Cordial Greetings,

On this forum, I've noticed repeated references to how a bad battery can cause all sorts of faults, OBD-II codes, and warning messages. Many people have sworn a new battery fixed everything, and I am not doubting them. However, I am wondering exactly how this happens. I have a theory and am interested in what others may think.

I'm an electronics tech (aviation), so have a decent working man's knowledge of electronics. I rarely work at component level (i.e. replacing a resistor on a circuit board), but work more at troubleshooting entire systems. I got to thinking about battery issues while up on my AC Ripple Soapbox(tm):

https://www.jaguarforums.com/forum/s...galore-117374/

BTW, this is NOT to be confused with my Misfire Soapbox(tm), which is an entirely different subject. So I get to thinking (don't worry, it's social thinking and I can stop any time I want), what if the battery itself wasn't at fault. What if AC ripple was the primary fault, and the battery was acting like a giant capacitor to mask the problem. As the battery ages, its ripple smoothing ability might decrease. Replace an otherwise fine battery, and voila, the problem seems to go away. Meanwhile, you've only masked the problem instead of actually fixing it.

Whaddya think? Part of what stirred my thinking was how some hi-tech battery testers run an AC current through the battery and then run some calculations to get an idea of battery health.

 

Interesting theory, but it operates on the premise that since most/all S-types display the classic set of codes when they want a battery, most/all S-types have ripply power.

How would you explain that abnormality?

 

In most cases, malfunctions due to a weak battery that still shows around 12 volts unloaded happen because of the large voltage drop on the weak battery during the process of starting.

The electronics modules might see 8 volts or less momentarily, and this is enough to cause the some of the module voltage regulators to drop out of regulation, causing soft faults in the digital citrcuits.

 

The same alternator...

I think even some nonstock alternators (such as the ones that won't extinguish the battery light) are built on the same basic series with the same diode packs.

That's my theory, anyway.

 

The onset of codes with a failing battery is not unique to S-types or Jags, many modern complex cars do similar things under similar circumstances. Do they all have flawed alternator designs?

I tend to side with the conventional wisdom which states that low system voltage during cranking gives the computers mini-fits as stated above.

 

I used to drag race and put batteries in the trunk of all my drag cars , always had issues with the voltage drops cos of the long leads from trunk to engine... I wonder if this is perhaps an issue with the S-type?

 

It's not, except during starting, as you can readily check with a meter (or ETM, which is a kind of meter).

 

I'm not saying the alternator is a necessarily a "flawed design". All alternators are prone to AC ripple, especially as they age. My hunch involves a combination of factors at play. I'm just not sure what all those factors may be, but read on to hear about an additional possibility.

But first, one thing to consider is that plenty of alternators out there (all brands of vehicles) are pumping out excessive AC ripple, but few people know to look for it or that it can even be a problem. Heck, on older vehicles with minimal electronics, AC ripple probably doesn't cause much problem at all. Think of it like the OBD-II misfire monitor on all newer cars. Before that, how many vehicles were running on less than all cylinders and it wasn't even noticeable? I've turned several V7 engines back into a V8 again, but in the pre-monitor days, the fault really wasn't very obvious.

Now back to the present day and Jag battery sensitivity. IF my theory is correct (it's a big if), all those pesky codes and fault messages might be an inadvertent warning of excessive AC ripple, even though there's nothing in the system designed to watch for this particular fault.

I had an aha moment this morning after pondering the other question about the battery placement and long cable. Read this article. Even though it's for Mercedes, the principles are the same for all modern vehicles:

http://www.pvv.org/~syljua/merc/TooSeptST07.pdf

The article talks about how AC ripple can cause all sorts of seemingly unexplainable electrical problems. Here's the juicy part, on page 3:


"Where you measure ripple is as important as the tools you use to do it. The vehicle battery acts like a gigantic capacitor, smoothing out alternator ripple. The closer to the source of the noise -- the alternator -- the more accurate a measurement you’ll get. Ideally, you want to place your scope leads on the alternator B+ terminal and against the alternator case itself."

Reading between the lines, the greater the distance to the battery, the greater the potential for AC ripple issues. In the immortal words of Colonel Hans Landa (bonus points if you know this obscure reference): "That's a bingo!"

Maybe the long battery cable is the missing piece of the puzzle explaining why Jags seem to be more sensitive to battery condition. I need to stress I'm not saying this conclusively. I'm just suggesting further brainstorming is warranted. I wish I had access to a Jag or two exhibiting the symptoms. I've only got the one Jag in my driveway, and so far it's fine.

However, through this forum, we do have virtual access to a large fleet. How difficult is access to the alternator terminals? Is there a convenient access point nearby? I think it would be interesting if we could post a simple, universal test that anybody could do and then compare the results. Factors to list include measured AC ripple, battery age, and the presence of other electrical faults. If anybody is interested and open-minded to a little experiment, we should first agree on a common test point (at or close to the alternator) to avoid skewing the results.

 

But they're not. Pretty well all brands of cars suffer the same whacko behaviour as the battery starts pining for the fjords irrespective of battery placement.

If the battery cables where the cause-why would the fault disappear when the new battery is installed?

 

As the ABS/DSC is typically one that flags the fault - and at engine crank/start - stick a scope on its power and tell us the ripple. And then explain why it nearly always only flags the ABS fault during a start with a failing battery. May as well measure the (DC) volts during cranking, too, preferably using a storage scope so you can see the spikes & min volts.

Now, the idea that a new battery will manage to be a better capacitor and remove the ripple (if any) despite the much longer leads to the battery than the ABS... sounds optimistic.

 

Can you measure this ripple? Even if I had a number I don't know what would be too much and what would be normal?

Cars have been using alternators for 50+ years. Has this always been present?
.
.
.

 

I'm not saying the long battery cable is the cause. It's more of a contributing factor specific to any vehicle with the long cable. The long cable exacerbates the problem, reducing the battery's ability to smooth the pulses (normal or otherwise) from the alternator.

Now before anybody jumps in to say the length of the cable doesn't matter, please read the linked article again. Mercedes is telling their mechanics to always check for AC ripple right at the alternator, not at the battery. Even with a big fat cable connecting the two, the pulses apparently smooth themselves out somewhat the further they travel. Unfortunately, most systems get their power supply closer to the alternator, not the battery.

Now for your question about why the fault disappears when the new battery is installed, that's easy. Amongst other functions, the battery also acts as a giant capacitor to smooth the pulses from the alternator. A capacitor is two plates (or series of plates) separated by an electrolyte, which is also the same basic description of a lead-acid storage battery. Change the chemical composition of the electrolyte, and the capacitance changes. This is one thing that changes as a battery ages. The other change is crud builds up on the plates, also changing the capacitance. Reduced capacitance equals reduced ability to smooth AC pulses.

Unfortunately for us, Jag placed the battery at such a distance as to limit its effectiveness. Not really a huge deal, but still a factor. Suppose you had two vehicles with virtually identical electrical systems and both had equally defective alternators pumping out the same amount of AC ripple. The only difference being one vehicle has the battery further away from the alternator, you would theoretically see one important difference as the batteries age. The one with the greater distance to the battery would be the first to experience AC ripple above a specified limit.

That's what I'm trying to suggest here. Not that any alternator design is inferior, etc. Simply that the long battery cable slightly restricts the battery's ability to smooth AC ripple, and even though a new battery can get that ripple back within limits, we might be overlooking a marginal alternator as the root cause. Or if not the sole root cause, maybe it's part of the issue in concert with a marginal battery and the limitations of the long cable.

Is this how Galileo felt when questioning the collective concensus of his day? <g>

 

I interpret the Mercedes article differently. They don't want the ripple being measured at the battery but at the alternator instead to avoid the smoothing effect you've mentioned. The length of the cables is irrelevant.

Do you have any studies indicating that batteries lose their ability to smooth as they age?

 

Are we reading the same article? Per Mercedes: "The closer to the source of the noise -- the alternator -- the more accurate a measurement you’ll get."

Now I'm going to out on a limb here and extrapolate that to mean the AC ripple in the cable decreases the further you get from the source. So even if the battery end measures relatively clean, the alternator end of the cable can be relatively dirty, and that's what is supplying most (but not all) of the vehicle's electrical system. The battery is working with one hand tied behind its back trying to smooth those pulses at the far end of the cable. So yeah, I'd say the length of the cable is important.

Nope, strictly conjecture on my part. I wish I had a car at my disposal with the classic symptoms (ABS fault, etc.) To check my theory, I'd measure AC ripple both before and after a battery replacement. Just for giggles, I'd also do before/after readings at the alternator and at the battery to confirm what Mercedes says about the cable length and its effect on AC ripple measurement.

 

These articles explain how to measure it. On modern computer-controlled vehicles, 0.5 VAC is a practical maximum limit:

http://assets.fluke.com/appnotes/automotive/beatbook.pdf

http://www.pvv.org/~syljua/merc/TooSeptST07.pdf

The second article has a lot of details of the gotchas when taking a measurement. An oscilloscope is preferred, but for most instances, an AC voltmeter is sufficient. However, because an AC voltmeter is designed to interpret the nice smooth sine waves of "clean" AC power, it may not accurately register the choppy spikes you might have in a failing alternator. Basically, the AC voltmeter is being asked to do something for which it wasn't really designed. So if the AC ripple test seems to pass, just be aware it's not conclusively good. By contrast, if it fails, then that's conclusively bad, but it is not 100% guaranteed if the test passes.



Yep. It's only a problem when the AC ripple exceeds what the onboard electronics can handle. Newer cars are more finicky about their power supply due to higher degree of computerization.

 

I see multiple issues:
1. we've not measured the a/c ripple so it's conjecture how much there is
2. many of the weird faults occur on starting in the morning, and clear if you re-start after a few miles - well, the battery's capacitance won't have changed nor will the alleged ripple
3. we know low battery voltage triggers the faults (it's even in JTIS)
4. the various modules I've had apart (i.e. looked inside) have internal regulators, so they won't care about a/c ripple unless it's HUGE - no evidence at all that it is

I don't see evidence for a/c ripple even being involved let alone the cause.

 

The long cables from bat to starter etc cause voltage drops when cranking.
I used to use 2 truck batteries in parallel to start some of my dragsters with the bats in the back.
Grounds are VERY important as is the resistance of the cables - I used to use massively thick welding type cable...
I use a Ctek charger on my jags battery - as it seems short trip city driving does not charge it up fully , also carry a portable jump starter in my trunk , have had to use it once or twice after the car refused to start...
Battery has been tested good , if I havent used the ctek for a week or 2 , I start getting battery related no start/swing/crank problems.

 

#1) Just for giggles, I did measure the AC ripple on my '02 V6 S-type. I have had no electrical issues with this car, other than a handful of times I'd get an intermittent code P1584 and Failsafe Mode, perhaps lasting 10 seconds at the most each time. The battery is about 3 years old. All measurements were taken with a Fluke 83 DMM. Readings were taken after a 45 minute drive, so I believe the battery was sufficiently charged. To load the electrical system, I had the high beams on, HVAC fan on high, and rear defroster on (No electric seat heaters).

Directly at the alternator, I measured 0.450 VAC at 2700 RPM, and 0.150 VAC at idle.

At the battery, I measured 0.150 VAC at 2700 RPM, and 0.030 VAC at idle.

The test point makes quite a difference, with the AC ripple 3-5 times greater directly at the alternator.

I don't currently have JTIS access (no home PC) to know Jag's limits, but 0.450 VAC comes in just under the 0.500 VAC generic limit recommended in other places. I'm not saying this alternator is bad, unless I hear otherwise about any limits specified by Jaguar. Au contraire, it appears to have normal AC ripple, and this simple test confirms it does drop dramatically closer to the battery.


#2) As far as the clearing of symptoms during subsequent restarts, two important things have changed:

The battery was previously dead but now has at least a partial charge. This changes the chemical characteristics of the electrolyte and so probably (strictly conjecture on my part) changes the overall capacitance.

The other change is temperature. The battery should be warmer after charging, but I've no idea if that changes the capacitance. The temperature of the alternator is potentially a far bigger factor. Now I'm not talking capacitance for this part. I've seen a lot of temperature related electrical faults. Heat can cause many components to act up, but on rarer occasions, I've seen just the opposite. I think that's from intermittent connections being forced together by thermal expansion, but don't know for sure. Rare, yes, but don't automatically rule out something like that.

As far as "alleged ripple" please see my #1 comments above. I guarantee ANY alternator will have some AC ripple. The only question is how much and whether it's out of limits.


#3 Not trying to be difficult, but what exactly does JTIS say about the battery and mystery faults? Does it specifically mention low voltage, or just say a battery fault? I'd like to read more, but am hampered without JTIS access.

#4 Agreed. The only catch is a guy would easily miss AC ripple if he doesn't think to look for it.

At my day job, I do a lot of electrical troubleshooting, usually under VERY tight time constraints. By nature, we usually have to shoot from the hip first, assuming we have a pattern failure. If that fails, we have to double back and do more in-depth troubleshooting. In the many cases where a new battery seems to be the fix on a Jag, that's the easy pattern failure and repair. For the situations where it doesn't work, I would highly suggest checking AC ripple. It took me all of two minutes on my car. In fact, the test was so simple, I'd even recommend it before replacing the battery. Maybe this would only be helpful in 5% of the cases, but if it can solve a problem that defies all other attempts, that two minutes will be well worth it.

Once again, I'm not trying to step on any toes, but am just trying to be helpful. Unfortunately for you guys, I'm also very pigheaded and I never forget anything that has helped me in the past. Now I just need somebody experiencing strange electrical symptoms to spend the two minutes to help prove or disprove my theory.

 

I can't say if you're theories are right or wrong, all I know is that changing the battery out in many cases seems to solve a whole host of issues.


I simply don't have enough knowledge to either agree or disagree...


Having said that, please don't feel that the guys are having a go at you in any way shape or form, Its simply their perspective / knowledge being put across as is yours.


Its what the forum is all about, the more ideas / suggestions we can get out there potentially, the better for all in creating a useful and worthwhile discussion.


Thanks

 

Thanks. Here's another interesting article, this one seeming to support both sides of the Great AC Ripple Debate of 2014, as it came to be known in Jag folklore:

http://www.motor.com/article.asp?article_ID=1766

On one hand, the test case vehicle was diagnosed with AC ripple high off the chart and he proved a change in battery capacitance fixed it. On the other hand, the actual fix was just a new battery and nothing else.

I guess in some ways, I should say I'm satisfied with those findings. My inquisitive nature had me wondering about the specifics of "why" a new battery is often the fix. I think it's the change in capacitance and resulting changes in AC ripple. Others think it's due to voltage issues. As my dear, prematurely gray-haired mother often said to me, "You're entitled to your opinion, no matter how wrong and boneheaded you may be."