auto voltage protection

Go To Last Post
12 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi,

 

I'm still working on my battery monitor ( hey , it functions!), and I want to clarify a small point.

 

I'm sure that I've read on the Freaks forum, that schottky diodes are good for protection against the many nasties of the automotive voltage environment.

 

But, when I go to Digikey et al, I keep getting to TVS diodes, which are NOT the same thing. They talk about zener and avalanche devices, and I don't know which would be better.

 

For Vcc, I'm planning to use an automotive grade regulator, and maybe some diode protection. 

 

 

For the voltage to be measured, I thought that I'd hook it up as shown below.

 

 

 

This is not  what's shown in the app notes -- they often show a single device from plus to ground.

 

So - 2 questions

1. Is a TVS the right device ( and also single or bi directional)?

2. Is my 2 diode circuit a reasonable idea?

 

Thanks,

 

hj

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The main benefit of a Schottky diode is that when it conducts it drops less voltage than a 'normal' diode.

Whether that is helpful depends on the circuit and for your input-protection it does not matter very much.

 

A TVS is essentially a very-good zener, and they are designed to quickly clamp high-energy spikes.

Unfortunately, the lowest available breakdown-voltage tends to be around 6 volts, therefore you will need to put it 'closer' to the Vin.

For an automotive application I would definitely use bi-uni-directional TVS's. (Don't forget to put one across your main-supply)

 

You schematic is wrong., the diode with the cathode to ground will shunt your Vin.

 

I would do something like the following ;

'

 

 

EDITED to use a uni-directional TVS.

Last Edited: Sun. Oct 22, 2017 - 04:22 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Why bidirectional? Surely you don't want the input going negative?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If the resistances are low enough, a unidirectional TVS might require a higher wattage R1 to handle a reversed battery fault. I was just double-checking that issue for a high-side driver with an internal (unidirectional) diode.

- John

Last Edited: Sun. Oct 22, 2017 - 01:33 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I've used a very similar setup to monitor a vehicle's power supply before.

This doesn't use TVS clamps, as the circuit is very old, and I wasn't very familiar with them, or their failure modes, at the time of the design.

 

One can argue whether or not the input diode, D2, is even necessary in this circuit.

(I sleep better at night knowing that it is there...)

 

With the values shown, the baseline current drawn in about 0.4 mA.

One could certainly scale up the resistors and decrease this a bit, if desired.

 

Ignoring the two diodes for the moment:

At Vin = 14 V, Vout = 5.0 V

At Vin = 12 V, Vout = 4.2 V

At Vin = 10 V, Vout = 3.0 V

 

The input, reverse polarity protection, diode will also drop the Vin a bit.

The Zener diode doesn't have an exact, sharp, cut-off voltage, but using the 4.7 V zener will keep the node below 5 V.

 

The cap is missing in your present design.

You will likely wish to add it.

It, coupled with the input resistors forms a LPF, and helps attenuate some noise on the signal.

 

Recall that in addition to reverse polarity protection, and classic "Load Dump" conditions, (Make sure you Google that!), if the vehicle is ever jump started by a heavy duty tow truck / wrecker they often use a 24 V jump source for a little extra boost to help get things going, (dead batteries, thick engine oil in cold weather, etc.).

 

Good luck with your project.

 

JC

 

 

Edit:

Very odd that I drew the zener on the left side of R10.

Electrically, of course, it makes no difference.

Visually, however, it is easier to see that R9 and R10 form a voltage divider when they are side-by-side, and then one can appreciate the fact that the zener just clamps the nodal voltage.

 

Oh well...

 

 

Last Edited: Sun. Oct 22, 2017 - 04:23 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You are correct., a uni-directional TVS would be better.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Just combine R1& R2 in front of the TVS which is then parallel with R3...so it has less current to fight against...better chance of using a more protective unidirectional.  Of course, TVS has a leakage component, so the R values need to be examined if they are very large.

When in the dark remember-the future looks brighter than ever.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The catch with the Doc's example is the 'knee' and leakage of the zener as avrcandies mentions. Unfortunately zeners aren't the device you hope them to be - they don't simply 'snap' when the voltage hits, say 4.7V. There's a curve. So before you get to 4.7V, the zener starts conducting and in association with the series resistor, the voltage starts to be attenuated. Then there's the leakage that is highly temperature dependent. This is also an issue with all semiconductors - schottkys are also particularly leaky.

MOVs and transzorbs also have a degree of leakage, so watch the value of that series resistor! Otherwise you'll find your ADC values get a bit wrong up the top end. If you're not using an ADC, then this is less of an issue.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

by the way ...

 

I'm still working on my battery monitor

 

What battery monitor??  What kind of battery??...you do have to worry about the monitor constantly loading the battery (for smaller cells)....a good technique is to switch the grounded  resistor (r3) with a fet, so when off, the divider draws "no" current .

I knew someone who simply used low resistances for the divider & felt lucky that his monitor "was finding those weak short-life batteries", when the constant drain was actually the culprit.  He finally notice all batteries had reduced life (by about 20x).

 

Watch out for avalanche diodes.  I used to keep a pile of assorted values around here.

When in the dark remember-the future looks brighter than ever.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You schematic is wrong., the diode with the cathode to ground will shunt your Vin.

facepalm -- yes, that's not what I wanted to do  (red with embarassment)

 

What battery monitor??  What kind of battery?

Sorry, I posted earlier looking for a voltage regulator ( thanks!). Just measuring auto battery volts during start and run -- mayhap after shutdown at times as well. 

 

Once again thanks for all of the comments, I will be sorting through them and I might actually finsh this simple task at some point.

Thanks,

 

hj

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman wrote:
... Unfortunately zeners aren't the device you hope them to be - they don't simply 'snap' when the voltage hits, say 4.7V. There's a curve. So before you get to 4.7V, the zener starts conducting and in association with the series resistor, the voltage starts to be attenuated....
Here are some curves that show the soft Zener "knee" that predominates in low-voltage diodes and makes shunts and dividers droop:

soft Zener knee

[ Retrieved from http://www.rohm.com/web/global/p... ]

- John

Last Edited: Mon. Oct 23, 2017 - 08:06 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'm getting there thanks for the excellent comments.

 

hj