New ON-semi BJT. anyone used it?

[rchua77]

MJL4281A, MJL4302A released by ON semiconductors.

Has anyone any experienced with these new discrete products? rated 350V

 

[Cradle22]

Hi!

So far I have used in my P3A:
MJ15003 / 15004 with BD139/140 as drivers
Sanken 2SC2922 / 2SA1216 with D1763A / B1186A as drivers,
and those new ON-Semi devices MJL4281A / MJL4302A with MJE15034 / MJE15035 as drivers

I have to admit that I did not experience a lot of differences between those three device combinations. The latest one, which I still use in the amp, handles my rail voltage (42 V) better, especially the drivers do not get as hot as the BD139/140 combination. And "hissing" (only hearable if you put your ear directly to the tweeter of my speakers) is best with those devices. I have yet to measure how much those samples ON sent me match, could be that they are better matched than my randomly used MJ15003/15004 combo.

Very sturdy devices, though I bet that they will be pricey if they ever get available through "normal" suppliers (so far I did not spot them in Germany).

Bye,

Arndt

 

[millwood]

I have used the 4281 in a jlh1969, and compared it vs. tip41c. Not much of a difference.

 

[Nelson Pass]

I haven't used it, but it looks like a nice part with a
lot of potential - fast, and with good linearity up to
an amp, making it a good candidate for small or
parallel operation.

 

[osey]

Ill be completing my first amp channel with them very soon. Ill let you know.

 

[Jean]

I just finished putting them in my p3a amp but haven't had a chance to compare them yet. I used mjl 193/194 pair prior to the new BJTs.

 

[rchua77]

Thanks for the great response everyone. I'm honoured that Mr Pass posted a reply even

Wonder how these new BJTs would compare with the legendary Toshiba 2SC3281 BJTs ? By any chance, has anyone used the ON semi's version of the 2SC3281 ... the MJL3281A, a review would be great ??

Regards
Roger

 

[millwood]

I have compared 3281 vs. 4281 (actually both against tip41c) in a jlh1969. Not much difference between 3281 vs. 4281 for that application.

 

[Andypairo]

Thanks for the great response everyone. I'm honoured that Mr Pass posted a reply even

Wonder how these new BJTs would compare with the legendary Toshiba 2SC3281 BJTs ? By any chance, has anyone used the ON semi's version of the 2SC3281 ... the MJL3281A, a review would be great ??

Regards
Roger

The ON version is claimed to be even better than the Toshiba, it has been discussed on this forum also, maybe the search function will help...

Also on the ESP site you can find some useful info.

Cheers

Andrea

 

[Fred Dieckmann]

ossssssscillations

"especially the drivers do not get as hot as the BD139/140 combination. And "hissing" (only hearable if you put your ear directly to the tweeter of my speakers) is best with those devices."

This could be a sign of parasitic HF or RF oscillations. BTDT. In four different manufacture's data sheets, I can't even find Ft for the BD139/140. They are really not suited for drivers for amps with high speed output transistors.

 

[Elso Kwak]

Re: ossssssscillations

"especially the drivers do not get as hot as the BD139/140 combination. And "hissing" (only hearable if you put your ear directly to the tweeter of my speakers) is best with those devices."

This could be a sign of parasitic HF or RF oscillations. BTDT. In four different manufacture's data sheets, I can't even find Ft for the BD139/140. They are really not suited for drivers for amps with high speed output transistors.

Hi Fred,
Strange thing is no one mentions the Ft of the BD139/140.
I don't remember were I found it but it appeared Ft is typically 50Mhz. Amplifiers mysteriously heating up up under no load and "hissing" might be an indication of oscillation indeed.

 

[fscarpa58]

Hi

I just downloaded the spice model for MJL4281A and MJL4302A from onsemi site.
They report a BF value of 191 and 505 respectively.
Is it normal for a complementary pair?
I also noticed a VTF=100000 for 4302a.
Maybe models will work but when physical parameters
are so different there is an possible case of non-identificability due to over-parameterization.

I also find the new online MJ21194 model reports BF=10000 VAF=10000
(updated sept 10,03) while the MJL21194 has BF=41.9 VAF=34.2.(Jul 29,97)


any comment?

Federico

 

[Christer]

BD139/140

Just for the h**k of it I checked the datasheets for these that
I happened to have downloaded, not expecting to find any
more info. But guess what, the Philips datasheets do indeed
specify ft. The typical values are 190MHz for BD139 and 160MHz
for BD140. Don't trust these figures to apply to other brands,
though.

 

[Elso Kwak]

Correction, Found it back

Hi ,
I have to correect my post as the 50MHz is a minimum gain bandwith spec. (Ft) for the BD139.
Found it in a very old Transistor D.A.T.A. book. For the BD140-6 & BD140-10 same value.

 

[Upupa Epops]

Hissing

To cradle 22 : When you will have any osciloscope, you will see, that human's ear sometimes is not " last arbiter " of quality of amp - oscilation, oscilation ..... 10 Mhz or 20 MHz or 50 MHz or 100 MHz ...

 

[Audiofanatic]

Jaeger Elektronik Book!

Hi ,
I have to correect my post as the 50MHz is a minimum gain bandwith spec. (Ft) for the BD139.
Found it in a very old Transistor D.A.T.A. book. For the BD140-6 & BD140-10 same value.

Hi Elso,

You are right!
In my book hey are marked as > 50MHz for both!

Best regards,


Audiofanatic

 

[Christer]

Is any of those data books for a particular brand, or just the
kind of generic data books? Data typically will differ between
brands, and I really wonder why so many manufacturers are
unwilling to supple certain important data such as Ft figures,
hfe vs. Ic diagrams etc. etc.

BTW, I had a hunch I had earlier dismissed those devices as
questionable for Ft reasons, so I checked the catalogue from
my local supplier Elfa, who sells both Philips and ON versions.
The ON had no Ft figures, just as in the data sheets. The Philips
ones had Ft figures differeing substantially from the datasheets,
in one case much lower, in the other case much higher. To check
if they had made an error or different data sheets from Philips
I checked they web catalogu with pdf datasheets available.
Turned out they had suddenly stopped selling the Philips ones,
so now one is left with the ON ones with unknown Ft
figures. I don't know what Elfa are up to nowadays, it
turned out they had also discontinued the NPN version of those
very good Toshiba drivers. Why keep the PNP but not the NPN It would have made at least some sense to do
it the other way around.

 

[pjacobi]

OnSemi models

[...]
I also find the new online MJ21194 model reports BF=10000 VAF=10000
(updated sept 10,03) while the MJL21194 has BF=41.9 VAF=34.2.(Jul 29,97)
[...]

I've seen very nonsensical parameters on OnSemi models, e.g. there is a D45H11 model with VAF=3.13283 and
TIP147 model with NF=0.28992.

I assume some automatic process (data conversion/extraction) has gone berserk and nobody checks the output.

Regards,
Peter Jacobi

 

[fscarpa58]

I agree

Federico

 

[millwood]

Re: OnSemi models

I assume some automatic process (data conversion/extraction) has gone berserk and nobody checks the output.

Regards,
Peter Jacobi

I suspect the process is some sorts of curve-fitting: they measure a typical transistor and plot various curves and input them into the computer which spits out those models. So it is possible that maybe more than one of those parameters don't make sense but in their totality, they do.