Post by thread:[EE]: Caps & automotive high speed transients

'[EE]: Caps & automotive high speed transients'

I am having a hell of a time with coil/points automotive ignition noise in

my power supply.

Do high speed transients ( 60v & < 5ns ) damage 50v aluminum capacitors?  If

so, what's the best way to suppress them? The 18v varistor (Digikey #

F2186CT-ND) and 600 ohm ferrite (240-1016-1-ND) can't stop the noise.

Would a new cd ignition generate less noise?

This project is killing me,

Jay

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

jayhanson <spam_OUTpiclistTakeThisOuTmit.edu> wrote:

> I am having a hell of a time with coil/points automotive ignition

> noise in my power supply.

Jay:

See Tom Williamson's Intel Appnote AP-125, "Designing Microcontroller 

Systems for Electrically Noisy Environments":  

    http://www.intel.com/design/mcs96/applnots/21031302.pdf

-Andy

=== Andrew Warren -- .....aiwKILLspam@spam@cypress.com

=== Principal Design Engineer

=== Cypress Semiconductor Corporation

=== 

=== Opinions expressed above do not

=== necessarily represent those of

=== Cypress Semiconductor Corporation

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

On Tue, 28 Sep 2004, jayhanson wrote:

> I am having a hell of a time with coil/points automotive ignition noise in

> my power supply.

>

> Do high speed transients ( 60v & < 5ns ) damage 50v aluminum capacitors?  If

> so, what's the best way to suppress them? The 18v varistor (Digikey #

> F2186CT-ND) and 600 ohm ferrite (240-1016-1-ND) can't stop the noise.

>

> Would a new cd ignition generate less noise?

Simple solution: buy a mains filter (Corcom etc) and put it in the supply 

of your circuit, after lifting all grounds from the car ground. It works 

for dc at low current. Otherwise a lc filter before the regulator is 

usually a good start (use a choke rated inductor, not a normal inductor).

Peter

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Peter L. Peres 04:51 2004-09-30:

>>Do high speed transients ( 60v & < 5ns ) damage 50v aluminum capacitors?  If

>>so, what's the best way to suppress them? The 18v varistor (Digikey #

>>F2186CT-ND) and 600 ohm ferrite (240-1016-1-ND) can't stop the noise.

Where do you get that transients picked up?

Are you sure that noise is real - you don´t get that noise picked up by the test leads?

Inductor, varistor and ceramic cap to GND should be enough on power supply input.

(After them maybe another choke and then alu cap.)

/Morgan

--

Morgan Olsson, Kivik, Sweden

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

I have built a digital motor and radiator thermometer using DS1820 for my

car and the leads for the sensor are passing right below the high voltage

choke and current generator. I have used just a choke I’ve picked up form an

old device, a current limiting resistor, a 25V zenner diode, a 470uF

aluminum capacitor, a 100n ceramic and the old LM7805 with 22uF and 100nF at

the output. I had no problem what so ever with this setup in more than a

year now.

I have designed and built electronics for big trucks (stone exploitation

Russian trucks and I can tell you there is a lot of noise in there) and

added a linear 15 regulator (a NPN transistor with a 15V zenner and a

100uF/63V in the base o the transistor) and had no problem with this either.

I don’t think you really have noise in the power line. When you are

measuring noise on the supply line pick the GND from the negative lead of

your filtering cap not from the end of the GND trace.

Hope it helps,

Sascha

{Original Message removed}
What is the difference between a choke rated inductor and a normal inductor, (besides, obviously, the words "choke rated" on the datasheet)?

-- Lawrence Lile, P.E.

Electrical and Electronic Solutions

Project Solutions Companies

http://www.projsolco.com

> {Original Message removed}
>>>Do high speed transients ( 60v & < 5ns ) damage 50v aluminum capacitors? 

>>>If

>>>so, what's the best way to suppress them? The 18v varistor (Digikey #

>>>F2186CT-ND) and 600 ohm ferrite (240-1016-1-ND) can't stop the noise.

>

> Where do you get that transients picked up?

>

> Are you sure that noise is real - you don´t get that noise picked up by 

> the test leads?

>

> Inductor, varistor and ceramic cap to GND should be enough on power supply 

> input.

> (After them maybe another choke and then alu cap.)

One or more stages of LC filter should do the trick. You can calculate approximate efficacy theoretically.

Depending on what the input and output voltages are and how much power you need, there is a simple trick which may help. We used this with great success long ago in a Telco environment. We simply used two stages of zener plus resistor filtering. ie input R, zener to ground, resistor to next zener to ground. Regulator feeds from second zener. Provided that the voltage drop in the resistors is acceptable to you plus the power in the zeners then this does a good & cheap job of greatly diminishing spikes to the stage where a regulator has no problems with them. If you have eg 12v in and 5v out then you have enough headroom to do this and power loss in resistors is possibly better than in the regulator anyway. Depends muchly on power level.

        RM

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

>What is the difference between a choke rated inductor and

>a normal inductor, (besides, obviously, the words

>"choke rated" on the datasheet)?

I suspect a choke rated one will have a rating for a DC component while

maintaining the inductance value without saturating. An ordinary inductor

may be rated only for an AC component rather than AC+DC.

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

> What is the difference between a choke rated inductor and a normal 

> inductor, (besides, obviously, the words "choke rated" on the datasheet)?

I'd guess he / they mean that it is rated to not saturate at a stated DC 

current. (or to have the stated inductance at the rated maximum DC current).

1 Henry is easy and can be small & light.

1 Henry at 1 Ampere, well ... :-)

        RM

>> Simple solution: buy a mains filter (Corcom etc) and put it in the supply

>> of your circuit, after lifting all grounds from the car ground. It works

>> for dc at low current. Otherwise a lc filter before the regulator is

>> usually a good start (use a choke rated inductor, not a normal inductor).

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Peter L. Peres 04:51 2004-09-30:

>>Do high speed transients ( 60v & < 5ns ) damage 50v aluminum 

>>capacitors?  If so, what's the best way to suppress them? The 18v 

>>varistor (Digikey #

>>F2186CT-ND) and 600 ohm ferrite (240-1016-1-ND) can't stop the noise.

>

>Where do you get that transients picked up?

>

>Are you sure that noise is real - you don´t get that

>noise picked up by the test leads?

>

>Inductor, varistor and ceramic cap to GND should be enough

>on power supply input. (After them maybe another choke and then alu cap.)

I believe the noise is real because my processor board resets every so often

with the motor running (I took out WDT for testing).  

Conceptually, this project is a walk in the park. In reality, I am about to

fall on my sword.

I think the problem is that I have created a special problem in this car.  I

have been screwing around with this car -- on and off -- for three years

now!!

When I designed the latest boards (yes, there have been several versions), I

took the advice of Bob Axtell:  Large varistors on sources, small varistors

on board, optoisolators, 47pf emi filters (P9826CT-ND) and 47pf caps on each

input, current operated (bipolar) transistors to run my high-current relays,

a two inch copper strap down the middle of the car for ground, and a TI

switching regulator for my processor.  ** I even have an totally-isolated

separate 12V battery for my processor board!!!

The car is unusual in that it's a 1973 VW thing.  It was designed to run on

a generator but was converted to alternator.  The noise appears to come from

the spark coil and points (both not original equipment).  I can measure

these 60v spikes on terminals of the car battery AND on the terminals of my

processor board with the separate battery!  Could it be radiating that much

energy through the air?

I measured another newer car and truck and they do not have spikes like this

in their electrical systems.  I am about to install a new capacitive

discharge ignition system (MSD, doing it today).

Jay

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

On Tue, 2004-09-28 at 20:14, jayhanson wrote:

> I am having a hell of a time with coil/points automotive ignition noise in

> my power supply.

I used this circuit in my pick-up: http://www.seetron.com/an_vpwr1.htm

I think I got it from the piclist a year or so ago. It works great on my

pick-up (a diesel).

Failing that, rip out the petrol engine and stick a diesel engine in. :)

Cheers,

Tim

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

----- From: "Tim ODriscoll" <tim@

> On Tue, 2004-09-28 at 20:14, jayhanson wrote:

> > I am having a hell of a time with coil/points automotive ignition noise in

> > my power supply.

> 

> I used this circuit in my pick-up:

 http://www.seetron.com/an_vpwr1.htm 

Add a 5 ohm 10W resistor in series with D1 and a cap to gnd. after it and

maybe a 0.1uf  to gnd on each side of D1 and you should be able to use the 7805 up to 1 amp.

The RC combination will reduce the voltage dropped across the 7805 and reduce noise on the 5V.

KF4HAZ - Lonnie

> I think I got it from the piclist a year or so ago. It works great on my

> pick-up (a diesel).

> 

> Failing that, rip out the petrol engine and stick a diesel engine in. :)

> 

> Cheers,

> 

> Tim

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

On Thu, 30 Sep 2004, Lawrence Lile wrote:

> What is the difference between a choke rated inductor and a normal 

> inductor, (besides, obviously, the words "choke rated" on the 

> datasheet)?

It works with dc through it and it has a low Q factor. This prevents it 

from ringing too much with large but fast spikes. A high quality inductor 

followed by a high quality capacitor used as a filter would cause enough 

ringing at the LC resonant frequency, when subjected to a spike or step 

pulse, to disturb downstream circuits. The low Q is usually obtained by 

winding it in a certain way so it has a large stray capacitance vs. 

ground. There are also hybrid chokes that use other methods to kill 

pulses (transmission line like).

Peter

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

On Thu, 30 Sep 2004, jayhanson wrote:

> I think the problem is that I have created a special problem in this car.  I

> have been screwing around with this car -- on and off -- for three years

> now!!

If it still runs it's ok. Try to put a 100pF cap between Vdd and MCLR 

directly at the chip (regardless of what else is there). The MCLR spec 

has the lowest noise margin on pics afaik. 0.5V below Vdd and you're reset 

afair.

> The car is unusual in that it's a 1973 VW thing.  It was designed to run on

> a generator but was converted to alternator.  The noise appears to come from

> the spark coil and points (both not original equipment).  I can measure

> these 60v spikes on terminals of the car battery AND on the terminals of my

> processor board with the separate battery!  Could it be radiating that much

> energy through the air?

Yes and no. Cables in harnesses that live in the same metal compartment 

couple into each other strongly. Also old cars do not have 'grounds' they 

have something that is most of the time below +12V. Unless you turn the 

headlights on or something ;-)

> I measured another newer car and truck and they do not have spikes like this

> in their electrical systems.  I am about to install a new capacitive

> discharge ignition system (MSD, doing it today).

The spikes are not supposed to be there at all. Poor wiring to the voltage 

regulator and/or generator can cause them, as well as picking power for 

the ignition from an arbitrary point in the system and grounding things 

arbitrarily. F.ex. check if the engine block still has a good ground (<0.2 

ohms to battery - including meter wires).

Peter

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

I am happy to report the problem has been fixed.  I was experiencing 60v,

<5ns spikes occurring at about (I think) 100 mhz.  I had an old coil,

points, and copper wire spark plug wire ignition system.  I had a cap on the

distributor but not one on the coil.

I fixed the problem by replacing the old ignition system with a new

capacitive discharge system from the MSD company.  I put in a new control

unit, coil, distributor, and resistor spark plug wires from MSD.  I am still

getting some 20V noise from the alternator but my circuits can handle that

just fine.

My advice to anyone who wants to put a PIC project into an old/modified car

would be to FIRST upgrade the ignition system to a modern capacitive

discharge system.

Thanks for all the help from all of you,

Jay

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Thanks for sharing, Jay. This is a continuing problem for all of us.

Every bit of sucess is cherished.

--Bob

jayhanson wrote:

{Quote hidden}>I am happy to report the problem has been fixed.  I was experiencing 60v,

><5ns spikes occurring at about (I think) 100 mhz.  I had an old coil,

>points, and copper wire spark plug wire ignition system.  I had a cap on the

>distributor but not one on the coil.

>

>I fixed the problem by replacing the old ignition system with a new

>capacitive discharge system from the MSD company.  I put in a new control

>unit, coil, distributor, and resistor spark plug wires from MSD.  I am still

>getting some 20V noise from the alternator but my circuits can handle that

>just fine.

>

>My advice to anyone who wants to put a PIC project into an old/modified car

>would be to FIRST upgrade the ignition system to a modern capacitive

>discharge system.

>

>Thanks for all the help from all of you,

>Jay

>

>

>_______________________________________________

>http://www.piclist.com

>View/change your membership options at

>http://mailman.mit.edu/mailman/listinfo/piclist

>

>

>  

>

-- 

Note: Attachments must be sent to

 attachKILLspamengineer.cotse.net, and 

 MAY delay replies to this message.

        520-219-2363

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist