|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
208.86.125.246
This is the second time I have seen this, both times it was a piece of tube equipment. Both times it was some problem causing a considerable B+ current draw, the diode(s) got so hot they unsolder themselves. The first time it was an HP oscilloscope and the diode unsoldered itself, fell down to the bottom of the case and the solder on the through holes re-bridged so it looked like the diode was never installed there.
The second time was yesterday, an ARC SP3 that has a lot of diodes in series to act as a fixed voltage source. This time the board was horizontal so the diode was still in place but unsoldered, with an intermittent connection.
I was curious as to how common this is. Have others run into it?
Follow Ups:
You are making me paranoid Fred. I used Wonder Solder for years. It has a low melting point.
I had a SP3A-1 preamp and did mods on it many years back. Seems I remember unsoldering the stock components was difficult. ARC must have used a fairly high melt temp solder?
I use Wonder Solder, too. At what temperature do you set your soldering iron? I seem to need the same 700-degrees F that I've used for all my other types of solder, whatever they were.
Thanks!
John Elison
Back in the 1980s I modified my H-K Citation V with Zener screen grid regulators. I used 5W parts and had to regulate each of the 4 screens independently to stay within the 5W rating. So each Zener also had a 5W resistor in series. These were each attached beneath its respective pentode. They worked fine, but eventually the heat generated by the Zeners and the voltage dropping resistor along with all the rest of the heat within the chassis caused the solder holding the diode to the standoff to melt leaving just the mechanical connection intact.
I realized that this was a bad situation, so I triode connected the pentodes.
and no room for heat sinking in the case?
probably a stupid question otherwise you would have, correct?
regards,
No, the problem was that there were too many components too close together and all of them producing heat and directly under a tube which was generating considerable heat too. My latest amp has screen grid regulation for the input tube and the output tubes, but the chassis is larger and better laid out for this kind of situation.
I built a full wave power supply for an attic fan made in China that use to be solar (panel failed). I wanted it to stay on longer and during cloudy days. Just a timer and set to max voltage with a variac. Better than me going up in the hot summer attic and doing more insulating.
Eventually the motor shorted out and the diode heat melted the solder. I used 25 amp diodes and they were still good after.
For a couple years it worked quite well anyway. Now I have one that's 120vac.
DC solar fan, good idea. I did that for pool recirculating pump but settled for brushed motor, brushless are difficult to find in that size. After about 6 years I got tired of changing brushes and added an inverter to run AC pump motor.
Same thing, diodes were good in both of my cases and I reused them.
I don't think I've EVER run into this. How much resistance could a diode be seeing to cause that much heat? We've all seen resistors cooking but never found any that have come undone.
Yep, that is weird.
I've seen it with resistors on through hole PCB a few times ... not where the entire part fell away though, just one 'leg' lifting when the resistor was installed 'torqued'
it's certainly a good argument for heat sinking eh?
regards,
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: