Topics

Low Beta transistors in Tek plug in


Tim Phillips
 

from Tim P (UK)
Sorry if this is more 'Electronics-101' than 'TekScopes', but I'm curious.
I sometimes find when checking transistors (on a Peak DCA55) that the Beta
is outside the range specified in the literature. Towers International is
popular in UK.
F'rinstance I'm looking at a 1S2 with a non-working pulser and a 2N3904 has
a Beta of 65 as against Towers 'minimum 100' I also see instances of
common-or-garden transistors with published Beta of -say- 200 and measured
at -say- 450.
Does a low or high + / - 50% mean an aging transistor? I know Tek had good
reasons for what they did, but these examples were not 'selected' or
TekSpec.
(Don't worry, I'm not about to shotgun the transistors and THEN call for
help !!)
many thanks
Tim


Chuck Harris
 

One thing you need to know about the "literature" on
transistor values is that beta's are measured at very
specific points on the transistor's operating curves.

If you are using a little hand held transistor tester,
it is highly unlikely that you are measuring at the same
operating point as was used in the literature.

A full blown curve tracer, is necessary to duplicate
the manufacturer's literature.

Critical applications of transistors require that you
use the full curves, and draw in load lines that represent
your operating conditions.

The purpose behind the little transistor testers is
two-fold, 1) to get your money, and 2) to test if
the transistor actually has any gain. Some gain usually
indicates a working transistor.

When you use a DVM to do a junction test on a transistor,
all you can tell for sure is that you found junctions that
do, or don't, behave like diodes.

Junction tests tell if the transistor is shorted, or open,
but not if it is good.

-Chuck Harris

Tim Phillips wrote:

from Tim P (UK)
Sorry if this is more 'Electronics-101' than 'TekScopes', but I'm curious.
I sometimes find when checking transistors (on a Peak DCA55) that the Beta
is outside the range specified in the literature. Towers International is
popular in UK.
F'rinstance I'm looking at a 1S2 with a non-working pulser and a 2N3904 has
a Beta of 65 as against Towers 'minimum 100' I also see instances of
common-or-garden transistors with published Beta of -say- 200 and measured
at -say- 450.
Does a low or high + / - 50% mean an aging transistor? I know Tek had good
reasons for what they did, but these examples were not 'selected' or
TekSpec.
(Don't worry, I'm not about to shotgun the transistors and THEN call for
help !!)
many thanks
Tim




Eric
 

Chuck, do you know good resources to help me understand load lines? I have
heard it come up a few times but feel I need to do a deeper dive in to the
subject. I dont mind the reading and I like getting in to the weeds. So
that wont be an issue.

Eric

On Tue, Jun 9, 2020, 11:12 AM Chuck Harris <cfharris@...> wrote:

One thing you need to know about the "literature" on
transistor values is that beta's are measured at very
specific points on the transistor's operating curves.

If you are using a little hand held transistor tester,
it is highly unlikely that you are measuring at the same
operating point as was used in the literature.

A full blown curve tracer, is necessary to duplicate
the manufacturer's literature.

Critical applications of transistors require that you
use the full curves, and draw in load lines that represent
your operating conditions.

The purpose behind the little transistor testers is
two-fold, 1) to get your money, and 2) to test if
the transistor actually has any gain. Some gain usually
indicates a working transistor.

When you use a DVM to do a junction test on a transistor,
all you can tell for sure is that you found junctions that
do, or don't, behave like diodes.

Junction tests tell if the transistor is shorted, or open,
but not if it is good.

-Chuck Harris

Tim Phillips wrote:
from Tim P (UK)
Sorry if this is more 'Electronics-101' than 'TekScopes', but I'm
curious.
I sometimes find when checking transistors (on a Peak DCA55) that the
Beta
is outside the range specified in the literature. Towers International is
popular in UK.
F'rinstance I'm looking at a 1S2 with a non-working pulser and a 2N3904
has
a Beta of 65 as against Towers 'minimum 100' I also see instances of
common-or-garden transistors with published Beta of -say- 200 and
measured
at -say- 450.
Does a low or high + / - 50% mean an aging transistor? I know Tek had
good
reasons for what they did, but these examples were not 'selected' or
TekSpec.
(Don't worry, I'm not about to shotgun the transistors and THEN call for
help !!)
many thanks
Tim






 

On Tue, Jun 9, 2020 at 06:14 PM, Eric wrote:


Chuck, do you know good resources to help me understand load lines?
Why not just start here:
https://en.wikipedia.org/wiki/Load_line_(electronics)

Raymond


 

On Tue, Jun 9, 2020 at 06:22 PM, Raymond Domp Frank wrote:


Why not just start here:
https://en.wikipedia.org/wiki/Load_line_(electronics )
Because it doesn't work...
Try this:
https://en.wikipedia.org/wiki/Load_line_(electronics)

Raymond


Dale H. Cook
 

On 6/9/2020 11:12 AM, Chuck Harris wrote:

The purpose behind the little transistor testers is
two-fold, 1) to get your money, and 2) to test if
the transistor actually has any gain. Some gain usually
indicates a working transistor.
I agree with Chuck. I have a transistor tester built into the veteran Beckman DMM that I use in the field, and I use it for the same purpose that I use a tube tester - to weed out bad parts.
--
Dale H. Cook, Radio Contract Engineer, Roanoke/Lynchburg, VA
https://plymouthcolony.net/starcityeng/index.html


 

On Tue, Jun 9, 2020 at 06:23 PM, Raymond Domp Frank wrote:




Why not just start here:
https://en.wikipedia.org/wiki/Load_line_(electronics )
Because it doesn't work...
Try this:
https://en.wikipedia.org/wiki/Load_line_(electronics )
It still won't, so Google

"Load line (electronics)"

without the quotes.

Raymond


 

On Tue, Jun 9, 2020 at 06:26 PM, Raymond Domp Frank wrote:


Try this:
https://en.wikipedia.org/wiki/Load_line_(electronics )
Copying and pasting the link "https://en.wikipedia.org/wiki/Load_line_(electronics)" results in a space before the closing ")".
Anyway, googling it gets a useful Wikipedia hit.

Raymond


Albert Otten
 

F'rinstance I'm looking at a 1S2 with a non-working pulser and a 2N3904 has
a Beta of 65 as against Towers 'minimum 100'
Hi Tim,

To illustrate Chuck's point.
That requirement is at Ic = 10 mA (and at Vce = 1 V in the datasheets) and is for the DC current gain. Being curious I tested 5 genuine Motorola 2N3904s at my 576. The gains were 130-170.
The requirements at smaller or larger Ic are weaker. For 151-0190-00 Tek requires >70 at 1 mA & 1 V.

Albert