SCR question

[Jason S]

Hi,

Bit of a dumb question here.....

Not having much luck with circuits triggering SCR's.
What I want to do is trigger an SCR, and I know they require a short burst
of current. Triggering them manually with a wire works fine.

Trouble is, how do I get another circuit to trigger it momentarily?
e.g. Let's just say I have a 555 timer IC (another circuit), and I want its
555 output pin to trigger the SCR, even if the 555 timer output changes back
to 'low' again. The SCR must remain on until power is turned off to the
actual SCR, even if later the 555 was switched off completely. This is just
an example.

I thought a non-elec cap and a high value resistor would work (providing a
pulse), but it didn't affect the SCR at all.. not sure if its due to the
type of SCR I'm using or not. If I manually trigger the SCR with a lead, it
works fine, so its not a connection problem.
The SCR I'm using is a C106Y and have also tried a BT169D (from Jaycar)..

This is what I tried, assuming would work:



PWR----||----+-----> To SCR gate

10n | (pulse)

|

_

| |

| | 10M

-

|

|

GND



I know about the electromagnetic induction method, and that a 555 can be
configured to work like an SCR, but isn't there a much quicker better way,
similar to the above?



Thanks in advance.

 

[ehsjr]

Hi,

Bit of a dumb question here.....

Not having much luck with circuits triggering SCR's.
What I want to do is trigger an SCR, and I know they require a short burst
of current. Triggering them manually with a wire works fine.

Trouble is, how do I get another circuit to trigger it momentarily?
e.g. Let's just say I have a 555 timer IC (another circuit), and I want its
555 output pin to trigger the SCR, even if the 555 timer output changes back
to 'low' again. The SCR must remain on until power is turned off to the
actual SCR, even if later the 555 was switched off completely. This is just
an example.

I thought a non-elec cap and a high value resistor would work (providing a
pulse), but it didn't affect the SCR at all.. not sure if its due to the
type of SCR I'm using or not. If I manually trigger the SCR with a lead, it
works fine, so its not a connection problem.
The SCR I'm using is a C106Y and have also tried a BT169D (from Jaycar)..

This is what I tried, assuming would work:



PWR----||----+-----> To SCR gate

10n | (pulse)

|

_

| |

| | 10M

-

|

|

GND



I know about the electromagnetic induction method, and that a 555 can be
configured to work like an SCR, but isn't there a much quicker better way,
similar to the above?



Thanks in advance.

Make sure your load is on the anode side.
Here's diode OR triggering:


+ ----------+ ---
| +---|<--- } | |
[Rload] +---|<--- } whatever --- ---
| +---|<--- }
--- |
\ / | ------
----- | | 555 |
| \ | | |
| ---[10K]---+--|<----|3 |
| | |
| ------
|
Gnd ---------+

Stuff a + square pulse into the gate. There is a turn
on time and a rise time - check the datasheet.
Ed

 

[Jason S]

Hi,

Bit of a dumb question here.....

Not having much luck with circuits triggering SCR's.
What I want to do is trigger an SCR, and I know they require a short burst
of current. Triggering them manually with a wire works fine.

Trouble is, how do I get another circuit to trigger it momentarily?
e.g. Let's just say I have a 555 timer IC (another circuit), and I want
its 555 output pin to trigger the SCR, even if the 555 timer output
changes back to 'low' again. The SCR must remain on until power is turned
off to the actual SCR, even if later the 555 was switched off completely.
This is just an example.

I thought a non-elec cap and a high value resistor would work (providing a
pulse), but it didn't affect the SCR at all.. not sure if its due to the
type of SCR I'm using or not. If I manually trigger the SCR with a lead,
it works fine, so its not a connection problem.
The SCR I'm using is a C106Y and have also tried a BT169D (from Jaycar)..

This is what I tried, assuming would work:



PWR----||----+-----> To SCR gate

10n | (pulse)

|

_

| |

| | 10M

-

|

|

GND



I know about the electromagnetic induction method, and that a 555 can be
configured to work like an SCR, but isn't there a much quicker better way,
similar to the above?



Thanks in advance.

I think it's working now. I hooked up a transistor between the cap/res and
SCR to amplify the short 'burst'. It seems like the burst may have been
extremely weak to trigger the SCR directly previously.
+V
|

|

____ |/

PWR----||----+-----|____|--K NPN
10n | 2K2 |\
| |
_ ------> SCR Gate
| |
| | 10M
-
|
|
GND



Any suggestions or comments regarding this are very welcome, as I'm still
learning! =)



Jason.

 

[Chris]

I think it's working now. I hooked up a transistor between the cap/res and
SCR to amplify the short 'burst'. It seems like the burst may have been
extremely weak to trigger the SCR directly previously.
+V
|

|

____ |/

PWR----||----+-----|____|--K NPN
10n | 2K2 |\
| |
_ ------> SCR Gate
| |
| | 10M
-
|
|
GND



Any suggestions or comments regarding this are very welcome, as I'm still
learning! =)



Jason.

Hi, Jason. A couple of things:

* SCRs have a spec called I(gt), the minimum gate current guaranteed
to turn on the SCR. The typical I(gt) for the C106Y is 50mA, and can
be up to 200mA. A 555 just can't source that much current.

* Your transistor booster circuit will work much better for the C106Y.
You're depending on the current gain of the transistor, though, to
limit the current going into the gate of the SCR. Not only that, but
you haven't taken the leakage current of the transistor into account --
for your other SCR, it could cause false triggering (although that's
not much of a problem with the C106 line). You might want to add a
couple of resistors like this (view in fixed font):

|
| VCC VCC VCC VCC
| + + + +
| | | | |
| .--o---o--. | |
| | | | .-.
| | | | ( X )
| | |.01uF | '-'
| | | 3|| ___ |/ |
| | o--||-|___|-o-| |
| | 555 | || R1 | |> V
| | | .-. | ___ -
| | | R2| | '---|___|-o/|
| | | | | R3 | |
| | | '-' .-.|
| | | | R4| ||
| | | | | ||
| '--o------' === '-'|
| | GND | |
| === o-'
| GND |
| ===
| GND
|
(created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)

If you've got a 12V supply voltage, try 1K for R1, 33K for R2, 33 ohms
1/2w for R3, and 220 ohms for R4.

Also, there are SCRs which are made to be triggered on low gate
current. They're called logic level or sensitive gate SCRs. Your
BT169D is one of those. The NTE equivalent data sheet says gate
trigger current can be less than 1mA. You should be able to drive it
from the 555 directly without the transistor, but you may want to bump
up the cap to 0.1uF or 0.22uF. This gives more time for the SCR to
turn on. It might not be wise to use the transistor follower to crank
so much currrent into this SCR -- you are getting close to the peak
gate current rating.

I hope this has been of help.

Good luck
Chris

 

[ehsjr]

Jason S wrote:




Hi, Jason. A couple of things:

* SCRs have a spec called I(gt), the minimum gate current guaranteed
to turn on the SCR. The typical I(gt) for the C106Y is 50mA, and can
be up to 200mA. A 555 just can't source that much current.

Ummm.... take a look at http://cache.national.com/ds/LM/LM555.pdf
The very first page says "Output can source or sink 200 mA".
You could drive a bucket full of C106 scr's from one 555 and a
bunch of current limiting resistors. The C106 line is senitive
gate.

And where on earth did you get the Igt for the C106Y?
According to the spec sheet:
http://www.datasheets.org.uk/specsheet.php?part=C106Y
or here
http://www.datasheets.in/datasheet-html/C/1/0/C106_ONSemiconductor.pdf.html
click on datasheet

The *maximum* trigger current I(gt) is 200 *micro* amps at 25 C
The *typical* I(gt) is 15 *micro* amps. Your 50 - 200 mA
is way out of line.

Maybe both of those datasheets are wrong? That's why I'd
love to see where you got the specs!

* Your transistor booster circuit will work much better for the C106Y.
You're depending on the current gain of the transistor, though, to
limit the current going into the gate of the SCR.

That point is valid and important. The gate current *must* be
limited to < 200 mA.

The circuit below is not necessary. His problem is dv/dt.
The 10 meg resistor has got to go. He should try his
cap/resistor circuit (no transistor) at 10nf and 10 or 15K
instead of 10 meg. But he hasn't posted a complete circuit,
so who knows what is connected to where.

Ed

 

[Jason S]

Ummm.... take a look at http://cache.national.com/ds/LM/LM555.pdf
The very first page says "Output can source or sink 200 mA".
You could drive a bucket full of C106 scr's from one 555 and a
bunch of current limiting resistors. The C106 line is senitive
gate.

And where on earth did you get the Igt for the C106Y?
According to the spec sheet:
http://www.datasheets.org.uk/specsheet.php?part=C106Y
or here
http://www.datasheets.in/datasheet-html/C/1/0/C106_ONSemiconductor.pdf.html
click on datasheet

The *maximum* trigger current I(gt) is 200 *micro* amps at 25 C
The *typical* I(gt) is 15 *micro* amps. Your 50 - 200 mA
is way out of line.

Maybe both of those datasheets are wrong? That's why I'd
love to see where you got the specs!


That point is valid and important. The gate current *must* be
limited to < 200 mA.

The circuit below is not necessary. His problem is dv/dt.
The 10 meg resistor has got to go. He should try his
cap/resistor circuit (no transistor) at 10nf and 10 or 15K
instead of 10 meg. But he hasn't posted a complete circuit,
so who knows what is connected to where.

Ed

Thanks for your comments guys.

The 555 was mentioned only as an example of triggering the SCR.... instead,
it could be a logic gate output, or maybe just a transistor, or push button.
Sorry to confuse the situation with the 555 example and the lack of
diagrams, but valid points were raised regarding the float, etc. which I'll
keep in mind.

Ed, I've tried taking away the transistor and replacing the 10M with 15K and
the ceramic cap and is working fine! I didn't have a 10nF ('103') like you
said to use, but I only had a cap marked "151"... It's working, that is the
main thing. I will be getting more caps this weekend and I'll try the 10nF
like you said.

I've hooked up a 22K drop-down resistor from the SCR gate pin to ground so
that it doesn't falsely trigger at power-on (thanks for that Chris....
although the 220R you suggested wasn't high enough a value to trigger).
ALL GOOD GUYS.

 

[Jason S]

Thanks for your comments guys.

The 555 was mentioned only as an example of triggering the SCR....
instead, it could be a logic gate output, or maybe just a transistor, or
push button. Sorry to confuse the situation with the 555 example and the
lack of diagrams, but valid points were raised regarding the float, etc.
which I'll keep in mind.

Ed, I've tried taking away the transistor and replacing the 10M with 15K
and the ceramic cap and is working fine! I didn't have a 10nF ('103')
like you said to use, but I only had a cap marked "151"... It's working,
that is the main thing. I will be getting more caps this weekend and I'll
try the 10nF like you said.

I've hooked up a 22K drop-down resistor from the SCR gate pin to ground so
that it doesn't falsely trigger at power-on (thanks for that Chris....
although the 220R you suggested wasn't high enough a value to trigger).
ALL GOOD GUYS.

Oh hang on, I just realised I don't need the 22K drop-down anymore now that
I removed the transistor, so just ignore that paragraph. Cheers.

Jason.

 

[Chris]

Ummm.... take a look at http://cache.national.com/ds/LM/LM555.pdf
The very first page says "Output can source or sink 200 mA".
You could drive a bucket full of C106 scr's from one 555 and a
bunch of current limiting resistors. The C106 line is senitive
gate.

And where on earth did you get the Igt for the C106Y?
According to the spec sheet:
http://www.datasheets.org.uk/specsheet.php?part=C106Y
or here
http://www.datasheets.in/datasheet-html/C/1/0/C106_ONSemiconductor.pdf.html
click on datasheet

The *maximum* trigger current I(gt) is 200 *micro* amps at 25 C
The *typical* I(gt) is 15 *micro* amps. Your 50 - 200 mA
is way out of line.

Maybe both of those datasheets are wrong? That's why I'd
love to see where you got the specs!


That point is valid and important. The gate current *must* be
limited to < 200 mA.

The circuit below is not necessary. His problem is dv/dt.
The 10 meg resistor has got to go. He should try his
cap/resistor circuit (no transistor) at 10nf and 10 or 15K
instead of 10 meg. But he hasn't posted a complete circuit,
so who knows what is connected to where.

Ed

Hi, Ed. Thanks for catching that. Since the C106 is no more, I went
to the NTE website, crossed to NTE5452, then misread their data sheet,
which does say Igt is 50uA typ. / 200uA max. Thinking about those old
clunker SCRs, I misread uA as mA.

ON semi also has this data sheet for the C106D:

http://www.onsemi.com/pub/Collateral/C106-D.PDF

The rest of your observations are also correct. I just missed that
one, which is odd, because I worked with the part in days past. Thanks
again for the spot.

Cheers
Chrs

 

[ehsjr]

Hi, Ed. Thanks for catching that. Since the C106 is no more, I went
to the NTE website, crossed to NTE5452, then misread their data sheet,
which does say Igt is 50uA typ. / 200uA max. Thinking about those old
clunker SCRs, I misread uA as mA.

ON semi also has this data sheet for the C106D:

http://www.onsemi.com/pub/Collateral/C106-D.PDF

The rest of your observations are also correct. I just missed that
one, which is odd, because I worked with the part in days past. Thanks
again for the spot.

Cheers
Chrs

The link you posted to Onsemi is helpful! I had to
"back into" getting there when I did my "fumble-fingered"
searching. And my result is a link that is convoluted.
Yours is straightforward. Nice.

Ed

 

[ehsjr]

Ed, I've tried taking away the transistor and replacing the 10M with 15K and
the ceramic cap and is working fine! I didn't have a 10nF ('103') like you
said to use, but I only had a cap marked "151"... It's working, that is the
main thing. I will be getting more caps this weekend and I'll try the 10nF
like you said.

Hi Jason,

I really didn't think about the value of the cap - I
just thought that was what you had on hand based on
your diagram. The issue is dv/dt, or from another
perspective, tc. dv/dt refers to the speed at which
the voltage level at the gate of the scr changes. tc
is the time constant of the R/C connected to the gate.

You can use a "diode or gate" to experiment with multiple
cap/resistor combinations, logic drives, switches,
whatever to turn the scr on. Chris made the important
point about protecting the scr from too much gate
current - be sure to use a series resistor to limit
gate current.

Ed

 

[Jason S]

Oh hang on, I just realised I don't need the 22K drop-down anymore now
that I removed the transistor, so just ignore that paragraph. Cheers.

Jason.

+12V

|

(LOAD)

|

___

\ /

____ -----

TRG >--||---o--|____|--o------/| C106Y

| R2 | |

R1 | |C2 |

.-. --- V

| | --- GND

| | |

--- |

| |

V V

GND GND



C1 = 10n

C2 = 2n2

R1 = 15K

R2 = 2K2



C2 prevents false triggering at power-on.

IT'S WORKING!

 

[Jason S]

Hi Jason,

I really didn't think about the value of the cap - I
just thought that was what you had on hand based on
your diagram. The issue is dv/dt, or from another
perspective, tc. dv/dt refers to the speed at which
the voltage level at the gate of the scr changes. tc
is the time constant of the R/C connected to the gate.

You can use a "diode or gate" to experiment with multiple
cap/resistor combinations, logic drives, switches,
whatever to turn the scr on. Chris made the important
point about protecting the scr from too much gate
current - be sure to use a series resistor to limit
gate current.

Ed

yup, noted.
Thanks again for your help Ed and Chris.

Jason.