fast high current transistors

[fsdenis]

Might anybody know of transistors with 300mhz
gain-bandwidth product good for 2 amps or more?<p>I'm parallelling 2N4401 npn's and 2N4403 pnp's to
make a driver good for 1.2 amps out but it takes
up too much board room to be practical.<p>These are similar to NTE123A and NTE159 transistors.<p>Performance of the parallel network is excellent.
I just need something physically smaller to do the
same job. <p>Thanks in advance for any help.

[russlk]

Higher current generally means lower bandwidth, so if the 2N4401 is working for you, try the MMBT4401 which is in SOT-23 package and is nearly as small as you can get without going to wire bonding dice.

[fsdenis]

Thanks, Russ.
It seems you are right about higher current meaning lower bandwidth. I have found nothing in
a TO220 package that has the low (250 ns) storage
time I need.<p>It looks like I will just need to come up with a
mechanical packaging technique to shrink this driver down. A lot!

[Externet]

Hi.
What about power output transistors used in VHF/UHF transmitters ? Motorola makes some.<p>Miguel

[fsdenis]

Thanks Miguel,<p>I'll look.

[Ron H]

High Ft doesn't necessarily equate to low storage time. Also, if you don't really need the gain-bandwidth that higher Ft gives you, you can reduce storage time with a Baker clamp if you can tolerate higher Vce(sat). Do a Google search on "Baker clamp". You can also reduce storage time by using an aggressive base turn-off circuit if you are using the transistors as switches.

[fsdenis]

RonH:<p>I've begun Google search "baker clamp" and have
already found ideas new to me that I might be able
to put to use in this or similar applications problems. I'll need time for digesting this stuff.<p>I've found examples of high Ft together with long
storage time in the higher current transistors that I might otherwise be able to use. Thanks.<p>There might be more aggressive base drive approaches than I'm now using. I've printed out a
number of PDF pages that seem to get into this for
off-line study.<p>I'll continue the effort to come up with an off-board mechanical package for paralleling the small
fast switching transistors and avoid the storage
problem altogether. <p>Thanks again for the good leads.

[Ron H]

Windmiller, why don't you describe what you're trying to do, or better yet, post a schematic, and we'll see if we can come up with a good solution.

[fsdenis]

RonH:<p>Since I last spoke with you, I have succeeded with
the off-board attempt to achieve a fast driver for
an N=20 array of TIP41C and TIP42C power bipolar
paralleled transistors. It should occupy about
1 cubic inch of box space once glued down.<p>However, I am mystified by the notion of storage
time associated with Icesat and Vcesat.<p>On my bench, I see storage time of about 1.8
microseconds in the TIP42C (npn) waveform with
collector open. This suggests to me that storage
time is a phenomenon associated with the base-
emitter junction only and independent of Ic.
Even, that saturation might be independent of Ic.<p>I would be glad to supply a schematic of what I am
up to if I knew how to send one to this forum.<p>If you are willing to stick your toe into this
mess, I would be delighted to hear your thoughts.

[fsdenis]

RonH:<p>Oops. TIP42C is pnp.

[fsdenis]

RonH:<p>It seems I may be able to post schematics on a
web site outside this forum. I'll start trying to
figure out how to do this. Thanks for the
suggestion.<p>I may have a solution for my immediate
question vs fast drivers. Thanks a lot for your
provocative answers.<p>I may be wrong to try to spread the original
question to attempt an understanding of the in's
and out's of storage time here. Maybe I should
think this thing through, make sure I know what
I'm seeing on my bench and if the storage question
won't go away, post it as a new question. And
maybe get it a bit more clear.<p>Many thanks for your help.

[Chris Smith]

Have you tried FETs?

[fsdenis]

Chris Smith:<p>I'm trying to avoid FET's for three reasons:<p>1) My lab is filthy and usually I have sparks
coming off the door knob, lamps, even transistors
due to the dry climate here. I understand that I
must practice static safe handling and storage
methods with FET's or risk blowing them up just
by touching them.<p>2) I know nothing about static safe handling nor
am I equipped for it. <p>3) Many people who experiment in electronics come
from other fields, attracted by the unbeatable
capability electronics has in measurement and
control. I'm one of them. Mainly interested in
extending the control capability to the half-horsepower level at 48 volts and below in the
alternate energy field.<p> My ignorance inclines me to the bipolar transistors as being tougher and more forgiving to one not primarily an electroniker.<p>Thanks for the idea, though. One day, when I know
what I'm doing, I might go for it.

[fsdenis]

A little follow-up from more benchwork:<p>6 each 2n4401 npns may be paralleled by putting
100 ohms in each base lead and then paralleling
the other end of the resistors.<p>6 each 2n4403 pnps may be paralleled by putting
100 ohms in each base lead and then paralleling
the other end of the resistors.<p>Parallel all 6 collector leads of the npns together. And parallel all 6 collector leads of the pnps together.<p>Parallel all 6 emitter leads of both the npns and
pnps together. <p>Put a 2.4 ohm 5 watt resistor in the npn collector bus and another in the pnp collector bus.<p>Use +5v to npn collector bus and -5v to the pnp collector bus.<p>Driven with about 13 ma per base (6*13 = 78 ma total, since either npn or pnp array will be driven at any one time, but not both, the output
(from common emitter junction into a 1 ohm resistor will be about 1.4 amps and very fast.<p>This is one way to have a lot of current with very
low storage time. And true inverted output.

[viveguy]

You can try putting a small capacitance from the base to collector,on the order of 10 - 100pF.This is an old technique that has been replaced by schottky devices.Oddly enough these were known as 'speed-up' caps.The capacitor keeps the base-emitter junction from going into saturation.
You know,there are some pretty hefty bus drivers around.I am surprised one of these won't fit the bill.