# Voltage limiting device

T

#### The little lost angel

Jan 1, 1970
0
I do lotsa tinkering lately with my PC and PSU, so getting worried
about what will happen if it blew up pPP

So I thought of building a sort of safety device for my PC that will
limit the voltage supplied if some kind of surge occurred when the PSU
fails. This is because there are horror stories of motherboard, hard
drives and other components dying when the PSU exploded (or something
to that effect).

Would it be possible to build such a widget that allows say a max of
13V to pass through. Anything higher is either clipped immediately or
the device trips fast enough to prevent the harmful voltage from going
through? If so, what kind of components and information should I be
looking for?

Thank you!!!

--
L.Angel: I'm looking for web design work.
If you need basic to med complexity webpages at affordable rates, email me
Standard HTML, SHTML, MySQL + PHP or ASP, Javascript.
If you really want, FrontPage & DreamWeaver too.
But keep in mind you pay extra bandwidth for their bloated code

C

#### Costas Vlachos

Jan 1, 1970
0
The little lost angel said:
I do lotsa tinkering lately with my PC and PSU, so getting worried
about what will happen if it blew up pPP

So I thought of building a sort of safety device for my PC that will
limit the voltage supplied if some kind of surge occurred when the PSU
fails. This is because there are horror stories of motherboard, hard
drives and other components dying when the PSU exploded (or something
to that effect).

Would it be possible to build such a widget that allows say a max of
13V to pass through. Anything higher is either clipped immediately or
the device trips fast enough to prevent the harmful voltage from going
through? If so, what kind of components and information should I be
looking for?

Thank you!!!

Look for "crowbar overvoltage protection" in Google. Crowbar protection is a
popular method which uses a thyristor whose gate is controlled by a Zener
diode. When an overvoltage occurs, the thyristor is turned on and
effectively shorts the PSU's output (which in turn blows a fuse).

I don't know for sure, but I suspect computer PSUs already have overvoltage
protection circuits internally, if so there's no need to worry about it.

cheers,
Costas

T

#### The little lost angel

Jan 1, 1970
0
Look for "crowbar overvoltage protection" in Google. Crowbar protection is a
popular method which uses a thyristor whose gate is controlled by a Zener
diode. When an overvoltage occurs, the thyristor is turned on and
effectively shorts the PSU's output (which in turn blows a fuse).

I don't know for sure, but I suspect computer PSUs already have overvoltage
protection circuits internally, if so there's no need to worry about it.

They usually do, kinda latch (some kind of technical method I don't
know?) the PSU to off position until AC is removed. But the problem
I'm thinking of is when the PSU fails, kinda like the
circuitry/caps/IC blows and sends a say 20V surge down the line to
stuff expecting a max of say 14V.

So I was wondering if it was possible to fix something between the PSU
and the devices to block such stuff. It would be quite useful for
friends who insist on using cheapo PSU, if I can make this thing for
say $5 pPP -- L.Angel: I'm looking for web design work. If you need basic to med complexity webpages at affordable rates, email me Standard HTML, SHTML, MySQL + PHP or ASP, Javascript. If you really want, FrontPage & DreamWeaver too. But keep in mind you pay extra bandwidth for their bloated code W #### w_tom Jan 1, 1970 0 Cheaper to buy the$80 (retail list) power supply to fix the
missing overvoltage protection circuit and many other missing
circuits in that cheapo power supply. You don't save money
buying a supply based only upon price. Minimally acceptable
supplies contain the overvoltage protection and other
essential functions because they also provide a long list of
numerical specs.

Overvoltage crowbars were defacto standard even 30 years
ago. And yet some supplies today do not have it? Yes. Today
we have an MBA education where only price is important;
technical facts be damned. Fix overvoltage protection - and
still have many other problems with that cheapo supply.

For example, overvoltage protection could kick in - and the
supply then burns up. No acceptable supply is damaged even
when all outputs are short circuited together! Intel specs
even say how big the shorting wire must be. Fix the problem -
not the symptom. Shoot the MBA; then get a real power
supply. Build the overvoltage crowbar only to learn about
them.

T

#### The little lost angel

Jan 1, 1970
0
For example, overvoltage protection could kick in - and the
supply then burns up. No acceptable supply is damaged even
when all outputs are short circuited together! Intel specs
even say how big the shorting wire must be. Fix the problem -
not the symptom. Shoot the MBA; then get a real power
supply. Build the overvoltage crowbar only to learn about
them.

Well, for another thing, I'm going to be fooling around with a couple
of PSU for fun and likely to blow up a couple in the process of
stressing them. So the device would be nice to protect my own systems
when I do that, just thought that it would also be nice to be able to
make the widget for friends who ignore advice to use better PSU.

--
L.Angel: I'm looking for web design work.
If you need basic to med complexity webpages at affordable rates, email me
Standard HTML, SHTML, MySQL + PHP or ASP, Javascript.
If you really want, FrontPage & DreamWeaver too.
But keep in mind you pay extra bandwidth for their bloated code

C

#### Costas Vlachos

Jan 1, 1970
0
The little lost angel said:
They usually do, kinda latch (some kind of technical method I don't
know?) the PSU to off position until AC is removed. But the problem
I'm thinking of is when the PSU fails, kinda like the
circuitry/caps/IC blows and sends a say 20V surge down the line to
stuff expecting a max of say 14V.

So I was wondering if it was possible to fix something between the PSU
and the devices to block such stuff. It would be quite useful for
friends who insist on using cheapo PSU, if I can make this thing for
say $5 pPP Yes, there are ways to limit the voltages, but I don't think it will cost as little as$5. Don't forget, there are 4+ rails to protect. Most of all, it's
a messy job... If you look at the PSU case, there is a whole bunch of cables
coming out of it, some going to the motherboard, others going to the
peripherals, etc. To do it right you'd need to open the PSU casing and add
the circuit in there. But doing so could cause much more harm than good if
you're not careful. Also, if your protection circuit is a little too
sensitive and gets triggered by a voltage spike or noise, the whole PC could
shut down on you in the middle of a job, meaning lost data, etc. I'd go for
a high-quality PSU. I've never had a problem even with cheap PSUs running

Thinking about it, I suppose you could add some sort of passive protection
to the rails by adding transient voltage suppressors in parallel with each
rail, rated just above the rails' voltages. These are basically high-power
Zener diodes that only conduct if the voltage exceeds a given threshold.
This solution could cost less that \$5, but again, I don't think there is a
real need for it.

Don't worry too much about it! Backing up your data regularly is a far safer
method than extra overvoltage protection.

Regards,
Costas
_________________________________________________
Costas Vlachos Email: [email protected]

W

#### w_tom

Jan 1, 1970
0
The overvoltage crowbar as described previously costs far
less than high power zener diodes AND has been industry
standard for over 30 years.

No reasons to think about it. Any acceptable ATX power
supply already has this protection circuit. Those with basic
power supply knowledge knows this circuit to be so important
that even Intel specs demand it:
The overvoltage sense circuitry and reference shall reside in
packages that are separate and distinct from the regulator
control circuitry and reference. No single point fault shall
be able to cause a sustained overvoltage condition on any or
all outputs. The supply shall provide latch-mode
overvoltage protection as defined below.

Suggest you seek a good source on power supply design to
learn the simple, less expensive, and more powerful
overvoltage crowbar circuit - an industry standard circuit
that uses 'low' power (much less expensive) zener diodes.

W

#### w_tom

Jan 1, 1970
0
I ever so strongly, with emphasis, encourage you to build
and test these circuits. Don't be afraid to make smoke. Part
blowing up supplies. Only defective supplies will be
damaged. Then continue on to find out WHY power supply was
defective (with a warning about 300+ volts stored on some
electrolytic capacitors inside that power supply - which is
why power supply capacitors are shorted out before applying
hands). There is great power in electrically dirty hands.
Dirty hands - something that too many 'newsgroup experts about
power supplies' don't bother to do.

C

#### Costas Vlachos

Jan 1, 1970
0
w_tom said:
The overvoltage crowbar as described previously costs far
less than high power zener diodes AND has been industry
standard for over 30 years.

No reasons to think about it. Any acceptable ATX power
supply already has this protection circuit. Those with basic
power supply knowledge knows this circuit to be so important
that even Intel specs demand it:

Suggest you seek a good source on power supply design to
learn the simple, less expensive, and more powerful
overvoltage crowbar circuit - an industry standard circuit
that uses 'low' power (much less expensive) zener diodes.

suggested it to the OP. The point is that there are 4 rails to protect, each
requiring an SCR and Zener. Plus it's an unnecessary job as most PSUs

cheers,
Costas

W

#### w_tom

Jan 1, 1970
0
Sorry. I did not understand that was your point.

T

#### The little lost angel

Jan 1, 1970
0
I ever so strongly, with emphasis, encourage you to build
and test these circuits. Don't be afraid to make smoke. Part
blowing up supplies. Only defective supplies will be
damaged.

I'm not worried about blowing up the supplies, I'm more worried about
blowing up the system connected to the supplies. It's easy enough to
replace a cheapo supply if I blow it up during a stress load, but NOT
the computer system!

--
L.Angel: I'm looking for web design work.
If you need basic to med complexity webpages at affordable rates, email me
Standard HTML, SHTML, MySQL + PHP or ASP, Javascript.
If you really want, FrontPage & DreamWeaver too.
But keep in mind you pay extra bandwidth for their bloated code

W

#### w_tom

Jan 1, 1970
0
An overvoltage protector circuit will not damage any other
device as long as (and this is important), ground integrity to
the OVP is maintained. In a multivoltage system, ground is
essential to all peripheral safety. Importance of ground is
why better busses design peripheral interface connectors with
grounds on each end of card; so that ground will always be
first to make and last to break.

OVP will not cause any power supply voltage to exceed
rating. And low voltage does not damage electronics; just
creates strange intermittents.

T

#### The little lost angel

Jan 1, 1970
0
An overvoltage protector circuit will not damage any other
device as long as (and this is important), ground integrity to
the OVP is maintained. In a multivoltage system, ground is
essential to all peripheral safety. Importance of ground is
why better busses design peripheral interface connectors with
grounds on each end of card; so that ground will always be
first to make and last to break.

OVP will not cause any power supply voltage to exceed
rating. And low voltage does not damage electronics; just
creates strange intermittents.

Erm, I think you're misunderstanding me. I don't mean to say the OVP
will cause any damage. I want an OVP device to prevent damage.

The damage likely to come from an overloaded supply blowing up. There
have been more than a few reports of supplies taking out motherboards
and harddisks when they die.

--
L.Angel: I'm looking for web design work.
If you need basic to med complexity webpages at affordable rates, email me
Standard HTML, SHTML, MySQL + PHP or ASP, Javascript.
If you really want, FrontPage & DreamWeaver too.
But keep in mind you pay extra bandwidth for their bloated code

B

#### Bill Vajk

Jan 1, 1970
0
The said:
Erm, I think you're misunderstanding me. I don't mean to say the OVP
will cause any damage. I want an OVP device to prevent damage.
The damage likely to come from an overloaded supply blowing up. There
have been more than a few reports of supplies taking out motherboards
and harddisks when they die.

Massive zener diodes.

W

#### w_tom

Jan 1, 1970
0
OVP will limit voltage IF properly sized. Size is why high
power zeners are not used - as was true 30 years ago. Current
is why silicon controlled rectifiers (thyristors) are used.
Each SCR sized sufficient to shunt maximum current for each
supply voltage.

A supply overloaded by an overvoltage crowbar shunt must not
be damaged; as required by design specs. But then, even if
supply is damaged, the OVP continues to shunt; guarantees no
damage to computer.

BTW, power supply does not explode. Internal components
simply fail - most often without external, visual indication.
This completely transparent to the load because the OVP
exists.

If in doubt, test with power supply disconnected from
computer. Power supplies running without connection to
motherboard easily accomplished.

Replies
11
Views
787
Replies
3
Views
1K
Replies
38
Views
3K
W
Replies
7
Views
1K
Jasen Betts
J
B
Replies
2
Views
1K
Barry
B