What is Operating Mechanism of Thyristor?

[Walter Contrata]

Greetings,

Please explain the physical mechanism for thyristor function. Why
does current continue to flow from anode to cathode after current to
the gate has been turned off?

I have found a diagram of the thyristor structure at

http://www.allaboutcircuits.com/vol_3/chpt_2/9.html

It shows an p+/n-/p/n+ stack, with the following ohmic contacts

top p+ layer = anode;
p layer = the gate;
bottom n+ layer = the cathode.

The bottom three layers, n-/p/n+, look like the collector, base, and
emitter of an NPN bipolar transistor, but instead of an n+ collector
ohmic, the thyristor has p+.

According to several sources, the thyristor is normally biased with
the anode positive and cathode negative, like an NPN bipolar
transistor. Like the transistor, negligible current flows from anode
to cathode until a positve current flows into the gate. However,
unlike the NPN, this current does not stop after the gate current
returns to 0. Why?

My apologies for multiple postings.

Best Regards,

 

[CFoley1064]

Subject: What is Operating Mechanism of Thyristor?

From: "Walter Contrata" [email protected]
Date: 4/4/2004 3:21 PM Central Standard Time
Message-id: <[email protected]>

Greetings,

Please explain the physical mechanism for thyristor function. Why
does current continue to flow from anode to cathode after current to
the gate has been turned off?

I have found a diagram of the thyristor structure at

http://www.allaboutcircuits.com/vol_3/chpt_2/9.html

It shows an p+/n-/p/n+ stack, with the following ohmic contacts

top p+ layer = anode;
p layer = the gate;
bottom n+ layer = the cathode.

The bottom three layers, n-/p/n+, look like the collector, base, and
emitter of an NPN bipolar transistor, but instead of an n+ collector
ohmic, the thyristor has p+.

According to several sources, the thyristor is normally biased with
the anode positive and cathode negative, like an NPN bipolar
transistor. Like the transistor, negligible current flows from anode
to cathode until a positve current flows into the gate. However,
unlike the NPN, this current does not stop after the gate current
returns to 0. Why?

My apologies for multiple postings.

Best Regards,

All your questions are answered in Teccor AN-1001, Fundamental Characteristics
of Thyristors

http://www.teccor.com/web/PDF Files/Power/an1001.pdf

Good luck
Chris

 

[John Larkin]

Greetings,

Please explain the physical mechanism for thyristor function. Why
does current continue to flow from anode to cathode after current to
the gate has been turned off?

I have found a diagram of the thyristor structure at

http://www.allaboutcircuits.com/vol_3/chpt_2/9.html

It shows an p+/n-/p/n+ stack, with the following ohmic contacts

top p+ layer = anode;
p layer = the gate;
bottom n+ layer = the cathode.

The bottom three layers, n-/p/n+, look like the collector, base, and
emitter of an NPN bipolar transistor, but instead of an n+ collector
ohmic, the thyristor has p+.

According to several sources, the thyristor is normally biased with
the anode positive and cathode negative, like an NPN bipolar
transistor. Like the transistor, negligible current flows from anode
to cathode until a positve current flows into the gate. However,
unlike the NPN, this current does not stop after the gate current
returns to 0. Why?

My apologies for multiple postings.

Best Regards,

It's just equivalent to a PNP transistor connected regeneratively to
an NPN...


A ------+
|
|
|
e < P
PNP b----------+ < N
c |
| c
G-----+----------b NPN < P
e < N
|
K-------------------+


This has two stable states: both transistors off, or (with current
flowing) both on. If either transistor starts to conduct, it feeds the
other, which feeds the first... If the current drops to the point that
the product of the two betas falls below 1, the whole mess turns off.

The PNPN SCR structure just merges the PNP and NPN transistors by
sharing layers. Often there's a diffused resistor added to each B-E
junction to reduce the feedback gain; this reduces spurious triggering
from noise or high-temperature leakage. "Sensitive gate" SCRs don't
have the resistor on the NPN side.

If you bring out the PNP base as the gate, you get a
negative-triggered SCR. This is the PUT (programmable unijunction)
structure.

Lots of linear bipolar and CMOS IC structures have junction isolation
designs which create unwanted PNPN structures, and have nasty latchup
problems in consequence. The old CD4000A series parts were notorious.
LM35 is still very bad, and Bob Pease won't fix it.


John

 

[Walter Contrata]

Chris,

Thanks for the speedy reply. This reference looks great.

 

[Walter Contrata]

****Omitted***


It's just equivalent to a PNP transistor connected regeneratively to
an NPN...


A ------+
|
|
|
e < P
PNP b----------+ < N
c |
| c
G-----+----------b NPN < P
e < N
|
K-------------------+


This has two stable states: both transistors off, or (with current
flowing) both on. If either transistor starts to conduct, it feeds the
other, which feeds the first... If the current drops to the point that
the product of the two betas falls below 1, the whole mess turns off.

The PNPN SCR structure just merges the PNP and NPN transistors by
sharing layers. Often there's a diffused resistor added to each B-E
junction to reduce the feedback gain; this reduces spurious triggering
from noise or high-temperature leakage. "Sensitive gate" SCRs don't
have the resistor on the NPN side.

If you bring out the PNP base as the gate, you get a
negative-triggered SCR. This is the PUT (programmable unijunction)
structure.

Lots of linear bipolar and CMOS IC structures have junction isolation
designs which create unwanted PNPN structures, and have nasty latchup
problems in consequence. The old CD4000A series parts were notorious.
LM35 is still very bad, and Bob Pease won't fix it.


John

John,

Thanks for clearing that up so well. I wish I had figured it out.

Sorry to hear about Mr. Pease... Should he fix it?