How does a battery act to suppress voltage ripple on a recified DC supply? 2

[pacmangonomnomz]

Hello everyone.

First of all, my apologies if this is in the wrong forum and for the perhaps silly question (I have no background in electrical science/engineering as you'll see below)

Following on from friend having their vehicle ECM damaged after an autotechnician decided to disconnect the car battery whilst the engine was running, I was asked why this happened and I have absolutely no idea. I've done a bit of research based on the somewhat limited information out there, but it has opened up a whole new subject matter that I'm keen to (try and) understand.

Firstly, why would a supply voltage from the alternator change sufficiently when the battery is disconnected to damage electronic components?

Secondly, I'm interested in the technical principles/process behind how a lead-acid battery suppresses voltage ripples in the DC supply. Is the ability to suppress the voltage determined by a particular characteristic such internal resistance or capacitance? Does this characteristic change over time?

Thank you in advanced.

Adam

 

[VEBill]

One possible answer is simply the response time of the voltage regulator being ill-prepared to deal with far too fast time constant of the electrical system sans battery.

Or, if the battery was charging at the time, then perhaps a load dump - and in the worst case of also not having the battery in parallel.

Standby for others to chime in. They may have better info.

 

[PeterKsiazek]

I'm with VE1BLL, the prime culprit is a load dump caused by the battery charge current having no where to go. Basically, let's say you have a charge current of around 25Amps and someone opens the circuit. If you recall one of the 'constitutive equations' of electricity we have: V=L*(di/dt) (voltage across an inductor is proportional to inductance times the rate of change of current. The stator of your alternator is basically a big inductor so if you rapidly change the current you will generate a huge voltage spike which can easily get into the 50-60V range! (for a very short period of time) This is the same process that is used to actually create the spark across your spark plug. An inductor (coil) is charged by closing a switch (usually an IGBT) and the switch is then opened which creates a high voltage across the spark plug which generates the spark.

In terms of your battery question: You can think of a battery as a really large capacitor with a higher than normal series resistance which varies based on the SOC (State of Charge) of the battery. So like a capacitor it integrates the current reducing the ripple greatly.

 

[itsmoked]

It was a load dump as already described, nothing to do with 'filtering' at all. I just wanted to correct the magnitude as it is normally considerably greater than Peter's suggested 60V. Robust vehicle electronics are typically required to be designed for +200V load dump spiking and full reverse battery voltage,(dead battery jumping by idiots). If a device is not equipped to reject these two circumstances it will be unreliable and short lived in automotive applications. It appears that the ECM was either, remarkably, not designed for load dump or in this case the load dump was actually greater than the typical 200V.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[MacGyverS2000]

You sure about that high value, Keith? Seems like the worst you could do with a car's typical component set is a voltage-tripler... inductor, couple of diode, etc. That would put it in the 55V range for a well-charged battery/alternator output, with 60V seeming like a good upper limit.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[zappedagain]

Or this 'autotechnician' who doesn't know about load dump isn't telling the rest of the story about what other stupid things he did...

Z

 

[itsmoked]

Hi Dan,
Well looking it up again, currently the highest expected voltage for load-dump is 101V for 12V systems and 202V for 24V systems. I've always designed for the 200V, after reading 200V in a semiconductor appnote, and never had a product failure from load-dump.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[IRstuff]

It's been mentioned in some car forums that removing the load can result in transient voltages as high as 200V, which is the L*di/dt of the alternator windings.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

Of course I can. I can do anything. I can do absolutely anything. I'm an expert!

 

[pacmangonomnomz]

Hello everyone,

Thank you for the responses. Unfortunately I only have my friend's non-technically minded word to go off regarding the lead up to the damage of the ECM so I can't say for sure that nothing else was interfered with on the vehicle prior to, or after, the ECM met it's demise. The vehicle was operating a 12 V system (Volvo V50 circa 2008 model - UK).

From the information posted, I'm quite surprised by just how high peak voltages can get under those circumstances.

Regarding the 'It was a load dump as already described, nothing to do with 'filtering' at all.'. That was my mistake for not clarifying, I was rolling two separate questions into one thread. I had tried to research what was happening prior to my post here but found most of searches were redirecting to the ripple suppression by the battery - which I found more interesting to be honest.

Thanks again,

Adam

 

[VEBill]

A question for my curiosity:

If the battery happened to have been fully charged, with not much more than trickle charge current (20 mA?) going into it at the point when it was disconnected, would that still cause a load dump?

 

[PeterKsiazek]

VE1BLL, if the charge current into the battery is not very high there would still be a load dump, however the overall magnitude and time would likely be reduced as the energy available to generate the pulse is much lower. However, I should point out there will always be some current going to/from the battery because of the filter current mentioned earlier (the output of the alternator is a three phase rectified sine wave). So, in general it is ALWAYS a bad idea to remove the battery when the engine is running. If your automotive technician did this then I wouldn't much trust them.

IRStuff,

It appears we are both a little bit off in our numbers, I've found an ST app note which states a typical load dump is around 80-100V peak I suspect that this number is going to vary greatly depending on the alternator, cable length, etc.

ST Application Note

There is also this link for a paper by an engineer in England which provides the numbers from ISO7637 (I don't have access to the standard so I can't verify it) which states for 12V systems the range is 65V to 87V and 123V to 174V for a 24V system

T.P. Jarvis Paper

I hope this helps, I know I learned a bit!

 

[OperaHouse]

I'd like to think an automobile ought to be designed to handle this event. A similar event with my FORD that could have resulted in personal injury. I was waiting to turn into traffic at a sidewalk with my foot lightly on the brake. Suddenly the engine reved to over 2,000 rpm and the vehicle surged. I thought my foot must have slipped to the gas. A few minutes later at a stop light it happened again, lurching me into traffic. I stopped at a parking lot and looked under the hood for something obvious. It just so happened that I was charging a SLA battery from my cigarette lighter. Charge limiter was some low wattage like 10 ohm resistor. When I tried to start the vehicle, it didn't start and that resistor went up in smoke. Obviously, the vehicle battery was disconnected. Opened the hood and the sheet metal battery clamp had snapped. The computer reset, didn't know the engine was running, opened the Idle Air Control valve fully as it would when the engine is first started. Try telling that explanation to a cop after you smashed into a vehicle!

 

[MacGyverS2000]

A new cause for "unintended acceleration"... but this time, it's repeatable.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[itsmoked]

Adam; To the ripple rejection.

A couple of points:

1) You realize that automotive alternators are three phase output right? That means the ripple is a very small fraction of what single phase generation would produce. Hence not a 'lot' of filtering is really needed.

2) The frequency of the aforementioned 3 phase power is crazy high compared to dawdling 60Hz from your local outlet. I think the typical gear ratio is something like 4 to one so even at idle the alternator is spinning at something like 3000 rpm and it just goes up from there.

3) The battery acts like a large slow capacitor, being peak charged by the peaks coming by on each cycle. This is all moderated thru the battery's internal resistance which is low on a car battery.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[pacmangonomnomz]

Thank you once again everyone - really interesting to read the responses. I'll take a look at the linked papers this evening.

I've got a few scenarios I'm playing out in my head with this new insight now, but I'm struggling to decide the outcome. Please correct me on any of the following...

Let's say a vehicle battery has a such a low SoC that the starter motor won't even turn over (maybe even has a bit of permanent sulfation). As I understand, the internal resistance of the battery is relatively high when in this state. In the case of removing the battery from the vehicle during engine running, we go from a very low internal resistance device (battery) in parallel in the circuit to an infinitely high resistance 'device'. This rapid change in resistance is responsible for the load dumping event... is this correct?

If this is the case then a similar scenario would be played out but with less extreme values during jump-starting(?). I can't remember the last time I jump-started one of my cars, but this thread has got me very, very slightly nervous now!

Can a fast 'switch' to a low SoC battery in the scenarios described below cause load-dumps (or increased ripple? Since it's effectiveness of ripple suppression/rejection is affected by the internal resistance)?

1) What risk is there to automotive electronics when disconnecting jump leads following a jump-start? Will there be a notable peak voltage? Will there be increased ripple initially until the charge of the battery increases?

2) Alternatively what if a vehicle is jump-started by a standalone spare battery? Once the engine has been started and the leads are removed, would there be a load dump/spike/increased ripple in this circumstance? I'm guessing it would be similar to the above?

In both cases, is this why it is recommended to continue being connected to the donor vehicle post-jump start for a few minutes?

Hoping I haven't out-stayed my welcome with the numerous questions!

Regards,

Adam

 

[sbkenn]

From my knowledge, a battery looks like a voltage source with a huge capacitor in parrallel. A sever load dump is only likely to occr if the battery is disconnected while then engine is running. One solution is here:

http://www.st.com/web/en/catalog/sense_power/FM114/CL1801/SC1496