt for a standard silicon power diode, and around 0.1 volt for a Schottky diode. A replacement has to be equal or lower.
Recovery time
This is the time it takes a diode to enter the non-conducting state when the voltage is reversed after passing a specified (heavy) current. When the voltage on a diode is rapidly reversed there is a short period when there are residual charge carriers in the chip/dice and this results in a high reverse current until they are swept out. An "an ultra-fast-recovery diode" has small charge storage and a short recovery time to non-conduction. A replacement has to be equal or shorter.
HTH
Cool Roly thanks man!!
Especially for letting me know the recovery time concept as equal or lower for replacements. Now I see the light! :tu:
You must sort out the transistors insulation from the heatsink. The transistors meant to be insulated should have some kind of mica spacer or silpad, along with some kind of plastic bushing for the screws.
If all the transistors are insulated, then you should not have any collectors shorted to the heatsink when they are all tightened (as the heatsink itself is floating, correct?). If they all have insulators, one of them must be cracked or damaged. Find out which one.
Yeah I will keep checking to see why *if* the insulation is failing, but it did not seem to be the connection. I will even pull out some mica insulators and just go that route in the end too. The power board has the four screw mounts connecting to the heat sink that have trace connections to the board. Still I am gonna have to figure it out before I feel comfortable plugging it in to the main voltage.
Well, if the heatsink screws connect to traces, then you are probably ok. Remove those particular screws and if your transistors are all insulated from the heatsink you should be good.
Sounds like you have already done this but doesn't hurt to double check.
Edit: Looking again, it appears the heatsink is grounded (it screws to chassis), so you shouldn't have any collectors or parts of the bridge rectifier shorted to the heatsink.
You appear to have shorts from both + and - rails to ground. Find the source of your shorts.
:tu:
If you can mount the transistors and check they are fully isolated before you connect their legs to anything, do so. Keep an eye out for damaged insulating washers on the screws; sometimes the spacer sleeve gets damaged at lets the side of the screw shaft touch the device tab.
Generally, with a mica insulator in particular you need to be sure that it and both surfaces are clean and free from any grit before you apply thermal compound. They are very thin and brittle and grit can punch through. Those rubbery insulators are normally much less picky.
Yeah I ordered TO-3P silicon type insulators last night, so yeah mica would have been nightmare in this case. New insulators are probably needed here in this instance. This type of transistor does not require insulating washers on the screw so that is not a worry. Definitely will figure this out once new insulators arrive and will not proceed powering up unit till I figure this out. Thanks for the input guys!! :tu:
Yup got it working now! It was the bad insulation and it was just strange that each side(screwed down) would be fine until I started screwing in the the other side. So the insulation was failing when both sides were tightened down and that just had me confused. Thanks to all that contributed to this success...
Enzo, Roly & G1 you guys Rock! :tu: