Tascam 388 problems Part 2

[The Great Cobb]

If you recall a few weeks ago there was a post about a Tascam 388 that had burnt out the fuse in the F3 position, and that the heat sink on the back was getting morbidly hot within a short amount of time being on. That was my room mate's and my 388. He has since moved out and now it is down to me to fix it up.

I am about to just take it down to the repair shop, but the guy wants 80 an hour and wont give me an estimate or take a look at it to see what kind of problem it might be. So i have no idea wither its going to be 3 hours of shop time or 20. Basically i am posting this to see if anyone has any idea of what might be wrong with it based on the description of the problem, so i have a remote idea of the repair costs. I am going to have to go into debt to pay for this thing to be fixed, which is terrible.

This is all i know if its problems:

WORKING:

The machine plays, records, all mixer channels work, all transport functions are fast and tight. The unit has a new belt. The tape is new and there is no gunk on the heads.

PROBLEMS:

Fuse in the F3 position was blown and replaced. After about 10 minutes of being on the heat sink gets very hot, hotter than it ever should, starting with the left hand side of the heat sink, eventually spreading to the whole sink.

Something i noticed today, my room mate replaced all the fuses a few weeks ago, but now, after having had been powered up for a total of maybe 30 minutes since, the fuse in the F3 position is starting to yellow. So whatever is happening that happened before is happening again. Ive left it off and unplugged since i made this discovery.

Does anyone have an idea of what might be causing this?

[sweetbeats]

Transistor Q1 is part of a regulator circuit I believe. It is on the left hand of the heat sink. Remove those plugin cards so you can see where Q1 mounts to the little PCB at the back of the heatsink. Look to see if the PCB is discolored/brown where Q1 connects to the little PCB (the TR PCB) with its three terminals.

I'd start by scoping the power supply rails (checking each rail with an oscilloscope...it tells you how clean the power is and will identify a power rail with bad regulation).

My guess is that the regulation is bad so the gate is too open...too much current flow. I wouldn't operate it until you can get a handle on it.

D105, R105, Q100 and Q101 on the PSU PCB would also be good to check.

[The Great Cobb]

I looked inside at the Q1 and the other ones, nothing. No discoloration whatsoever on the PCB or anything else.

I dont have an oscilloscope. Can you recommend one that will do the job?

Thanks for your help.

[sweetbeats]

I'm a scope newbie myself. Have you used one before? Its a whole new world.

Pull the power supply PCB. That is the plugin card closest to the right of the cardbay. Can you take some hi-res pics of both sides of the card and post them here?

I really think you need to find a reputable tech to thoroughly check out the power supply. I don't know what your background is so I'm not trying to discourage you but the whole oscilloscope thing I know was a huge step for me because it takes mentoring to understand:

1. How to use the thing
2. How to read what shows up on the screen
3. How to take what you read on the screen and interpret what it means in reference to the thing you are measuring

You might be able to find a scope for as little as $50 but expect there will be some more expense if something doesn't work or it is missing accessories or the manual. Figure maybe $100 ~ $200 for a scope that is ready to go for the duration.

I'd offer to have you send me the Power Supply PCB and i'd check it out for you but I really can't take on something like that right now.

If the tech your talking to won't give you a flat rate to diagnose and then give you an estimate for the repairs I'd be looking for another tech. Something is allowing too much current to pass -OR- something is drawing too much current and the checking starts at the power source.

[dodgeaspen]

80 an hour?!?! MAN!

[The Great Cobb]

Well i wrote to the dude and told him what was wrong with it and i told him ive got a budget of 300 bucks, which i am going to have to go into debt to get. I asked him if he can fix it for 300. I will not go higher than that. If he cant do it for 300 in cash then i dont know what to say.

Basically with his system, you bring your stuff down, then he fixes it, and whatever time it took him he then bills you. At that point it could have been a 2 hour job or a two week job. And you are at his mercy because he has your stuff. The guy hasnt been anything but kind so far, and i have no reason to doubt his trustworthiness, but i think its a bit of a hassle to not be able to get a repair quote for an 80 dollar an hour shop.

Ive got a room full of old half broken crap that id love to get fixed. If this guy had slightly more accessible customer service as far as quotes go, even estimated quotes, it would be money in his pocket. But instead the "no estimates" policy just kinda has me turned off from the whole idea. Ive worked in a shop before that billed people 75 an hour, and it was a number we just pulled out of nowhere and even if something took us 5 hours wed usually just bill people for 2. Thats how you get people back in with more orders.

Oh well lets see what he says.

[sweetbeats]

Cobb...the situation makes me itch.

Does anybody know of a reputable repair shop in Arizona? Not saying the folks you are talking to aren't good or reputable but their policy sounds like one they implemented to control business load...

Another thought, Cobb...might sound nuts, but you are about a 6 hour drive from Montebello California, the ultimate Tascam repair facility: Teac/Tascam headquarters. You might call them up and see if they do estimates, or even talk to their analog support department and get an idea of what the problem might be...they might give you a ballpark of the cost.

[The Great Cobb]

Since i dont own a car it would be next to impossible for me to take it to Tascam themselves.

Yes, let it be known that even car-less, my priority is still analog reel to reel decks and spring reverbs... haha

Sweetbeats, i dont really love this situation either, but its one that i seem to be on a neverending cycle to have to confront. If you want, i can PM you the website of the guy who runs the shop, but its pretty normal. There are only two shops i know of here, and the other is really high end stuff, like custom consoles and full rebuilds of hi-fi tube stuff. So i think theyll be even more demanding price wise. I think i recall them charging like 250 PER CHANNEL to recap mixers or something like that. So this guy seems like almost a deal.

The sad part for me is... i cant even tell if this thing is broke! Its just getting really hot on the heat sink, which seems a bit abnormal based on how hot its getting. Everything else functions on this machine.

[The Great Cobb]

On a much lighter 388 note, here is an album recorded using a 388. Sounds fantastic to me. Especially track 8. If you listen closely the echo and drum tone is almost exactly like the drum and echo tone on the album "King Tubby Meets Rockers Up Town" on the last 2 tracks, which is a massive compliment.

http://music.myspace.com/index.cfm?f...bumid=11700110

[sweetbeats]

I'll have to check out that link...I'm a huge fan of classic dub...King Tubby, Mad Scientist, etc.

Urg...I wish I had an answer for you on what to do with the problem but it is over my head...

[sweetbeats]

When you say "left side of the heatsink", am I understanding correctly that you are referring to the side closest to the mixer section?

[The Great Cobb]

Yes, its the mixer side that heats up first. Once the machine is on,, its ok for a few minutes, then the left side starts to get warm while the right stays cool, then after a little longer the whole thing gets really hot. At aroun 10-15 minutes of being on, the heat sink is untouchable. Keep in mind there is no smoke, everything functions, there is no burning smell or anything. In fact, the vent area on top of the machine stays relatively cool even when the sink is raging.

Like i said, i pulled those cards and saw ZERO signs of heat damage anywhere. BTW i did see that same slight discoloration on both sides of those diodes that you had in yours (as shown in your 388 story thread).

Any more basic tests i should try? Its fairly clean inside. The transport never misses a beat and springs to life after any command is given.

[sweetbeats]

Can you put up a pic of the fuse board?

[The Great Cobb]

AH... i might have to get a friend over who has a camera. I used to have one, but i dont anymore. I replaced all the fuses with new ones except the very last bottom one, mainly because it was a different kind, and because it didnt appear to be messed up in any way. Does the bottom fuse potentially have anything to do with what is heating up the sink so fast?

If you recall, the machine originally had full transport functionality but wouldnt pass any signal in the mixer section. After replacing a fried fuse in the F3 position, that problem was solved and now the mixer section works fine. But, like i said the heat problem concerns me and i can see the fuse in the F3 position is starting to get a little yellowish, and the machines only been on for short amounts of time since it was installed.

[evm1024]

There are 2 general types of contracts. Level of effort and firm fixed price.

If you take your deck into a shop you will get level of effort at some shop rate (and $80 / hr is not bad, I've been paid up to $250 and hour) for as many hours as needed to fix it. This is also called time and materials. You can set an upper bound on the cost. Level of effort up to $300 for example.

In the case of firm fixed price there is significant risk to the tech in terms of how long it will take to fix it. In general when I've done firm fixed price is 2 to 4 times the estimated (at lease internal estimate) cost to do the job. This sets a fixed price to the customer and has enough extra built in to cover the risk of overruns.

Jumping over to the problem you are having.....

The heatsink should not get so hot that you cannot touch it. The system is going into thermal runaway and the fuse blows to protect it from catastrophic failure. Verify the fuse rating and do not change to a higher value.

Something is not right... (we duh)

I have a very blurry pdf of the 388 manual so it is hard to read the part numbers. (If anyone has a good schematic on pdf let me know)

Counting from the top of the schematic, which fuse is blowing? and which device on the heatsink is getting too hot (Q1 etc)

Fromt here I'll see if I can suggest something.

Regards, Ethan

[The Great Cobb]

Yes, i too have a blurry copy of the 388 manual in PDF format. The hard part is, in the manual it lists the values of the fuses as F1 = 3A 250V, F2 & F3 = 2A 250V, F4 = 1A 250V, and F5 = 3A 250V. But, on my actual PCB it lists the values as F1 = 4A 250V, and all the rest as 3A 250V. When i wasnt getting signal from the mixer section, i replaced the burnt out F3 position fuse with the same value as shown on the PCB. After that the mixer worked fine again. Then, since i had a package of them i also replaced the ones in F2 through F5 because on the PCB it says they are all 3A 250V.

The left hand side of the heat sink starts getting hot way faster than the right does. It is an obvious difference even after 1-2 minutes of being on. It is the side that is marked Q1 the furthest to the left.

Thank you for your help.

EDIT: I was thinking the difference in the manual might be because they changed the design a little at some point. I remember hearing that they added vents to the top later, and mine does have the vents, so maybe the manual thats floating around the internet in PDF is from an older version of the 388.

[evm1024]

OK , I had a little time to look at the schematic. Fuse F2 and F3 protect the transformer winding for the (presumed 15) plus and minus 15 volt supply. This winding goes through those fuses and into a bridge rectifier. This provides unregulated + and - DC to the dual tracking voltage regulator chip (U100 I think) and to their series pass transistors Q1 for +15 and Q2 for -15.

I assume that you do not hear a loud hum so this tells me that the bridge is most likely not blown. If it were I would expect lots of hum....

Test 1: Measure the voltage on the collectors of Q1 and Q2 relative to ground. You should be able to see those voltages on PS connector P4 pins 2 and 5. I'm guessing that they should be around 20 volts DC or so. I'm looking for differences in the voltages, one should be + and the other - to the chassis.

That fuse F3 is blowing first may only mean that it went before F2....

Test 2: Measure the dc voltage on connector P5 pins 3 and 2 (to ground again). Should be + 15 and -15

Test 3: remove PS connector P4 and see if the heatsink gets hot. P5 sends the + and - 15 volts to the system. By pulling the plug you isolate the PS from the system. If it gets hot then the problem in on the PS board. If it is cool then the problem is in the system somewhere....

Test 4: remeasure the dc voltage on connector P5 pins 3 and 2 (to ground again). On the PS board. We are looking for a difference to the results of test 2.

Caution: be careful not to short anything (like yourself!)

Regards, Ethan

[The Great Cobb]

I know this is going be a real eyeroller, but this is my first time using a multimeter on anything. When you say "to ground" what do you mean? Should the power be ON during these measurements?

I know this probably sounds ridiculous but im just trying to make sure im not about to screw anything up or kill myself.

Thank you.

[evm1024]

Originally Posted by The Great Cobb

I know this is going be a real eyeroller, but this is my first time using a multimeter on anything. When you say "to ground" what do you mean? Should the power be ON during these measurements?

I know this probably sounds ridiculous but im just trying to make sure im not about to screw anything up or kill myself.

Thank you.

I'm actually quite happy that you are willing to ask rather than get hurt.

There is risk of hurting yourself or your equipment if you touch something or short something out. These tests are done with the equipment turned on.

In most cases a measurement is done with the black meter probe on the sheet metal (which is grounded most often) and the red meter probe on the test point in question. This gives us the voltage from the test point to ground.

We are measuring DC voltages which are less than 200 volts. So you would start with your meter set to the 200 v scale and then reduce the scale to the 20 volt scale when you see that the voltage is less than 20. (if you try to read 21 volts when your meter is on the 20 volt scale it will indicate overscale by blinking at you)


Be careful and work slowly.

Regards, ehtan

[The Great Cobb]

"Test 1: Measure the voltage on the collectors of Q1 and Q2 relative to ground. You should be able to see those voltages on PS connector P4 pins 2 and 5. I'm guessing that they should be around 20 volts DC or so. I'm looking for differences in the voltages, one should be + and the other - to the chassis."

So i have a couple questions, and before i ask the first one let me say thanks again.

1. Do i pull the power supply PCB card out, then turn on the machine?

2. Im trying to understand what i touch and where... so once the PSU card is out, i turn the machine on, set the meter to 200 volts DC, touch the black probe to the chassis, then touch the red probe to the little metal pins (2 or 5) that the PSU card normally plugs into?

3. At this point i am looking for both to read + or - 20 DC volts right?

Sorry to be so thorough, but the more obvious the instructions are the better i feel about sticking metal probes into powered up vintage electronics.

[evm1024]

Nope. You leave the power supply installed but the box open to you can get to it. It needs to be in working order while measuring. Getting to what youu need to measure can be difficult at times.

I tend to use an alligator clip jumper wire to connect the black (negative) input wire to the chassis. This way I only have one probe to manage. Of course you need to place them where they don't short things out if they move.

Depending on the thickness of your probe tips you may be able to slide it up into the plug along the wire until it touched the metal connector inside. OR locate the trace on the PCB that comes off that pin.

In any case, the PSU needs to be powered and connected to the rest of the system except for test 3 where you remove one of the connectors on the PSU. Actually, you might want to do that first and see if it over heats.

Regards, Ethan

N.B.

PSU - Power Supply Unit
PCB - Printed Circuit Board

[The Great Cobb]

"Test 3: remove PS connector P4 and see if the heatsink gets hot. P5 sends the + and - 15 volts to the system. By pulling the plug you isolate the PS from the system. If it gets hot then the problem in on the PS board. If it is cool then the problem is in the system somewhere...."


Since the P4 is what the card plugs into, and i cant plug it in without plugging them all in (P2-P6), how do i go about unplugging just that one? Do you mean to pull the entire card and power it on?

Also, when you say PS, do you mean the power supply PCB?

The reason i ask this is because the card plugs in to either all or none.

MAJOR thanks for patience...

[evm1024]

Interesting, I've never seen inside on so don't know the layout. It looked from the schematic that there was a number of connectors. And I assumed that the PS PCB was mounted with standoffs and that there were 6 or so connectors that plugged into it.

I'll have to look at the exploded diagrams to get the payout of the land.

--Ethan


PS is the whole power supply which includes the series pass transistors and 3 terminal voltage regulators on the heatsink and the transformer. The PS PCB is just the board.

REgards, Ethan

Originally Posted by The Great Cobb

"Test 3: remove PS connector P4 and see if the heatsink gets hot. P5 sends the + and - 15 volts to the system. By pulling the plug you isolate the PS from the system. If it gets hot then the problem in on the PS board. If it is cool then the problem is in the system somewhere...."


Since the P4 is what the card plugs into, and i cant plug it in without plugging them all in (P2-P6), how do i go about unplugging just that one? Do you mean to pull the entire card and power it on?

Also, when you say PS, do you mean the power supply PCB?

The reason i ask this is because the card plugs in to either all or none.

MAJOR thanks for patience...

[sweetbeats]

Gents,

I'll be of more help when back home next week.

Ethan, the PSU PCB is of the plugin sort so the only way cobb can disconnect one connector is to jumper the connections to be left intact for the test. Cobb, I'll await Ethan's advisement but I *think* it'd be pretty quick and easy for me to make some jumper connectors so you can have the power supply card out of the unit but still connected and then be able to disconnect one or more of the connections.

Also, to assist in getting to Q1, Q2 etc., look on the back...the heatsink attaches to the chassis with just two screws. Take those out and you can tip the heatsink out and better access that tranasistors and the PCB to which they are attached. Do this with the 388 off and disconnected from the mains so you can properly position the heatsink and do whatever you have to to ensure nothing short circuits when you turn it on. I really like using that mesh rubber shelf grippy stuff you can get in rolls in the housewares section at, say, Wal Mart. Non-conductive and keeps stuff from sliding around.

[evm1024]

I see it now. the PSU plugs into the motherboard.... Oh well. One could unsolder the connector on the PSU PCB but that is a bit invasive....

We will see if Cory can come up with a jumper so that the PSU board can be removed from the system and then isolate the +-15 volts to the PSU.

Regards, Ethan