BF469, 470, 471, 472 vs. MJE340/350 are ON better?

[peranders]

Hi!

Does anybody have reliable information about MJE340 and 350? I wonder about Ccb, Cbe and Cce capacitance. I wonder also about the transition frequency. The datasheets from ON are very poor in that sense.

BFxxx are 20 years old, have worked fine for but those types are only 2 Watts. Does anybody have anything good (or bad) to say about the ON-types? I'm mostly interested in technical performance, speed, noise, peculiarities.

My application is cascodes for diff input stages and high gain second stages, and of course high output power amps +- 70-100 volts.

 

[HarryHaller]

MJE340 MJE350

SRC=MJE340;MJE340;BJTs NPN;Power;300V 500mA
*SYM=POWBJTN
.SUBCKT MJE340 1 2 3
* TERMINALS: C B E
* 300 Volt .5 Amp SiNPN Power Transistor 08-16-1995
Q1 1 2 3 QPWR .67
Q2 1 4 3 QPWR .33
RBS 2 4 6.5
.MODEL QPWR NPN (IS=36.8F NF=1 BF=212 VAF=311 IKF=.5 ISE=138P NE=2
+ BR=4 NR=1 VAR=12 IKR=.75 RE=.663 RB=2.65 RC=.265 XTB=1.5
+ CJE=18.8P VJE=.6 MJE=.3 CJC=12.2P VJC=.22 MJC=.2 TF=7.65N TR=294N)
.ENDS
* MJE340, MOTOROLA BIPOLAR POWER TRANSISITOR DATA BOOK,
* 1992, P.3-876.
**********
*SRC=MJE350;MJE350;BJTs PNP;Power;300V 500mA
*SYM=POWBJTP
.SUBCKT MJE350 1 2 3
* TERMINALS: C B E
* 300 Volt .5 Amp SiPNP Power Transistor 08-16-1995
Q1 1 2 3 QPWR .67
Q2 1 4 3 QPWR .33
RBS 2 4 6.5
.MODEL QPWR PNP (IS=36.8F NF=1 BF=365 VAF=311 IKF=.225 ISE=96.8P NE=2
+ BR=4 NR=1 VAR=12 IKR=.337 RE=.663 RB=2.65 RC=.265 XTB=1.5
+ CJE=39.1P VJE=.6 MJE=.3 CJC=25.4P VJC=.22 MJC=.2 TF=7.65N TR=294N)
.ENDS
* MJE350, MOTOROLA BIPOLAR POWER TRANSISITOR DATA BOOK,
* 1992, P.3-880.

 

[JonMarsh]

I've used the 340/350 series in past designs with good results. Hfe fall off at higher temperature and current is one drawback.

Another nice set of transistors in a reasonable voltage range are the Zetx FZT696 and FZT796, as long as 180 - 200V is OK.

-Jon

 

[peranders]

Sorry, I'm not a SPICE-guy (yet). Can anyone decode the SPICE data for me, please?

The subject line (topic) became funny. I ment if the MJE-types are better than the BF-types. Of course, the MJE are better (in power and in current) but in what sence besides that?

 

[JonMarsh]

Why you want to know SPICE a bit even if you don't use it...

A SPICE model is a mathematical description of the semiconductor behavior. A good SPICE model will give good insights into the behevior of the device, and it's nonlinearities as a function of voltage and current. (A "good" SPICE model is not always provided- many have shortcuts or simplifications- many parameters can be omitted, or incorrect, and the model still "works", after a fashion).

Besides it's usefulness in a simulator, if you chose to develop with simulation as an initial evaluation point for circuits, it's a very "compact" method describing in detail the device behavior- much more so than the many paragraphs of text which would be required to substitute, and which would be prone to ambiguity in comparison to the "numbers".

This may have been HarryHaller's "indirect" point, in addition to the actual information he provided. If you have a good grounding in semiconductor theory (math included), SPICE notation is a non-issue to learn. If you don't have a grounding in semiconductor theory, the rest of the discussion is a moot point, until you do. It's usually necessary to walk before running, yes?

Lastly, although it's helpful to have comments an information from others, there's no substitute in my experience for testing and characterizing the parts yourself - if you don't like On's curves, measure your own. And evaluate the performance in the circuit blocks you're considering using. I've built up a number of "standard" (for me) test fixtures with set operating conditions so I can easily compare devices in specific circuit functions, and evaluate distortion and bandwidth. This separates the wheat from the chaff pretty quickly.

Regards,

Jon

 

[peranders]

JonMarsh, I understand what you meen but sometimes it's much faster to know from datasheets and then design compared to testing. But if you know if a transistor has 10 MHz fT you can disgard it right away and save lots of time.

 

[Bonsai]

BF469/470

I'm doing some development work on an amp at the moment and have identified a problem with the MJE devices when used as the upper devices in a cascode configuration. If the amp is driven into saturation (i.e. clipping) it is prone to ringing as it comes out of clipping. If I run the simulation with the BF4xx devices (these were high speed, high voltage video amplifier transistors), the problem solved. I've actually had the problem on a prototype board and presumed the problem was due to stability problems. I now have reason to believe the issue is to do with higher base storage on the MJE devices - I ran a simple simulation and the MJE devices are quite 'sticky' and switch on and off after the BF4xx types. A Baker clamp does not solve the problem by the way.

I checked for the BF4xx parts on the Philips website - they are discontinued, and replaced with some similar SMD devices (SOT223 I think).

Does anyone know of a close equavalent to the BF4xx devices from another manufacturer?

I'm using LT Spice by the way.

cheers

 

[peranders]

Do you have any anti-clipping (or anti-saturation) diodes?

Can you post the schematics?

One trick is to avoid total saturation with diodes.

I'm also after good TO126 or TO220 substitutes.

 

[andy_c]

Re: BF469/470

(...) A Baker clamp does not solve the problem by the way.(...)

Have you tried looking at the base currents in simulation to make sure the Baker clamp is working as it should? There's some discussion of this issue here:http://www.diyaudio.com/forums/showthread.php?postid=298754#post298754 I had to do some special biasing to make sure the base voltage bias didn't get affected by the current in the Baker clamp diode, which can get rather large depending on the circuit topology.

 

[jcx]

since the signal in the collectors is current, input cascode devices can use series V dropping zeners or R||C

i haven't tried any but hi-V dmos might be interesting to explore in vas cascodes, Supertex (depletion mode too!) and Zetex have some fairly lo C devices up to 400 V

 

[Upupa Epops]

If you can replace BF types, use Sanyo 2 SA 1540 / 2 SC 3955. They are approximately twice expensively then BF types, but they have ft 300 MHz. Other parametres are good too : Pc 7 W, Uce 180 V, Ic 100 mA , case TO 126 ML ( full insulated ). They are specificated as " video transistors ", same as BF. Look also on same Sanyo pages - this firm have in program many other interesting types.

 

[steverb]

From memory BF469 & BF470 were
used in alot of power amp kits from designs in Australian magazines.
Ive had alot of 469.470's fail in one
particular kit design :may not have been up to spec (or the designer didnt check
soar curves) Now the Aus designs & kits use 340/350

 

[Samuel Jayaraj]

Transistors which have similar specs to the BF 469-472 devices are the BF869, 870, 871 & 872 with higher Pd.

What about 2SB649 and 2SD669 for driver stage. Very stable devices. Check Randy Sloan's OPTIMOS designs.

 

[Christer]

What about 2SB649 and 2SD669 for driver stage. Very stable devices. Check Randy Sloan's OPTIMOS designs.

Slone use them in all his designs in the power amp book too.

Seeing that you are from India, I wonder if you may know
anything about CDIL versions of these devices? I ordered
B649 and D669 from a swedish company, expecting to get
Hitachi devices - I didn't know anyone else made them and
there were even rumours they were going out of production.
The D669s I got were, or at least seem to be, genuine
Hitachi. However, for the other ones I got CDIL devices
marked P3B649A. I hadn't even heard of CDIL before but
found them on the net as a big Indian company. One question,
of course, is whether CDIL has a good reputation and can
be expected to do "true" second sources of the Hitachi
devices (don't get me wrong here, I am sure India is fully
capable of setting up a first-rate semiconductor plant, but
CDIL seems not very well known). The more interesting
question is whether these transistors are even true CDIL
ones? Looking at the CDIL site I could not find any devices
called P3B649A or even similar in their list of products. This
is very puzzling since I can't thus even confirm whether they
are supposed to be replacements for 2SB649, although one
might suspect so from the similar name.

Could you or anyone else from India perhaps shed some
light on this?

 

[ashok]

CDIL-India

Hi Christer,
CDIL has been around a long time. I was not aware they were still around. I used quite a few devices made by them. I never had any problems. They used to be a well organised company and are not fly by night operators.
Most of the current crop of power devices I think are imported in wafer form and just packaged here. This has been going on for long. I don't believe we have any significant wafer manufacturing plant now . The biggest we had burnt down ( how ??) a few years ago.
These days I get all my devices through friends from Singapore because most 'were' not available or were suspect to be fakes. I hear Bombay and Delhi has lots of fakes. In Chennai the local device dealers told me that there are many fakes here also . Most fakes are now of Chinese origin.
Your devices 'should' be Ok I think , if the part is really the same type.
Cheers.
Ashok.

By the way, I don't think any devices made here ( packaged here)and exported would be fakes ,especially from CDIL or ECIL and such other companies.

 

[Christer]

Ashok,

thanks for the answer. So presumably these transistors are
actually made from Hitachi wafers? As for fakes, yes since
CDIL seems not very well known internationally I don't quite
see why anyone would bother to make fake CDIL devices
instead of fake Hitachis. The thing that puzzled me most,
however, was that their product list didn't include these
particular transistor. OTOH I think the list had not been updated
for very long when I last looked at it. Maybe I should have
a look again.

Edit: It still does not appear in the product list. However,
when searching for it i get the message that CDIL does not
currently produce any such device. Maybe that could
mean they have done so previously and there are still
batches lying around in various places around the world?

 

[Bonsai]

VAS Issues etc

Here is the LT Spice circuit and some notes and stuff.

rgds

 

[ACOUSTICS]

SRC=MJE340;MJE340;BJTs NPN;Power;300V 500mA
*SYM=POWBJTN
.SUBCKT MJE340 1 2 3
* TERMINALS: C B E
* 300 Volt .5 Amp SiNPN Power Transistor 08-16-1995
Q1 1 2 3 QPWR .67
Q2 1 4 3 QPWR .33
RBS 2 4 6.5
.MODEL QPWR NPN (IS=36.8F NF=1 BF=212 VAF=311 IKF=.5 ISE=138P NE=2
+ BR=4 NR=1 VAR=12 IKR=.75 RE=.663 RB=2.65 RC=.265 XTB=1.5
+ CJE=18.8P VJE=.6 MJE=.3 CJC=12.2P VJC=.22 MJC=.2 TF=7.65N TR=294N)
.ENDS
* MJE340, MOTOROLA BIPOLAR POWER TRANSISITOR DATA BOOK,
* 1992, P.3-876.
**********
*SRC=MJE350;MJE350;BJTs PNP;Power;300V 500mA
*SYM=POWBJTP
.SUBCKT MJE350 1 2 3
* TERMINALS: C B E
* 300 Volt .5 Amp SiPNP Power Transistor 08-16-1995
Q1 1 2 3 QPWR .67
Q2 1 4 3 QPWR .33
RBS 2 4 6.5
.MODEL QPWR PNP (IS=36.8F NF=1 BF=365 VAF=311 IKF=.225 ISE=96.8P NE=2
+ BR=4 NR=1 VAR=12 IKR=.337 RE=.663 RB=2.65 RC=.265 XTB=1.5
+ CJE=39.1P VJE=.6 MJE=.3 CJC=25.4P VJC=.22 MJC=.2 TF=7.65N TR=294N)
.ENDS
* MJE350, MOTOROLA BIPOLAR POWER TRANSISITOR DATA BOOK,
* 1992, P.3-880.

Are you sure CJE=18.8E-12 , CJC=12.2E-12 (MJE340)

 

[lineup]

Are you sure CJE=18.8E-12 , CJC=12.2E-12 (MJE340)

Here are 2 models I from MultiSim.
CJE and CJC is 1E-11 everywhere

.MODEL QMJE340 npn
+IS=1.03431e-13 BF=172.974 NF=0.939811 VAF=27.3487
+IKF=0.0260146 ISE=4.48447e-11 NE=1.61605 BR=16.6725
+NR=0.796984 VAR=6.11596 IKR=0.10004 ISC=9.99914e-14
+NC=1.99995 RB=1.47761 IRB=0.2 RBM=1.47761
+RE=0.0001 RC=1.42228 XTB=2.70726 XTI=1
+EG=1.206 CJE=1e-11 VJE=0.75 MJE=0.33
+TF=1e-09 XTF=1 VTF=10 ITF=0.01
+CJC=1e-11 VJC=0.75 MJC=0.33 XCJC=0.9
+FC=0.5 CJS=0 VJS=0.75 MJS=0.5
+TR=1e-07 PTF=0 KF=0 AF=1

.MODEL QMJE350 pnp
+IS=6.01619e-15 BF=157.387 NF=0.910131 VAF=23.273
+IKF=0.0564808 ISE=4.48479e-12 NE=1.58557 BR=0.1
+NR=1.03823 VAR=4.14543 IKR=0.0999978 ISC=1.00199e-13
+NC=1.98851 RB=0.1 IRB=0.202965 RBM=0.1
+RE=0.0710678 RC=0.355339 XTB=1.03638 XTI=3.8424
+EG=1.206 CJE=1e-11 VJE=0.75 MJE=0.33
+TF=1e-09 XTF=1 VTF=10 ITF=0.01
+CJC=1e-11 VJC=0.75 MJC=0.33 XCJC=0.9
+FC=0.5 CJS=0 VJS=0.75 MJS=0.5
+TR=1e-07 PTF=0 KF=0 AF=1

 

[AndrewT]

2639 days and nights in the desert.
Bob Cordell's proposed models might give good information.