Maker Pro
Maker Pro

Transient Voltage Suppressor diodes question

E

Efthimios

Hallo to all,

About ten days i posted a question how to protect a 1 Volt sensor from
a 100 Volt peak for about 0.2 sec/ 20 mAmp.

Some people suggested a Transient Voltage Suppressor. As I was reading
the specs of TVS i realised that brake down voltage is only 10-20%
from the stand off or operating Voltage. Not 100 times more. I
understand "brake down voltage" the highest voltage were voltage will
pass through the diode.

Am I missing something about TVS? Can a TVS protect my sensor from a
100 Volt spike?

If TVS can not protect is Varistors more suitable?


Efthimios
 
E

Efthimios

Not really.


I think the bridge rectifier / two diodes in series idea was the most
practical (Phil Allisons' suggestion IIRC?).

Thanks John. Although i do not understand clearly how two diodes is
seriec can work. I already posted a question in the previous
discussion.

Rgds

Efthimios
 
J

John Devereux

Efthimios said:
Thanks John. Although i do not understand clearly how two diodes is
seriec can work. I already posted a question in the previous
discussion.

Diodes start conducting in the "forward" direction at around 0.6V, so
two of them in series start conducting at around 1.2V.

(The exact voltage depends on the current you put through them, as
well as the temperature and the diode).
 
E

Efthimios

Diodes start conducting in the "forward" direction at around 0.6V, so
two of them in series start conducting at around 1.2V.

(The exact voltage depends on the current you put through them, as
well as the temperature and the diode).

They will start to conduct at 1.2 Volt but how they will stop the 100
Volt pulse flowing in the same direction?

Efthimios
 
J

John Devereux

Efthimios said:
[...]
Diodes start conducting in the "forward" direction at around 0.6V, so
two of them in series start conducting at around 1.2V.

(The exact voltage depends on the current you put through them, as
well as the temperature and the diode).

They will start to conduct at 1.2 Volt but how they will stop the 100
Volt pulse flowing in the same direction?

By diverting the current to ground.

(use fixed point font to view):

Input ------------------------
| |
V |
--- [Sensor]
| |
| |
V |
--- |
| |
0V ---------------------------

If the voltage goes above about 1.2V, the diodes conduct and keep the
voltage below about 1.5V. This assumes that the 100V source is not
able to supply more current than the diodes can stand. You mentioned
20mA, which is quite low, so any normal diode would be OK, 1N4148 or
1N4001 perhaps.
 
E

Efthimios

Efthimios said:
Hallo to all,
About ten days i posted a question how to protect a 1 Volt sensorfrom
a 100 Volt peak for about  0.2 sec/ 20 mAmp.
[...]




Am I missing something about TVS? Can a TVS protect my sensor from a
100 Volt spike?
Not really.
If TVS can not protect is Varistors more suitable?
I think the bridge rectifier / two diodes in series idea was the most
practical (Phil Allisons' suggestion IIRC?).
--
John Devereux
Thanks John. Although i do not understand clearly how two diodes is
seriec can work. I already posted a question in the previous
discussion.
Diodes start conducting in the "forward" direction at around 0.6V, so
two of them in series start conducting at around 1.2V.
(The exact voltage depends on the current you put through them, as
well as the temperature and the diode).
They will start to conduct at 1.2 Volt but how they will stop the 100
Volt pulse flowing in the same direction?

By diverting the current to ground.

(use fixed point font to view):

   Input ------------------------
              |                  |
              V                  |
             ---              [Sensor]
              |                  |
              |                  |
              V                  |
             ---                 |
              |                  |
   0V ---------------------------

If the voltage goes above about 1.2V, the diodes conduct and keep the
voltage below about 1.5V. This assumes that the 100V source is not
able to supply more current than the diodes can stand. You mentioned
20mA, which is quite low, so any normal diode would be OK, 1N4148 or
1N4001 perhaps.

Thanks John you saved my day.
 
R

Rich Grise

About ten days i posted a question how to protect a 1 Volt sensor from a
100 Volt peak for about 0.2 sec/ 20 mAmp.

Some people suggested a Transient Voltage Suppressor. As I was reading the
specs of TVS i realised that brake down voltage is only 10-20% from the
stand off or operating Voltage. Not 100 times more. I understand "brake
down voltage" the highest voltage were voltage will pass through the
diode.

Am I missing something about TVS? Can a TVS protect my sensor from a 100
Volt spike?

If TVS can not protect is Varistors more suitable?

I once worked at a place that made specialized HV supplies (like an SEM);
they originallly used MOVs (Metal Oxide Varistors) and they were an
endless source of problems - MOVs wear out, and they have to send the box
back to get them replaced.

Then the Transzorb press releases came out, and we not only started using
them, but we issued an EO to take out all of the varistors and throw them
away, replacing them with Transzorbs.

How "stiff" is that 100V? Transzorbs are spec'd in joules, I think, so
should drag down the 100V source if it's draggable, But if all it is is
a spike, it should be just fine and dandy. :)

Cheers!
Rich
 
I once worked at a place that made specialized HV supplies (like an SEM);
they originallly used MOVs (Metal Oxide Varistors) and they were an
endless source of problems - MOVs wear out, and they have to send the box
back to get them replaced.

Then the Transzorb press releases came out, and we not only started using
them, but we issued an EO to take out all of the varistors and throw them
away, replacing them with Transzorbs.

How "stiff" is that 100V? Transzorbs are spec'd in joules, I think, so
should drag down the 100V source if it's draggable, But if all it is is
a spike, it should be just fine and dandy. :)

Cheers!
Rich

I had some burn-in failures that were clearly due to electrical
overstress, not the kind of thing you see in burn-in. Transorbs on the
burn-in boards took care of the problem, even though the person in
charge of burn-in was certain there were no spikes on his well
conditioned supplies.

Regarding the OP question, the answer is yes, but I would spec the
transorb a bit higher than 1V since they turn on quickly.
 
J

John Devereux

Rich Grise said:
I once worked at a place that made specialized HV supplies (like an SEM);
they originallly used MOVs (Metal Oxide Varistors) and they were an
endless source of problems - MOVs wear out, and they have to send the box
back to get them replaced.

Then the Transzorb press releases came out, and we not only started using
them, but we issued an EO to take out all of the varistors and throw them
away, replacing them with Transzorbs.

How "stiff" is that 100V? Transzorbs are spec'd in joules, I think, so
should drag down the 100V source if it's draggable, But if all it is is
a spike, it should be just fine and dandy. :)

He wanted to limit the 1V supply to less than 1.5V IIRC. No transorb
will do that. And the "spike" is ~0.2s long. More of a "surge", I'd
say.
 
R

Richard The Dreaded Libertarian

He wanted to limit the 1V supply to less than 1.5V IIRC. No transorb will
do that. And the "spike" is ~0.2s long. More of a "surge", I'd say.

Then cascade a ferrite bead, an intermediate transzorb, maybe 25-40V,
another ferrite bead, and a string of two 1N914's. Maybe another ferrite
bead after them.

Good Luck!
Rich
 
Top