Hey Guys,it's been a while I know. Hope everyone is well. I have a nonstompbox related question for you.
I have a receiver unit that has come out of an old stereo console. It is a Zenith Z940 Chassis 25ZT120. It utilizes all PNP transistors throughout and they are germanium too. The unit is from the 60s. It came to me from a guy who refurbs the cabinet and I do the electronics for him, 3 as of now. I recapped it, changed lamps, replaced the mica insulators and regreased the output transistors, etc. I listened to it for a few days without issue. The speakers in the console must be more efficient than my bench speakers because he brought it back to me with noise on one channel. I heard the noise with my ear right up to my bench speaker and traced it down to a Zenith 121-399 transistor that appears to be a driver for the output trannies on that channel. I have no schematic. NTE has a one fits many replacement, NTE102A, for $8.83 a piece. I would buy two and replace the other one in the other channel but we want to keep cost down. I have replaced it and the corresponding one on the other channel with 2N2907 silicons in the TO-92 package. The original 121-399 were metal can like the NTE replacement listed below.
https://vetco.net/products/2n222-pnp-transistor-nte102a
My question is related to heat. The germanium ones had thin metal round sleeve heatsinks that I don't think fit the TO-92. I know germaniums are very heat sensitive and silicons are less so. The issue is that the TO-92 seems to run a little warm to the touch. My cheap laser thermometer reads only 96F on the top but they feel much hotter than that. Got up to 105F at one point. They don't burn you but you still are hesitant to leave your finger there for long. The datasheet says these can operate at 150C or 302F. I can't imagine running a TO-92 anywhere near that and have never seen one run that hot that didn't blow up. I know I'm no where near that value but I am still concerned.
The things has been running ok for an hour now and sounds good. Any thoughts? Am I worrying too much like usual?
Quote from: bloxstompboxes on December 20, 2023, 10:03:48 PMThe datasheet says these can operate at 150C or 302F.
That's the maximun junction temperature allowed.. :icon_wink:
(https://i.imgur.com/0pEUP7F.png)
You can work on thermal resistance calculation (https://www.giangrandi.ch/electronics/thcalc/thcalc.shtml) to find a proper heatsink, like below:
(https://i.imgur.com/Lju2aPg.png)
Thanks antonis. I think I'm being overcautious, which would make me a poor design engineer in today's bean counter, throwaway, short life design society. I will, due to the minimum cost and likelihood of needing more of these in the future, go ahead and order and recommend to my customer that we install heatsinks like the one you listed above.
Quote from: bloxstompboxes on December 21, 2023, 09:04:56 AMI will, due to the minimum cost and likelihood of needing more of these in the future, go ahead and order and recommend to my customer that we install heatsinks like the one you listed above.
I presume a TO-92 package with a brass clip on heatsink will be more efficient to heat dissipation than a TO-1 one..
NTE102A is rated at 90
oC junction temperature, significantly supported by its metal casing, but both packages are rated at 600mW..
Quote from: antonis on December 21, 2023, 11:27:59 AMQuote from: bloxstompboxes on December 21, 2023, 09:04:56 AMI will, due to the minimum cost and likelihood of needing more of these in the future, go ahead and order and recommend to my customer that we install heatsinks like the one you listed above.
I presume a TO-92 package with a brass clip on heatsink will be more efficient to heat dissipation than a TO-1 one..
NTE102A is rated at 90oC junction temperature, significantly supported by its metal casing, but both packages are rated at 600mW..
Yeah, it's a no brainer. Mouser has them for like 1.41 a piece. Kind of a rip off for small piece of thin metal, but then again they are not in high demand and it is a small price to pay for peace of mind. Much cheaper than the NTE substitute which I'm not even 100% sure, most likely it would, will work.
If a silicon works in that position, consider using a silicon device in the TO220 package. The TO220 can dissipate up to about 2W with no heat sink at all. There are some higher-gain TO220 driver devices that should work OK. You'll have to cruise the data sheets to pick one with a gain over 150-200 for likely success.
If you are CONUS, i can drop these in a envelope if want. Some are npn unknowns but the heat sinks are usable.
(https://i.postimg.cc/wyvVpJ4M/20231221-122521.png) (https://postimg.cc/wyvVpJ4M)
Did a look at Mouser. You might be able to use https://www.mouser.com/datasheet/2/308/1/ENA1799_D-2311172.pdf (2SA2222SG, 0.85 plus shipping) with a gain of 150-450, or maybe KSB1015, MJE15035, or TTA1452B. These are all under $2 each. Shipping from Mouser will set you back about $8, but they're fast at shipping. They're all massively more rugged than any germanium or a TO-92 silicon, even with a heat sink.
Thanks R.G. and mozz. Everyone here is still so helpful and generous!
I went ahead and ordered the heatsinks and some more 2n2907 as they work elsewhere in these receivers too. These guys are socketed, so unless I did some modding or mounted them to the backside of the sockets I'll have to stick with TO-92. Working on these old things is a bit of a pain as they are all point to point wiring too. But it's good practice though. I will, however, keep all these suggestions in mind if this fails.
I think it'll be alright since they worked fine for an hour at around 2/3 volume without heatsinks. Adding them on should just help future-proof the repair. At 106F without heatsinks, they ought to be a lot closer to room temp with them. The 2N2907 are rated at 625mW Pc and the NTE102a at 650mW. The addition of a heatsink should suffice. The gain on the originals was IIRC around 200. The 2907 were just a working drop in without any rebiasing needed.
Quote from: mozz on December 21, 2023, 12:28:28 PMthe heat sinks
That type, ARRL Handbook says you make them at home. Pick a drill bit with shank slightly smaller than the transistor. Wrap it and use a vice to crimp it tight. Drill for a bolt & nut to make it tight on the transistor. The hood metal from a classic Jaguar or MG is good alloy and suitable gauge................. :icon_rolleyes:
Actually, at this level steel is adequate sinking and I'm throwing out a karaoke DVD player, the lid would give a lifetime supply of such sinks, or brackets, deflectors...
Here is the finished product if anyone is interested. The heatsinks and trannies in question are the two black rectangular ones in between the long silver metal heatsinks going front to back behind the dial.
(https://i.ibb.co/xm45m1B/IMG-4682.jpg)