KSM27 Failing, almost dead

[antichef]

I decided to take one for the team and go get that KSM27 that the shop said was "distorted" - at least it was cheap . At first, all I got was whooshing and popping - I took it apart to see what was inside. Indeed, SMT:

Nothing visually wrong (on the other side, either) but who can tell with all that little bitty stuff? After I put it back together, I recorded a little bit, and actually heard a semblance of my voice coming through the din:


http://www.recursor.net/ksm27/ksm27.wav (warning - turn your monitors down). Really messed up, because I normally sound just like James Earl Jones.

Any ideas? Worst case, I guess I can pull out the capsule and do something with it.

 

[dgatwood]

http://www.recursor.net/ksm27/ksm27.wav (warning - turn your monitors down). Really messed up, because I normally sound just like James Earl Jones.

Any ideas? Worst case, I guess I can pull out the capsule and do something with it.

Unless the popping sound is associated with physically moving the mic or only starts after it warms up, that sounds pretty much like what I'd expect if the diaphragm has substantial condensation on it. The lack of bass response reinforces that theory; the capsule is nearly shorted from the condensation. Put the mic under about a 100W for an hour or so and try again. Leave it longer if necessary, but don't let it get too hot; you want to dry it, not cook it.

If that doesn't solve the problem, though, it might be a cold solder joint somewhere. The solution I always heard suggested for that was to reflow the solder using a toaster oven. (I'd unsolder and remove the capsule first, then reattach it afterwards. Mylar doesn't like temperatures nearly that hot....)

http://www.instructables.com/id/Toaster-Oven-Reflow-Soldering-BGA/

You can ignore steps 3-5 since you're just reflowing existing solder joints. Make SURE you do not do this if the board has a significant number of parts on both sides. I doubt it does, but it is worth checking. The last thing you want would be parts falling off the underside.

 

[Marik]

For comparison something a little bit simpler, but still sounds OK (at least what they say... eh):

I hate those SMDs. It is nearly impossible to trace even with schematics. On top, it seems yours is double sided crosseyes: ).
If it is not a capsule or cold joint I would not waste time, but gut it, get a nice xformer and build something worthwhile... say Neumann 87Ai style with no feedback (I suspect Shure capsule does not need HF EQ).
Simple, elegant, good sounding--can easily put that SMD thingie into shame. You can do it on a perfo-board.

Best, M

 

[Gus]

in the picture the top left large device should be a fet center leg gate, note the nice design of the PCB with the cutouts to isolate the high resistance section of the board.

first simple things

I looked at shure site it has pictures of the 32 and 44 replacement capsules but not the 27. Can you take the grill off and see what kind of capsule they are using in the 27 it could be dirt if it is more like a 87 capsule.

Next go to the shure site the 27 microphone is specked at 5.4ma, so you should have 48VDC -18.36VDC(2.7ma(1/2 of 5.4ma) X 6.8K) about 29.64VDC at pins 2 and 3 in the microphone. Maybe your preamps phantom supply does not supply enough current.

If not dirt get a mag lens and check the solder at the larger longer parts first then work down to the smaller part the longer one can have more stress with expansion and contraction from heat and handling at the pad to part solder connections.

there is more that can be done but I would check the phantom under load first.

 

[antichef]

I've got about -27.5 V when I check between pins 1 & 2 and between pins 1 & 3. I've tried a few different preamps (admittedly all of them may be underpowering the mic), and things are the same.

I took the capsule out and put it in a MXL 2001 (mounting system was totally different, so I just kind of set it in there) -- sounded fine - nice and smooth. I tried listening to the KSM27 without the capsule attached, and I got all the same popping and white noise. I didn't try putting the MXL capsule in the Shure. Put the Shure capsule back in place, and it sounded like the recording I posted again.

I poked around with the soldering iron on the larger solder joints, especially the XLR pins, but I can't make a dent in the problem. The circuit board is certainly 2 sided - there are quite a few (but not as many) components on the other side, including a 14 pin surface-mount IC chip. I suppose I could try baking it carefully - it can't get much worse . Drag -- from the little taste I got of the capsule, I bet it was a pretty good sounding mic at one time.

Guess maybe Marik called it -- this can be Microphones 102 for me? -- I might first try to mount the capsule better in the MXL2001, though - I think I could do it by drilling one hole in the capsule base-plate of the that mic.

 

[dgatwood]

So I got this same basic sound out of a mic I was modding a few minutes ago, minus the popping part. Turned out to be a shorted film capacitor (brand new). Not likely to be your problem, but it is certainly possible.

Have you inspected it carefully to look for any solder dendrites causing shorts between nearby contacts?

If you happen to be able to find a second one that works, it should be easy to track down the problem. Start with a volt setting and check the capsule supply voltage. My guess is you will find it to be nearly zero like I did. Work your way back through the power supply side of things using a second mic to figure out where you should have power but don't. When you suddenly get power on the defective mic, chances are you've just jumped past the defective or improperly soldered part.

 

[jonnyc]

I decided to take one for the team and go get that KSM27 that the shop said was "distorted" - at least it was cheap . At first, all I got was whooshing and popping - I took it apart to see what was inside. Indeed, SMT:

Nothing visually wrong (on the other side, either) but who can tell with all that little bitty stuff? After I put it back together, I recorded a little bit, and actually heard a semblance of my voice coming through the din:


http://www.recursor.net/ksm27/ksm27.wav (warning - turn your monitors down). Really messed up, because I normally sound just like James Earl Jones.

Any ideas? Worst case, I guess I can pull out the capsule and do something with it.

I'd email Shure, tell them you just bought a ksm27 and that it's acting up. Hell, maybe they'll send you a brand new one or at least repair that one. It might be a little dishonest but you might end up getting a nice mic out of the whole deal.

 

[Marik]

I've got about -27.5 V when I check between pins 1 & 2 and between pins 1 & 3. I've tried a few different preamps (admittedly all of them may be underpowering the mic), and things are the same.

I took the capsule out and put it in a MXL 2001 (mounting system was totally different, so I just kind of set it in there) -- sounded fine - nice and smooth. I tried listening to the KSM27 without the capsule attached, and I got all the same popping and white noise. I didn't try putting the MXL capsule in the Shure. Put the Shure capsule back in place, and it sounded like the recording I posted again.

So the problem is with the circuit. Your voltages look pretty close to the ones Gus posted. With 27.5V on either pin 2 or 3 the mic consumes about 6ma, which is within tolerances.
Do you have and oscilloscope and generator to view a waveform?

Best, M

 

[Marik]

I'd email Shure, tell them you just bought a ksm27 and that it's acting up. Hell, maybe they'll send you a brand new one or at least repair that one. It might be a little dishonest but you might end up getting a nice mic out of the whole deal.

Sorry, not sure if that's worth it. In my experience, it is much better to be straight to start with. This way the karma would not get back on you.

Best, M

 

[antichef]

wow - here's an article about a production batch of KSM27s with faulty capacitors:

http://shure.custhelp.com/cgi-bin/s...y5zZWFyY2hfbmwmcF9wYWdlPTE*&p_li=&p_topview=1 -- not exactly my problem (because in this case, there's no issue for the first 5 minutes, and my mic is instantly bad), but who knows?

I guess if this was mostly about getting the mic back up and running, I could call Shure and see how much it would cost to get it serviced -- yeah, I should have provided more detail in my first post about how I got it -- "AS IS", for cheap, with full knowledge that it was hosed -- trying to get Shure to take on the problem would definitely be a karma buster in this case. But maybe they do have some good out-of-warranty service options? I might check and see. (edit: their website says they have a flat rate $128 fee for KSM27 work, whatever that means-- maybe you just pay shipping and that fee? That would still keep me slightly below what looks like the used ebay price for one of these (but I'm not sure I'd buy one ))

(another edit: if I proceed with all my plans here, I could then call them and maybe get on their list of memorable tech support calls -- "Yeah, I read on the internet that I should bake it, so I put it in the oven. It got all toasty and melty, and I guess it does sound a little warmer now, but it still don't work right. Now about that warranty...")

I'll check the capsule supply voltage, but I doubt I'll get my hands on a working unit to compare with. Also, I don't have an oscilloscope or generator (soon, I hope ). So this may turn into a long term Frankenmic project for me, which is fine, too

 

[Gus]

I read the link and I would call Shure about what the fix might cost. You might have the same problem caps being bad all the time.

One can work on SMD with regular irons and things but you need practice.

I would not bake the board.

What does the capsule look like?

 

[antichef]

What does the capsule look like?

I'll try and post a pic later -- it's a 1 inch, encased around the edges in a plastic or nylon frame (it's loose, but you might have to destroy the frame to get it out) and attached to the frame is a stalk of the same material that seats in a rubber-ish base and is screwed into the frame.

 

[dgatwood]

I would not bake the board.

Agreed---not if it really has a significant number of parts on the back side---unless the parts on one side are all glued, but only if you are sure that they are all glued.

Definitely check the voltage at the capsule.

 

[Marik]

their website says they have a flat rate $128 fee for KSM27 work, whatever that means-- maybe you just pay shipping and that fee? That would still keep me slightly below what looks like the used ebay price for one of these (but I'm not sure I'd buy one ))...

Also, I don't have an oscilloscope or generator (soon, I hope ). So this may turn into a long term Frankenmic project for me, which is fine, too

Well, if you don't have the scope that might be a little challenging rolleyes, unless you have some shorted parts, which you could check with DMM. Indeed, the SMD takes some practice, so...
If you like the capsule then it might not be a bad idea to put those $128 towards a nice transformer... you got my idea .
Otherwise, let the folks in the Shure company to take care of it (most likely, they'd just change the whole board--much faster, less man-hour, and much less pain in the butt) and then sell on ebay.

Best, M

 

[antichef]

I called Shure - very good customer service!! I hardly ever call these numbers (and have mostly used equipment), so I'm probably easy to impress. Anyway, a replacement main board is pretty cheap, so I might go that route.

They acknowledged the capacitor issue, but since I couldn't tell them when the mic was purchased originally, they said I'd need to send it in for evaluation -- mainly to interpret the date code on the mic. Anyone know what this would look like and how to interpret it? I've seen discussions of other Shure products that mention a 4 digit thing where the last digit is the same as the last digit of the year, and the prior three digits is the ordinal number of the day of the year on which it was made. But I don't see any 4 digits (yet) that correspond to that pattern.

Anyway, worst case is that I send it to them, and they call and tell me it's not covered, in which case I can either pay the $128 general repair fee (which doesn't seem that appealing) or I guess ask them to send it back and then order the ~$35 main board and replace it myself. Guess it can't hurt to send it to them, so that's probably what I'll do. If things don't work out, then it's back to the lab

 

[antichef]

An update on this little saga: Over the last few weeks, I've had a very positive experience with Shure - I can see why they have a good reputation, and why they'd be desirable to work with for professionals in live sound and studios.

The mic was definitely not under warranty, and also not covered by the capacitor switch-out thing they had going on. They offered to fix it for their standard ~$128 fee, but when you add that to what I paid for the mic, it's a decent used price, but not a killer deal - instead, I ordered a replacement main board for $36 including shipping, and they sent back the mic as is with no charge (even for shipping!). Now I've got both (the board arrived before the mic), so I'll attempt the surgery soon. I'll take pics if I can.

 

[antichef]

It lives!!!

I had to pull one little transistor-y looking component off the old board to put on the new, but other than that, it was just a matter of attaching the capsule leads, the 5 leads from the two switches, and the XLR pins -- works like a champ now!

 

[pylet2000]

"I love it when a plan comes together."

 

[Pony!]

antichef I have the same problem with a ksm27. Do you have any idea what the part # would be, so I could order it direct from shure, as well. Also, could you, in detail, explain your replacement procedure?

 

[antichef]

Part no. 90A10698 "RoHS PRINTED CIRCUIT BOARD, KSM27"

I was lazy and didn't take pictures, but:

1) loosen tiny hex nut at base
2) somehow unscrew locking ring -- I used the hex wrench in one of the notches in the ring - I bet Shure techs have a tool for this
3) unscrew the four screws holding the grill in place (then half-turn and remove the grill -- careful not to touch the capsule!), then the three screws holding the circuit board in place
4) use a desoldering tool to remove as much solder as you can from the places where the XLR pins touch the board, then use an xacto knife or something to pop them loose if necessary
5) desolder the two capsule lead wires from the existing board (noting where they attach - it's pretty easy to remember, and the green one's marked)
6) slide the existing board out the side of the mice assembly
7) desolder the switches from the board (three pins for one, two for the other), paying attention to where they go -- again, easy to remember, and marked, I think.
8) desolder and remove the small three-legged transistor, one leg of which is right next to where the red wire was attached (it spans a hole in the circuit board) and solder it to the new board (my board didn't come with this transistor, so I had to grab the old one -- I also did this at the end because I didn't see the problem)
9) slide the new board into place, and put in the three screws to keep it there.
10) solder the switches to the new board
11) solder the XLR pins to the new board
12) reassemble and test

I reckon your warranty vaporizes at around step 1 or 2, btw