With apologies to Wayne Carini.......
One of the things about working on old equipment is that sometimes the parts are hard to find, especially if they were declared obsolete years, or decades, ago.
Now while most passive components like resistors and capacitors are readily available, some of the semiconductors are not. Diodes are easy to find, and even the infamous "Darlington Power Packs" now have discrete versions available, but most of the Small Signal Transistors have been extinct for quite some time.
The issue then becomes finding suitable replacements, of modern manufacture. Most "US" transistors have a "2Nxxx" number on them, like one of my old favorites, the 2N2222.
Japanese transistors use a completely different numbering system, as explained on Wikipedia:
"The Japanese technical standard JIS-C-7102 provides a method of developing part numbers for transistor devices.[1] The part number has up to five fields, for example in the number 2SC82DA:
- The first digit "2" indicates this is a 3 lead device ( a diode would have a prefix numeral 1)
- The letters "S" is common for all EIAJ registered semiconductors
- The following letter designates polarity and general application of the device. For transistors:
- A PNP high frequency
- B PNP low frequency
- C NPN high frequency
- D NPN low frequency
- E P-gate thyristor
- F N-base unijunction transistor
- J P-channel field effect transistor
- K N-channel field effect transistor
- M bidirectional triode thyristor
- The numerals following indicate the order in which the application was received, starting at 11
- A suffix letter indicates improved characteristics....."
OK, so that at least tells me about what they are, but to find out exactly what they are, I had to get the datasheets for each (obsolete) part number. The datasheets list all the parameters for the described part number, and from there, you can use the selection guides on the vendor websites (I like Mouser the best) to find currently available parts with matching specifications.
But what if you can't find a datasheet? Remember, these are obsolete parts, and sometimes a datasheet hasn't been scanned and posted somewhere. Well....then you're getting down to hand-to-hand combat. First you look at the schematic, and see what the applied voltages are, and make a best guesstimate as the the current and power dissipation required.
This is the schematic for the right channel of the power amp :
At this point I've found current replacements for all but three of the original transistors, but it's going to take a bit longer to decide what to use for the last three.....
%20American%20Airlines%20Flight%20Tracking%20and%20History%2004-Aug-2024%20(KDFW-KEGE)%20-%20FlightAware.png){kind=tracking}
Engineers are a special breed. I was once told an engineer can be one of two things; absolutely wrong or approximately right. I'm humbled by your patience.
ReplyDeleteI'm obligated to finish a project I've taken on for someone else, extraspecially when there's a cost involved. I don't mind doing all the sleuthing, and last night I found a recommended sub for one of the missing three parts I don't have the datasheets for.
DeleteAnd I've just about finished the spreadsheet to load into your little radio.
Electronics voa the Sherlock Holmes method....
ReplyDeleteWhatever works!
DeleteAs I recall, I used to use 2N2222 and 2N2907 complimentary pairs a lot in my early days of cobbling experimental circuits together at Hughes, until the microcircuits revolution came along. Then it was off to the races with those. Fun times.
ReplyDeleteYou Betcha! Their Ft is 300 MHz, and I've used plenty of them, too. 2N3904 is another oldie-but-goodie!
DeleteI remember working on an old 45 remember player with blown final transistors. The TO-3 cases had water filled cans attached to them as heat sinks. They were also PNP geraniums. Talk about a nightmare.
ReplyDeleteThe only thing I have with germanium transistors is a Heathkit GC-1A "Mohican" shortwave receiver.
DeleteThirty six years ago or so, I worked on a program making replacement printed wiring assemblies (populated printed circuit boards) for the Lance Missile (MGM-52, tactical missile) GMSTS (Guided missile system test set). Since the Lance was capable of carrying nuclear warheads redesigning any test equipment was difficult because it had to go through stringent approvals therefore most of the system stayed the way it was designed. The test set was designed in the mid 1960s. It used discreet diode/transistor logic and it used LOTS of it. Even though we were using the same 2N transistors as it was designed with, improvements in the design and manufacture of those transistors made them "better". But for us that didn't work. We would run into timing problems because some of the new transistors were switching faster than the originals. To fix this they had come up with a test set to classify the switching speed of the transistors. We would test a lot of them and put aside the slower ones for repair use. The rest of the transistors were put into production. Similar issues make it interesting to select a replacement for out of production unobtanium transistors or even know replacement transistors.
ReplyDeleteI did something similar when I worked for a medical electronics company in the lat 1970's. We had a product that was picky about the Hfe of a certain transistor, so we had to hand-select them. We had a Teradyne components test system, programmed with magnetic strip cards. We wrote a routine that would measure the beta, and then I sat in front of the rack and tested transistors to find the ones with the right beta, and separate them from the rest. The selected ones were marked with a color dot, and went to production/spares, and the remainder were used elsewhere.
Delete