Reverse Polarity and Over Voltage Protection Design

From DIY contraptions to sophisticated FreeEMS-specific designs! Plus general hardware development!
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Reverse Polarity and Over Voltage Protection Design

Post by Fred »

Over Voltage Protection

I was having a play with a spreadsheet earlier and using a zener with series resistor will result in large power dissipation with rather small over voltage conditions. Granted the device shouldn't see more than 15 at all ever, but in the event that it does, our scheme needs to behave appropriately. The only way to ensure a zener protected scheme behaves appropriately is to use a marginal sized fuse. Given we will be drawing around 500mA at some point in the future just to keep the CPU going the fuse needs to be 1 Amp at least to provide reliable service. Furthermore, it will need to be able to handle inrush current at turn on time (for the sensor and reference channel at least) too. What this means is that if we fuse for reliable long term operation then a marginal continuous over voltage will result in huge heat output and component death before the fuse blows. Not good. Additionally, the inline resistors will have to be 5w or larger and having 2 or three of those takes up some space.

I'm not sure if we need over voltage protection on the dirty 12v feed at all, but we certainly want it on the two regulated channels in some form.

Reverse Polarity Protection

I was also thinking about the inline diode and voltage drop through that and crow bar circuits etc and came across some good articles.

I'll post something more about both of these tomorrow. I just wanted to put the discussion out there in advance so people could put their thoughts forward.

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
jharvey
1N4001 - Signed up
Posts: 1607
Joined: Tue Jun 10, 2008 5:17 pm

Re: Reverse Polarity and Over Voltage Protection Design

Post by jharvey »

OV on the input reg's? Aren't most 7805's good for about 60V? Perhaps 5v LDO doesn't mean 7805.

Sound to me like the double diode is the way to go for protecting the inputs and outputs.
User avatar
Delta
LQFP112 - Up with the play
Posts: 111
Joined: Fri Jul 25, 2008 8:04 pm
Location: Perth, WA, Australia

Re: Reverse Polarity and Over Voltage Protection Design

Post by Delta »

my preferred method for voltage protection

Image

Due to the large current drawn by the darlington pair, putting a fuse between the voltage source and input diode means on fault condition the fuse blows. The input diode will have to be a TO220 type, the zener will be fine at 1W. The 0.01Ohm resistor input to darlington pair should be 5W but can probably just get away with 1W (perhaps).
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Reverse Polarity and Over Voltage Protection Design

Post by Fred »

Delta wrote:my preferred method for voltage protection

http://i35.tinypic.com/1494ca1.jpg

Due to the large current drawn by the darlington pair, putting a fuse between the voltage source and input diode means on fault condition the fuse blows. The input diode will have to be a TO220 type, the zener will be fine at 1W. The 0.01Ohm resistor input to darlington pair should be 5W but can probably just get away with 1W (perhaps).
Having thought about this some more I'm happy with this design with a couple of changes.
  • To provide some isolation and cleanliness of power there should be a 1 Ohm resistor inside the protection inline right where it says vregin.
  • The input diode should be replaced with a FET biased to act like a zero drop diode. Could a VNP20N07 function like that?
  • The regulator should have a diode across it to prevent reverse voltage across it.I realise many have them internally, but some don't.
I think if we make those changes it should be robust. It's basically a crow bar circuit so we need to set the voltage limit as high as possible without damaging components behind the protection. Perhaps 20 Volts would be a good figure? That way false trips and fuse blows are minimised and isolated to cases with blown alternator regulators.

How high can the voltage go before the BJT exceeds its peak rating of 20 Amps?

This design or a variant of it is fine for the two regulator circuits, but what about the flyback supply? Does your flyback/snubber/etc mind if the voltage is reversed? If so, how do we protect that? Or does it not need protection?

Thanks for the effort you've put in so far btw, good work!

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
L98TPI
TO220 - Visibile
Posts: 11
Joined: Thu Aug 14, 2008 4:59 am

Re: Reverse Polarity and Over Voltage Protection Design

Post by L98TPI »

A crowbar seems a bit over kill to me, more parts, more space, more cost. Can we not depend on the input voltage rating of the regulators not to pass a surge on? A zener diode to blow the input fuse on over the regulator rating V and reversed polarity seems enough for me. <KISS>

ME
Any tool can be the right tool...
User avatar
jharvey
1N4001 - Signed up
Posts: 1607
Joined: Tue Jun 10, 2008 5:17 pm

Re: Reverse Polarity and Over Voltage Protection Design

Post by jharvey »

I can't see the above pics at the moment, (web filter issue) but a crow bar might not be as nice as we would like. Many crowbars use SCR's. These can be tripped by a small voltage spike. So lets say a little RF makes it through, we pop our fuse, then we start looking for a short circuit, when we should be looking for noisy plug wires. Not exactly intitutive is it?

If we use an SCR as a crowbar, we want to pay specail attention to filtering the input trigger. We want a high pass filter that is set high enough to prevent small RF glitches from triggering it, but low enought that it will trigger before significant damage is done.

I don't see a real need for OV protection. If we have a spike on the input over 60V, you've done something wrong. If you have a spike on the regulated side, you've also done something wrong. Quite rare for these conditions to happen.

Perhaps this is better for 2.0?
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Reverse Polarity and Over Voltage Protection Design

Post by Fred »

Crow bar is the wrong term, it only draws current while the over voltage condition continues. It is a pair of high current transistors that just limit voltage by clamping down with a series resistor of 0.1 Ohms.

If the over voltage was not continuous (short spike/RF/etc) it would do nothing other than keep the power smooth.

The trouble with using "just a zener" is that they are expensive to obtain with suitable current ratings to achieve that. They aren't all that easy to find. Using a transistor controlled by a zener to perform the same task is just a nice way of beefing it up without using exotic high current/power zeners etc.

For all of these threads we need to define what the goal is. If we really KISS it then the board is effectively a temporary board and the chances of people wanting to fork out on a throw away solution are slim IMO. If we offer some basic niceties like power protection and ignition control and low Z control and an intelligent system for IO polarity that isn't easy to change in software/be reset by upgrade/etc then we are making it a long term solution with expandability and a lot more appeal.

None of this stuff matters if it's a temporary thing, but you seem to be approaching it in a non temporary way, and if so then all aspects should be done properly, whatever that is.

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
Brian
QFP80 - Contributor
Posts: 42
Joined: Sat Jul 26, 2008 1:08 pm

Re: Reverse Polarity and Over Voltage Protection Design

Post by Brian »

I guess the regulator needs to be chosen first, and then decide what type of protection is required. The LM2940T has a lot of inbuilt protection, and is guaranteed to work within spec up to 26V, and will apparently survive up to 60V for <100mS. I have tested the reverse-polarity protection with an 8V supply, and the regulator and the micro it was feeding both survived and are still in operation. There may be other regulators out there that are better, I haven't searched for a long time. Maybe one could be tested to see what the real-world limits are, e.g. how does it handle a short on the output, what happens with a prolonged 35V input etc.

From the National Semiconductor datasheet:

Designed also for vehicular applications, the LM2940/LM2940C and all regulated circuitry are protected from reverse battery installations or 2-battery jumps. During line transients, such as load dump when the input voltage can momentarily exceed the specified maximum operating voltage, the regulator will automatically shut down to protect both the internal circuits and the load. The LM2940/LM2940C cannot be harmed by temporary mirror-image insertion. Familiar regulator features such as short circuit and thermal overload protection are also provided.

Brian.
User avatar
Delta
LQFP112 - Up with the play
Posts: 111
Joined: Fri Jul 25, 2008 8:04 pm
Location: Perth, WA, Australia

Re: Reverse Polarity and Over Voltage Protection Design

Post by Delta »

Brain is right here, TL750M05CKCSE3 is a part available from digikey at 1.14 per piece singly TO-220. It already copes with reverse voltage upto 60V transient 26V continuous 750mA output etc etc. I'll have a word to a few people here and see if I can get it at a reasonable price in Aus.

If people find its too damn hard to get parts like this - then we'll have to think more about my circuit. But for now if we CAN get a good piece it uses up less space and is designed to do the exact job we are looking at.
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Reverse Polarity and Over Voltage Protection Design

Post by Fred »

That is all very well and good for the regulator, but if we are to have some caps on board for filtering, it is not well and good for them. Tantalum are very sensitive to even short term reverse voltage and can go with an almighty crack. Likewise electro, though it may handle a touch in reverse, hold it there for more than a split second and bang you have fluid and paper everywhere. For the over voltage side of it, just using 50 or 60V caps will solve it, but for reverse hookup something is required if they are not to explode. Thoughts?

I can vouch for the over temp protection, and over current protection Image

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
Post Reply