Hockey puck SCRs

K

KW

Guest
I recently bought a hockey-puck SCR, without realising it needs to be
clamped to a heatsink. I'm trying to decide whether to return it or
whether to build a mount.

The SCR in question is an Internation Rectifier 250PA140, specs pasted at
the end of this topic.

First of all, I haven't managed to find any information about how much
force you need to clamp these things with. I'm assuming 2 metal plates
with 3x6mm bolts should be ample.

Secondly, do I need to apply force or displacement? For example, if I need
to apply exactly 1mm of displacement to the plates of the puck, then the
puck contracts by 0.1mm this might be a disaster, unless there is some
kind of springy contact inside that will take up the slack. Perhaps I
should use some kind of springy bar to apply a constant pressure despite
changes in temperature etc...

Thirdly, I presume that too much force will break the die, and too little
will mean the contact has insufficient area for the device to function
properly? How can I know when I've tightened the clamp the right amount?

Finally, I understand there is meant to be a snubber resistor between the
gate and the cathode. But I get very low resistances of < 0.1 ohm and I
wonder whether this implies it is broken?

If I(GT) Max. is 150m and the parallel resistance is < 0.1 ohms then as I
see it the voltage can't rise high enough to cause the device to trigger -
I'm assuming it needs at least a diode drop.

Thanks in advance for any help.

K


V(DRM) Max.(V)Rep.Pk.Off Volt.=1.4k
I(T) Max.(A) On-state Current=250?
@Temp. (?C) (Test Condition)=70
I(TSM) Max. (A)=4.2k
I(GT) Max. (A)=150m
V(GT) Max.(V)=2.5
I(H) Max.(A) Holding Current=500m
I(D) Max. (A) Leakage Current=11m
@Temp. (?C) (Test Condition)=125
V(T) Max. (V)=2.3
@I(T) (A) (Test Condition)=250
dv/dt Min. (V/us)=200
t(q) Typ. (s)=60u
Package=TO-200AB
 
On Wed, 12 Dec 2007 03:15:03 +0000, KW <n8023299@burnmoorncl.ac.uk>
wrote:

I recently bought a hockey-puck SCR, without realising it needs to be
clamped to a heatsink. I'm trying to decide whether to return it or
whether to build a mount.

The SCR in question is an Internation Rectifier 250PA140, specs pasted at
the end of this topic.

First of all, I haven't managed to find any information about how much
force you need to clamp these things with. I'm assuming 2 metal plates
with 3x6mm bolts should be ample.

Secondly, do I need to apply force or displacement? For example, if I need
to apply exactly 1mm of displacement to the plates of the puck, then the
puck contracts by 0.1mm this might be a disaster, unless there is some
kind of springy contact inside that will take up the slack. Perhaps I
should use some kind of springy bar to apply a constant pressure despite
changes in temperature etc...

Thirdly, I presume that too much force will break the die, and too little
will mean the contact has insufficient area for the device to function
properly? How can I know when I've tightened the clamp the right amount?

Finally, I understand there is meant to be a snubber resistor between the
gate and the cathode. But I get very low resistances of < 0.1 ohm and I
wonder whether this implies it is broken?

If I(GT) Max. is 150m and the parallel resistance is < 0.1 ohms then as I
see it the voltage can't rise high enough to cause the device to trigger -
I'm assuming it needs at least a diode drop.

Thanks in advance for any help.

K


V(DRM) Max.(V)Rep.Pk.Off Volt.=1.4k
I(T) Max.(A) On-state Current=250?
@Temp. (?C) (Test Condition)=70
I(TSM) Max. (A)=4.2k
I(GT) Max. (A)=150m
V(GT) Max.(V)=2.5
I(H) Max.(A) Holding Current=500m
I(D) Max. (A) Leakage Current=11m
@Temp. (?C) (Test Condition)=125
V(T) Max. (V)=2.3
@I(T) (A) (Test Condition)=250
dv/dt Min. (V/us)=200
t(q) Typ. (s)=60u
Package=TO-200AB
Click to expand...
I found another device datasheet with TO-200AB. At the end of the
datasheet, it says 700-900 lb, or 320-400 Kg.

http://www.powersemiconductors.com/images/pdfs/phase_control/DT03N%20%20series.pdf

http://tinyurl.com/2c8fau


I don't think you'd be looking at displacement for a ceramic case.

--
John
 
On Wed, 12 Dec 2007 03:15:03 +0000, KW <n8023299@burnmoorncl.ac.uk> wrote:

I recently bought a hockey-puck SCR, without realising it needs to be
clamped to a heatsink. I'm trying to decide whether to return it or
whether to build a mount.

The SCR in question is an Internation Rectifier 250PA140, specs pasted at
the end of this topic.

First of all, I haven't managed to find any information about how much
force you need to clamp these things with. I'm assuming 2 metal plates
with 3x6mm bolts should be ample.

Secondly, do I need to apply force or displacement? For example, if I need
to apply exactly 1mm of displacement to the plates of the puck, then the
puck contracts by 0.1mm this might be a disaster, unless there is some
kind of springy contact inside that will take up the slack. Perhaps I
should use some kind of springy bar to apply a constant pressure despite
changes in temperature etc...

Thirdly, I presume that too much force will break the die, and too little
will mean the contact has insufficient area for the device to function
properly? How can I know when I've tightened the clamp the right amount?

Finally, I understand there is meant to be a snubber resistor between the
gate and the cathode. But I get very low resistances of < 0.1 ohm and I
wonder whether this implies it is broken?

If I(GT) Max. is 150m and the parallel resistance is < 0.1 ohms then as I
see it the voltage can't rise high enough to cause the device to trigger -
I'm assuming it needs at least a diode drop.

Thanks in advance for any help.

K


V(DRM) Max.(V)Rep.Pk.Off Volt.=1.4k
I(T) Max.(A) On-state Current=250?
@Temp. (?C) (Test Condition)=70
I(TSM) Max. (A)=4.2k
I(GT) Max. (A)=150m
V(GT) Max.(V)=2.5
I(H) Max.(A) Holding Current=500m
I(D) Max. (A) Leakage Current=11m
@Temp. (?C) (Test Condition)=125
V(T) Max. (V)=2.3
@I(T) (A) (Test Condition)=250
dv/dt Min. (V/us)=200
t(q) Typ. (s)=60u
Package=TO-200AB
Click to expand...
My old GE manual specifies 700-900lbs force for their
press pak mounted SCRs. I mounted some where I didn't have
to be concerned about dissipating heat (single pulse operation)
and found an online calculator to determine clamping force resulting
from a given torque on a bolt then used a small torque wrench to
tighten it up. I don't remember the numbers, but it took surprisingly
little torque to get 800lbs clamping force. You might try googling
"presspack" to see what you come up with.

Mike
When truth is absent politics will fill the gap.
 
Thanks for the suggestions.

I found that googling 'torque bolt calculator' came up with some useful
widgets. I shall finally have to invest in a torque wrench though!
(probably no bad thing).

K
 
"KW" <n8023299@burnmoor.ncl.ac.uk> wrote in message
news:pine.SOC.4.62.0712130338541.25376@aidan.ncl.ac.uk...
Thanks for the suggestions.

I found that googling 'torque bolt calculator' came up with some useful
widgets. I shall finally have to invest in a torque wrench though!
(probably no bad thing).

K
Click to expand...
Torque wrenches can be obtained pretty cheaply, but you can improvise by
hooking a spring scale on a wrench at, say, one foot from the center, and
40 pounds pull = 40 ft-lb.

Be aware that the pressure exerted based on ft-lb on bolts depends on
lubrication of the threads. Also remember that you need to add the
pressures from two bolts, and it is important to tighten evenly. Some
clamps have a heavy steel bar with a pressure point in the center, and some
have a built-in pressure gauge which relies on the bending of a spring
steel bar clamp.

That 0.1 ohms gate to cathode sounds like it's shorted. Most big puks read
12 to 35 ohms, and generally require a minimum of about 150 to 500 mA gate
current for turn-on, which is about 0.5 to 1 VDC.

The snubber is placed from anode to cathode, and is designed to limit the
dV/dt across the SCR when voltage is suddenly applied, which can trigger
the SCR on. A typical snubber will be about 40 ohms with a 0.1 uF capacitor
rated for AC applications.

Good luck,

Paul
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top