DIY JE-990 DOA

[kazper]

Just a few tips that I wanted to share, turn the tit on the LM394 down with a sharp pair of Needlenose or clip off. Someone else posted to install the LM394 first before Q5.

Use heat shrink to connect the Q3/CR3 together before sodering.

I soldered CR9,CR10 and heat shrinked them and put them through the holes on Q8 and Q9 keeping the diode in the center of the transistor. I then gently pulled the leggs into place and it seemed to look beter than trying to wrap the leg around. I then coverd it all with a larger peice of shrink and enclosed it to gether around the sides.

I think it would be easyer to mount the pins if they were stuck in a socket and then soldered to the board.

 

[evm1024]

I was a bit confused in reading that it was hard to get 30 gauge in the inductors. Just to be sure ther is no confusion the higher the gauge number the smaller the wire.

Gauge Diameter (inches)
22 0.0253
26 0.0159
28 0.0126
30 0.01

I'm using the red wire from that same radioshack 3 wire package. It is 30 gauge and about 0.01 inch in diameter. John Hardy appears to use 28 gauge wire in his 990. Also, The wire in John Hardy inductors is not insulated. Just 28 gauge tinned copper.

I'm at a loss as to why R7 would fail. Looking at the schematic I see that it is the emitter resistor for Q6. I don't see any obvious path to -24v for it...

I've not had the time to test my 990 out yet. I was hoping to get to that this week. As for Q3/CR3 and CR9/10. I used epoxy to bond them.

Hey, Kazper. I'm in Vancouver also. Just off 164th. I'd like to see what you have some time.

 

[Lowfreq]

Am I doing something wrong here, or has anyone else had trouble soldering the wire from the inductors to the board?

For the life of me, I can't seem to get the solder to grip the wire.......

 

[Ptownkid]

Did you strip the lacquer at the end where you'd like to solder?

 

[sbranco]

Hey EVM,

Thanks! Smaller guage = larger diameter wire. Its not marked on the Rat Shack spools but I should have been able to figure it out from the overall package and how much wire was supplied for each guage.

No wonder I had trouble threading that 22 guage wire in the inductor! I think I will switch to the 26 guage (green stuff) or maybe just tighten up the 30 guage (red wire) and solder it closer to the board.

 

[Lowfreq]

[quote author="Ptownkid"]Did you strip the lacquer at the end where you'd like to solder?[/quote]Yep, did that. Still no luck.

I'll try it again and see what happens.

 

[kazper]

I striped it on the end of a table with a razor blade leaving about 1/32" under the core left, I did it on both sides and had not trouble, you should see nice virgin copper. Try that and see what happens.

 

[Ptownkid]

weird, I had no problems at all.

 

[kazper]

I will admit that the first one will be lucky to make noise, the second unit looks like I bought it from the store. I only have 4 more to go, but I'm going to wait till I fire these 2 up first. Need to build up a test jig and then take them with me to work to use a good power supply.

I think I should order 2-4 more and use them in my future build of the Fred Forsell opto comp.

 

[evm1024]

I built 8 990 that have been waiting to get tested. They have been sitting for months waiting for a JLM Dual 99v preamp to get completed. Well the dual pre is mostly finished and ready for testing. I've configured it like the Jensen twin servo.

So, I breadboarded up a power supply and a test jig and popped the first 990 in. Power on and Poof up goes a puff of smoke. Poking around (egg on face mode on) I discover that I had swapped the output transistors (egg on face mode off). All 8 built the same way. Not sure what I was thinking. Perhaps I just think that the positive rail should be on top.

Pulled the blown transistors and put in some BD139/140 that I have around and give it power again. No problem. The jig is set up non inverting gain of 10. Fed it DC to 200 kHz without problems.

It is going to be a pain to remove the transistors and put them in the correct locations. I might try solder wick rather than a sucker....

 

[kazper]

Just so I don't mess mine up, they face the inner part of the chip correct, back faces the outside if the board. I think thats the way it shows on the board and the picture...

 

[evm1024]

The front of the transistor (with the writing) faces the inside of the board. THe back of the transistor faces outside.

Swapped Q8 and Q9 (NPN in PNPs place...

 

[sbranco]

Regarding the smoked R7, I replaced it with a new one although even smoked, it still registers 180 ohms. Anyway I tried replacing Q6 too. The doa still makes a crazy screeching and lots of noise. Have to track it down but for now it seems easier to just put together another one. At least we know it wasn't just a bad Q6.

 

[sbranco]

Ok I just finished my third 990 DOA. This one has the same problem as the other one with a problem. Whatever it is, I did it again. Sounds exactly like the other one with buzz or screeching or whatever you want to call it. FRUSTRATING. All the components are exactly the same in all three. I haven't fried R7 yet, but I haven't left it on for more than a couple seconds. I think I've been very careful checking over the components and one of them works fine. AAAAARRRRGGGHHHH!

 

[evm1024]

[quote author="sbranco"]Ok I just finished my third 990 DOA. This one has the same problem as the other one with a problem. Whatever it is, I did it again. Sounds exactly like the other one with buzz or screeching or whatever you want to call it. FRUSTRATING. All the components are exactly the same in all three. I haven't fried R7 yet, but I haven't left it on for more than a couple seconds. I think I've been very careful checking over the components and one of them works fine. AAAAARRRRGGGHHHH![/quote]

Post a couple photos from different angles and the bottem of the board. Perhaps other eyes will see something.

 

[sbranco]

I'll try and post some pics today. I have gone over it again and again. And I have one that works and two that do not. I can't tell any difference in the one that works from the others except that it looks like I used less solder.

Its a pain to replace components in these little babies. When I built the third one, I checked each component as I put it in against the other two and all three appear to have identical components. As far as I can tell, I didn't do anything obvious like swap a 4250 transistor with a 2484 transistor.

All the diodes are placed the same way. None of the caps are polarized so that's not an issue. Having two doa's with the same faulty component seems unlikely.

It is definitely the doa because I can pop the good one in and it works. Swap either of the bad ones in and neither works. And both bad ones emit the same rotten sound and pass nothing from the mic.

Pics coming.

 

[sbranco]

OK here are photos, the good, the bad and the ugly....

To tell them apart quickly, the good one has a check mark on the top of the LM 394 can. The third try at it has the green wire in the inductors.

OK here is the second try that did not work

And finally the third try that did not work.

My soldering skills are certainly questionable. I tried to deflux the third one a little on the bottom and it ended up looking pretty bad. I need to get a brush to use with the defluxer to get more of it off and more defluxer as well. The bottom of the first one looks the best and it is chuck full of rosin.

Thanks for looking over these photos. Not sure they are a builder's how to since two of the three do not work. But they are some detailed pics... that might help some.

 

[Sender]

Try using some flux-off in the bottom side. I has an instructor told me that flux can, in some case, be conductive.

Even if the flux is not causing a short, cleaning up the bottom will give you a better view of the bottom where you might able to spot a short. Sometimes I also like to srape (very gentely) the flux off. I have a set of dental tools that is great for this :thumb:

 

[Kev]

cooool


:green:
no surface mount crap here

and
real lead solder

that's proppa DIY

I use an acrylic thinners from the local Bunnings (DIY shop) and a combination of brass/wire brush and finish with folded up hand paper towel.

The paper flakes a apart and takes the flux and crap with it
makes a real mess on the floor and gets stuck to the feet
LOVE IT ! ( I should get back to doing some DIY )
and then as the paper drys out it will polish the solder to a mirror shine

a quick spray with PCB laquer and they you have it
purrrfection
:green:

 

[evm1024]

[quote author="sbranco"]OK here are photos, the good, the bad and the ugly....

To tell them apart quickly, the good one has a check mark on the top of the LM 394 can. The third try at it has the green wire in the inductors.



SNIP!

My soldering skills are certainly questionable. I tried to deflux the third one a little on the bottom and it ended up looking pretty bad. I need to get a brush to use with the defluxer to get more of it off and more defluxer as well. The bottom of the first one looks the best and it is chuck full of rosin.

Thanks for looking over these photos. Not sure they are a builder's how to since two of the three do not work. But they are some detailed pics... that might help some.[/quote]

I'm looking at your photos and at a working board of mine. Everything that I see looks OK. This moves me into speculation mode.

If you look at the schematic of the 990 you will see that R7, C2, C3 and the emitter of Q6 are all tied together. Nothing there looks like it could fry R7. But when looking more closely at the boards I see that C3's leg (the one closest to Q6) is surrounded by the ground plane. If this leg had a solder bridge to the groundplane you would have the + supply voltage across R7. Assuming 24 volts that would equate to 132 mA or 3.16 watts. I would expect a 1/4 watt resistor to smoke with 3 watts dissapating in it.

SO, a close visual inspection and perhaps measure the voltage on Q6's emitter (Top of R7 is the place). It should be close to the + rail (24 volts) and not close to ground.

Of course with a blown R7 you will not be able to measure properly. Replace it and smoke it again. Or perhaps you could check the resistance to ground. It should be greater than 0. Presumably this will not blow up Q6.

Regards