astable multivibrator semiconductors

[John Larkin]

On Fri, 19 Dec 2014 15:39:45 -0800 (PST), mrdarrett@gmail.com wrote:

On Thursday, December 18, 2014 6:29:36 PM UTC-8, John Larkin wrote:

...

With 12 volts or so Vcc, you'll be zenering the bipolar b-e junctions,
which is A Bad Thing. Spice will typically not model b-e breakdown.



--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com


Ok, so I should stick with 6V max Vcc then?

Absolute max Vebo = 6V, from here.
https://www.fairchildsemi.com/datasheets/2N/2N3904.pdf

I'm surprised I didn't barbecue the 2n3904s then...

Thanks!

There can be long-term degradation of beta from zenering the bases,
but I don't have numbers on that. Maybe someone does, or would like to
measure it.

The immediate effect will be to make the timing less predictable.


--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

 

[John Larkin]

On Fri, 19 Dec 2014 12:07:10 -0600, Tim Wescott
<seemywebsite@myfooter.really> wrote:

On Thu, 18 Dec 2014 18:29:27 -0800, John Larkin wrote:

On Wed, 17 Dec 2014 14:37:58 -0800 (PST), mrdarrett@gmail.com wrote:

On Wednesday, December 17, 2014 2:14:30 PM UTC-8, Tim Wescott wrote:
On Wed, 17 Dec 2014 13:47:53 -0800, mrdarrett wrote:

On Wednesday, December 17, 2014 1:40:26 PM UTC-8, mrda...@gmail.com
wrote:
On Wednesday, December 17, 2014 12:06:22 PM UTC-8, Tim Wescott
wrote:
On Wed, 17 Dec 2014 10:40:20 -0800, mrdarrett wrote:

On Wednesday, December 17, 2014 10:15:32 AM UTC-8, Tim Wescott
wrote:
On Wed, 17 Dec 2014 10:14:10 -0800, mrdarrett wrote:

For an astable multivibrator, would 2n7000 mosfets be more
efficient than 2n3904 bjts?

How long is a string?

--
www.wescottdesign.com


0 <= L < oo ?

I built one with 2n3904s and was wondering if I should order
some 2n7000s from Mouser to make it more efficient, is all.

Without any more detail than that, no correct answer can be
given, any more than one can answer "how long is a string?".

Post a circuit, and what you're trying to do with it, and we'll
try to help out -- even if we have to start by asking for more
detail yet.

--

Tim Wescott Wescott Design Services http://www.wescottdesign.com



A basic astable multivibrator:
http://en.wikipedia.org/wiki/Multivibrator#mediaviewer/
File:Transistor_Multivibrator.svg

R1=R4=10k R2=R3=10k C1=C2=0.1uF Q1=Q2=2n3904 or 2n7000 (which is
why I asked)

I'm tapping into the node between R4 and Q2's collector to run the
gate of a MOSFET (irf530) to power a blower motor.

Michael


Sorry, R2=R3=60k

If that circuit ever got into a situation where both transistors were
turned on, they would saturate, both their collectors would be sitting
at low voltages, and the base current delivered by R2 and R3 would be
sufficient to keep them saturated.

Moreover, that circuit depends on the base current of a BJT to work:
with a pair of FETs, the average voltage of the gates will rise to
equal the supply voltage, they'll both end up turned on hard, and the
circuit won't work.

You could come up with more elaborate biasing arrangements -- but why?

A 555 will work at 12V -- why not use a 555 and a 12V linear
regulator? You should get plenty of drive to your MOSFET gate, and
you'll have a circuit that won't fail 0.1% of the time, while leaving
you unable to diagnose problems the other 99.9%.

To answer your original question: the base current of the BJT's is
small compared to the power getting burnt in the resistors; the
reduced power consumption (if any) would be lost in the noise of
component variations from unit to unit.

--

Tim Wescott Wescott Design Services http://www.wescottdesign.com


Ah, ok. So, to summarize, did I understand this correctly:

1) the astable multivibrator circuit with discrete components requires
BJTs to operate properly; the system will not oscillate with FETs

2) there is a 0.1% probability of the system failing to oscillate; use a
555 to avoid this

3) to reduce power consumption, increase R1 and R4

Why... I just get a kick out of using discrete transistors instead of an
IC. That's why.

Thanks!

Michael

With 12 volts or so Vcc, you'll be zenering the bipolar b-e junctions,
which is A Bad Thing. Spice will typically not model b-e breakdown.

It's actually mentioned in the Wikipedia article he cites. The solution
is protection diodes so the B-E junctions never go more than one diode
drop in reverse -- which means even more parts on the thing.

If you mean b-e diodes, that has its own problems.