Antigravity Batteries rock

A5/S5/RS5
#12

So we’re looking for a bit of help from Antigravity battery owners who also have a VCDS. Some owners are experiencing an over voltage situation where the car can go so far as to shut down. I went for months and months with no issue and had a big one on Tuesday. Cycling the ignition off and back on resets everything and the car runs as normal.

I know of at least three other owners who have experienced this problem and I’d like to get the “hive” together on this one to solve it once and for all.

We have another member who’s composed a list of various VAG OEM battery fitments in different sizes (amp hours), battery type (flooded, AGM, etc…) and cold cranking amps. We’re experimenting with those different codes to see if one is more compatible than another.

To further confuse everyone, there are a number of owners who have zero issues and use their cars hard (on track even). So it’s sort of a mystery as to what the root cause of this is.

I know I have several codes when I do a full scan. My first theory is if there’s anything “upsetting” the CAN-Bus, it can cause issues. I currently have a code for seat back out of range adjustment (never been able to get rid of that one!), a HVAC flap code (that I induced and just haven’t run through adaptation to remove) and an MMI code as I used to have it set up where I could use my phone as a mobile hot spot with the MMI. Now that I have CarPlay, that isn’t needed.

The first time I had an issue, I had an intermittent headlight bulb at the same time. I replaced the bulb and continued to drive along without any issue until yesterday. I had tried a different BEM code this time around to see if it would potentially change the charge routine and work for others with issues. I’ve now changed it again to an AGM-type battery, approximately 50Ah.

The following was procured and collected by VolvoFan and I’m reposting with his permission. These are a list of different battery codes (not the full BEM) and what they mean in terms of amp hours, size and battery type. I changed my code to a “CC” this evening but probably should have tried the CB. The “c” portion means it’s an absorbed glass mat (AGM) battery, which, according to Antigravity, comes closest in terms of the recharging logistics of a lithium ion battery. So there’s “logic” to the coding and they’re not random letters.

All part numbers are 000915105** where the ** is filled in by the two letters below.
2 LTRS AMP HRS DIN AMPS TYPE (I think?)

DB 44 220 Conv
BB or DC 51 280 Conv
CB 58 340 AGM
EB 59 320 EFB
AC or DD 60 280 Conv
AD or DE 61 330 Conv
CC 68 380 AGM
DF 70 340 Conv
FC 70 420 EFB+
AE AF or DG 72 380 Conv
CD 75 420 AGM
ED 79 420 EFB
AG or DH 80 380 Conv
AJ or DJ 85 450 Conv
CE 92 520 AGM
EE 93 520 EFB
AH or DK 95 450 Conv
CF 105 580 AGM
AK or DL 110 520 Conv
DM 115 600 Conv

Batter vendors and their associated code:
5DO = JFF/Boading
TU3 = Exide
MLA = Moll
JCB = JCI/JCB
VA0 (or VAO) = Varta
UM5 = Akuma
BA2 = Banner

So VolvoFan used this code with his battery:
000915105EB VA0 081120AGRA

The first 11 digits are the battery type identifier. In that set, the last two tell us the “type” as in amp hours and whether it’s a flooded, agm or conventional. The middle three are the manufacturer and the last set of digits are just “made up” using the date he installed the battery and AGRA representing Antigravity.

I was using the same code but with the “EB” switched out to CB. I tried using DB as the 44ah is close to the Antigravity’s 40AH but the VCDS would not take it.

I’ve now switched it to CC to see if coding it as an AGM battery will help.

One theory I have is that the regenerative braking is causing an issue. Audi’s alternators have something where they use the engine braking to load up the alternator and charge the battery instead of loading the engine under acceleration. I’m finding my voltage creeping up as I brake and my incident happened after braking and then accelerating rapidly.

What I have planned next is for anyone who has a VCDS to log certain parameters while driving and we can compare numbers as well as BEM codes.

If we have any electrical engineers, I’d love to hear your thoughts on possible causes to the over voltage issue as well as possible remedies to the situation.

When I have a bit of time this weekend and next week, I’m going to list which electrical parameters I’d like logged by everyone who participates.

Any ideas greatly appreciated.

#13

I would say most instances of my overvoltage faults occurred during the same conditions you mention, braking then on to mostly aggressive accelerations. I think that is a start. Wonder if the system can be recoded? Is that the convenience module that controls the charging?

#14

I don’t think it’s in the convenience module, you’d log in the main electrical system module and the can bus but I’m going to dig into it and see if I can log the requested and actual alternator output and a few other variables to see what’s connected to what and try to determine where the problem is occurring. Going to try and enlist the help of the Ross Tech guys too.

#15

I don’t want to complicate this too much and you might have said something about this already… But on the 5.2 S6 there is a battery adaption that you have to do when you install a new battery. Some cars have no issues others have issues if you don’t do the adaption. The over voltage seems odd because the battery just receives voltage when the car is running. I know the battery is constantly giving feedback to the computer but it would be interesting to see what voltage is causing to give the over voltage code.

#16

**I am not an electrical engineer. The possible easy fix: Can you put the battery in extended life mode. In order for there to be spare capacity (capacitance?) when recuperation kicks in?

A standard battery charges faster with a lower state of charge, so I figure if it kept below for example 60 % it will be more likely to be able to take the charge. To my knowledge the same holds for EVs. So better trick tell the ECU the true Ah than tell it the battery has more.

The charging is active during coasting as well, but I figure no sudden spike then.

Try switching off the car more often at red lights, as the technology was developed for start-stop cars mainly (I believe).

#17

Any thoughts or suggestions are totally welcome! Not sure what years you’re referring to but I think Audi has changed the “logic” with their battery management system. Now it’s just code in and forget with the actual code telling the system what the amp hours, cca, and type of battery is being installed. It does remember up to three batteries you’ve installed in the past too.

But I’ll look into that and see if there’s some way to run adaptation on it.

The voltage isn’t coming, I believe, from the battery end, but from the alternator end of things. The lithium ion batteries can charge and discharge far more rapidly and they have less than half the internal resistance of a normal lead acid/agm type battery. That’s why a small 40Ah battery can crank over the RS5’s motor with such authority. There’s no difference in the cranking/starting between the Antigravity and my old 110Ah battery. In fact it probably turns over a hair more quickly.

#18

^^^typed that up this morning but forgot to hit send. Busy day.

Anyway…I literally stumbled across an interesting bit of information. There is a product called RS Nav which is an Android Auto/Carplay interface for those who don’t know. It’s a stand alone device which interfaces with the MMI.

On later model RS Nav’s, they have something called “dual clutch support” in their setup menu. If you have a dual clutch-equipped Audi, and don’t check that box, it can cause the car to have almost the exact same issues as what’s being seen with the AG batteries. Dash lights up with multiple warnings.

I am trying to find the root cause of the RS Nav issues and what exactly checking that box does to change the electrical behavior and keep the errors at bay. Just thought the parallels were too close to be a coincidence. They somehow have to be related and it might be a coding issue.

I have an older RS Nav installed and Greyson had one in his car as well. With that said, Greyson’s car ran fine with the RS Nav installed and a normal battery. There’s a change in the later version RS Nav which must be causing the issue.

#19

Just to clarify, I was thinking the ECU would think the battery was half full when the lithium ion battery is in fact at >80 %. At that charge the rate of charging accepted by a typical li ion battery would be limited to less than what the alternator produces at hard braking (to extend the life of the battery). I am almost sure it is using voltage difference to determine the state if charge though, so it does not play in.

#20

Do have an error log code programmed on the battery?

Sort of a side note, did anyone manage to enable the battery screen on the MMI using the green screen?

#21

The new code I’ve entered is for a MUCH smaller battery, this time a 50Ah AGM battery. The voltage at the battery is always well above 13V.

Ross-Tech replied to Volvofan stating they didn’t think the Gateway wouldn’t support a lithium ion battery due to the car being seven years old. But gave no logic behind that very unscientific statement. I’ve asked him to elaborate as to what “wouldn’t support” means from an electrical standpoint.

Thinking last night, I have a feeling it may have to do with the alternator charging under braking (or charging period) and then when you accelerate, the increased rate of discharge from the battery may be causing issues, somewhere in that quick transition. It’s not a voltage dip, that’s for sure but the car does go into some sort of protection mode.

The other line of thought is it’s the AG battery itself which is shutting down when it sees more than 14.7V from the alternator.

#22

There are no battery-related errors when you pull up the VCDS and scan for codes. The codes only seem to exist in other unrelated modules.

I’ve not tried to find the battery menu in the green screen as I figured it’d be easier to accomplish this with the VCDS. I’ll look into it and see what options are in there.

I still find it odd an add-on CarPlay unit causes the same errors if the “dual clutch support” box isn’t checked in the unit’s setup menu. Maybe the circuitry inside the AG has some sort of firmware that doesn’t play nice the CanBus in a similar fashion.

#23

When I was having it on my 2013 S5 I got a couple of spikes all the way to 15volts. It would be a fairly violent safe mode kicking in. The transmission would kick into a neutral, without an indication of neutral, engine would go to idle, then about 5~10sec later it would recover. I would get a Transmission warning in the DIS but when I’d scan it at home with Vagcom there would be not transmission or engine faults. Still running the OEM battery in the S5 but would like to go with the AG in both the S5 & RS5. S5 Alternator 150amp Valeo, RS5 190amp Hitachi, same battery and electrical module part #'s.
From the B8.5 N. America press release: “The powerful RS 5 features an optimized engine and drivetrain, on-demand oil pump, and
energy-recovery system that conserve energy during coasting and braking.” And read the Recuperation part of this Audi brochure.

#24

As it cranks up effortlessly with the Antigravity battery, maybe the battery settings are a bit too good (=powerful) for what is needed. Can you limit/lower the output voltage slightly from the battery via an update over the bluetooth for cars with recuperation. Or change the fast charge setting so it can “trap” the spike.

#25

Or, since AGMs are allegedly capable of higher recuperation than regular lead acid batteries, try an adaptation for the latter?

#26

I tried the AGM on my S5 but I was using a small AG battery, half the amps of a OEM battery. Still cranked the car up fine, no difference. I’m sure the Reserve capacity if I let it sit off a trickle charger would have shown. AGM setting didn’t help in my case but we’ll see with a bigger AG battery that I’m going to install. I want to get this figured out here or on my S5 before I get the RS5 back on the road. Really can’t pass up on shedding almost 50lbs from the car.

#27

Exactly, I am also in the market for one.

I meant use a standard lead acid battery setting (“conv” in the list above?), with the li-ion Antigravity battery.

The reason being I am assuming the AGM setting would regenerate more aggressively as the AGM batteries are meant to allow for it. The idea that they can manage cars with start-stop functionality: in stop and go city traffic they need to recharge effectively at coasting and stopping so they can crank up repeatedly.

768×1024 417 KB

flooded lead acid

vcds

vcds guide

#28

Ok…have a better understanding of the issue with RS Nav and dual clutch gearboxes…apparently it’s a handshake issue where instead of just “listening” the RS Nav unit was trying to talk to the CanBus. So that’s a no go in correlating anything with the AG batteries as the only electrical hookup is the positive/negative and there’s no other interface into the car’s wiring harness to transmit actual data.

#29

I went out and did a bunch of logging with the new code I’m using and so far no issues. Volvofan is on to something though and if he doesn’t post here, I’ll post it up.

Right now he’s using a 1.5V battery and a circuit to trick the BMS into seeing a higher voltage than what really exists. Our goal is to keep voltage below 14.4v. The other thing that works is to just disconnect the BMS in the trunk altogether. There’s a two wire plug. Pull it and you’ll get a constant voltage. Don’t know what the long-term effects of that are.

As soon as I’m done with my maintenance marathon on my car, I’ll graph out my logs and post them. You can clearly see the alternator is producing recharging voltage levels on deceleration.

#30

Here’s volvofan’s post…
" Okay, getting closer to a solution. First of all, I’m going to take the info from my post immediately above and edit into the first post of the thread, so that folks don’t have to dig when referencing this.

Second, I am testing a couple different methods to spoof the Battery Monitoring & Control Module (BMCM, aka J367). I’ve had some success in “static” (aka car idling in my driveway) testing, but haven’t done any “dynamic” testing yet. Will report back. Once again, I’ll update the first post with these results in more detail, but I see the following methods as viable options for addressing this issue:

  1. CODING. This should be your first stop. Try re-coding the BMS with a battery part number that’s closer to the amp-hours of your AG battery (or other LiFePo4 replacement). Make sure you use a new serial number (second part of the BEM). I do NOT recommend trying the EFB or AGM part numbers… even though those types of batteries supposedly are closer to the charging profile of what the LiFePo4 batteries want, I found that I got MUCH worse overvoltage spikes when running that part number. Stick to the conventional ones. PROS: Free (if you have the necessary hardware) and minimally-invasive. CONS: You need to have the necessary hardware (VCDS, OBDeleven, etc), doesn’t seem to work for everyone.

  2. DISCONNECT BMCM. Unplug the two-wire connector at the negative battery terminal. This places the car into a “fallback” mode. Alternator puts out no more than 14.3V, whereas I was seeing up to 15V at idle with it connected. I haven’t done any driving tests to see if this has any other adverse effects. PROS: Free. CONS: You are defeating some of the functionality of the vehicle; this may have other side-effects of which I am not yet aware.

  3. DROP OUTPUT VOLTAGE OF ALTERNATOR. Install a high-amperage diode inline with what goes from the alternator to either the jumper block in the cowling under the hood, or to the main fuse block on the positive battery terminal. Not sure which would be better, as I haven’t tested this yet. The part is on the way, though, and I intend to. PROS: A “hardware” solution for limiting alternator voltage doesn’t rely on software reaction time. CONS: Costs about $25 plus installation time, may have other side effects, inefficient (wasting excess voltage as heat), and reduces alternator output voltage ALL the time (even at lower RPMs when you may not want voltage to be lower).

  4. SPOOF BMCM WITH BATTERY AND DIODE/POT. Put a 1.5V (AA, C, D, doesn’t matter) battery in series with the voltage-sensing lead from the positive terminal to the BMCM. Add either a diode (or diodes) to drop voltage “boost” to the level that gives you the best performance), or a voltage divider/resistor of some sort. This tells the BMCM that the battery’s voltage is higher than it actually is, as I theorized above. I tested this, and it DOES work. Adding about 0.6-1v seems to get the BMCM to tell the J533 module to tell the alternator to chillax and dial back its output voltage. It was a crude test, and I didn’t do any driving to see if this solves the problem under all conditions, but it’s promising. PROS: Relatively inexpensive, simple, retains functionality of J367 module to adjust alternator load to varying conditions without overcharging AG battery. CONS: May take some trial-and-error to tweak/tune, and the 1.5V battery will need to be changed out periodically.

  5. SPOOF BMCM WITH VOLTAGE BOOSTER AND OP-AMP. Same effect as described above… lie to the BMCM. Haven’t tested this yet, but the components come today. PROS: retains functionality of J367 module to adjust load to varying conditions without overcharging, most customizable solution (can use potentiometers to dial in precise voltage gain), no current draw when off, no 1.5V battery to change. CONS: Most complex solution, requires some rudimentary soldering and fabrication skills, costs about $30 in components (voltage booster, op-amp, relay, taps, wiring, board, etc.)

-Jon"

#31

So we’ve done more experimentation and Jon’s built an electronic interface that tricks the battery management system into seeing more voltage than is actually being produced. I’ll be building one myself to test.

In the meantime, I went and unplugged my battery management system from the battery. Takes two seconds, small, two wire plug on the battery’s negative terminal.

VIOLA! No more over voltage on deceleration. I typically see between 13.6 and 14.3. Battery charges as normal and there do not seem to be any downsides as far as I can tell. I did a 50 mile trip in full hooligan mode and no issues. I then ran my typical errand loop, multiple stops at various stores and again, zero issues. Car sits for a week and the battery voltage levels are where they should be. So it seems we’re good to go with that method.

I still want to try out Jon’s contraption as there will be a “soft” code if you were to do a scan with the VCDS and the BMS disconnected. NO DASH LIGHTS with this method however. Have a friend who’s testing on an RS3 and this seems to work for him just fine. Another client with a TTRS has had zero issues with his Antigravity. Go figure.

I’ll post up more info from Jon today, including the schematic for building his spoofer.

1 Like