I have a Crown Macrotech 600 I have to repair for a friend. I have oodles of MJ21194/94 TO-3 output transistors that I can use.
Can anyone tell me why they 21193/94 wouldn't work in place of the MJ15022/23. Seems like it would be an upgrade to me over the old stuff. The two specifications read almost exactly the same. We used the 21193/94 to build KSA-50's and KSA-100 Klones....
Thanks!
Mark
Can anyone tell me why they 21193/94 wouldn't work in place of the MJ15022/23. Seems like it would be an upgrade to me over the old stuff. The two specifications read almost exactly the same. We used the 21193/94 to build KSA-50's and KSA-100 Klones....
Thanks!
Mark
The prime difference between the 21193/4 and the older types is gain linearity. That is, Hfe doesn't fall off as quickly when output current rises above an amp or so and hence distortion is considerably lower as power increases with these newer types. Usually, that's a good thing, and your sound quality will be better too.
Be careful fitting newer faster transistors.
I rebuilt a Maplin 225WRMS bipolar amplifier with new 2n3055/mj2955 and it oscillated badly. I had to increase the VAS capacitor to stop the oscillation.
I put the new transistors on my semiconductor analyser and the gain was massive compared with older transistors that came with the amplifier.
This extra gain caused oscillation.
I rebuilt a Maplin 225WRMS bipolar amplifier with new 2n3055/mj2955 and it oscillated badly. I had to increase the VAS capacitor to stop the oscillation.
I put the new transistors on my semiconductor analyser and the gain was massive compared with older transistors that came with the amplifier.
This extra gain caused oscillation.
I've used both types in Macro/Micro/Geo/PB and similar crowns. No problems. I've never noticed any difference in that application...
John
John
Enzo told me to not intermix them is the worst thing. They are different and don't thermally track. I replaced 19 blown MJ15024/25 with 20 MJ21193/94 in a Peavey amp, so far so good. I don't think the Ft difference is as much in those two as 1970's 2n3055 (200 khz) versus >1985 2n3055 (>2 mhz)
Actual examples of 21193/4 may measure as faster than the older part types due to parameter spread and general technology advances over the years but I haven't found that a problem if the parts are matched properly or the whole set is changed.
The different basic DC gain and gain linearity could be a problem in a mixed set, as indianajo/Enzo may be suggesting. At high power, the Gain differences between them are quite significant and that is enough to cause problems with current sharing when it really matters.
I don't think the OP is suggesting a mix of parts but it's good to have that possibility raised.
The different basic DC gain and gain linearity could be a problem in a mixed set, as indianajo/Enzo may be suggesting. At high power, the Gain differences between them are quite significant and that is enough to cause problems with current sharing when it really matters.
I don't think the OP is suggesting a mix of parts but it's good to have that possibility raised.
I also wasn't suggesting mixing types in one output stage. I've always replaced complete sets. Also, buying from one batch from a reputable supplier I've found the matching to be quite good - easily as good as Crowns numbered matching parts.
- Status
- Not open for further replies.