Honda Insight Forum banner

1 - 20 of 108 Posts

·
Linsight Designer
Joined
·
1,778 Posts
Discussion Starter #1
Howdy folks, here's a video showing a quick, simple modification that prevents the ECM from controlling whether the DCDC is charging the 12 volt battery. This means the DCDC will always keep the 12 volt rail at 14.x volts when the car is running (or in autostop), unless the IMA battery is discharged (because the MCM can still control the DCDC converter). Battery charge voltage is still controlled by the engine temperature. Thus, the DCDC behaves much more like a true alternator, in that if the engine is running, then the 12 volt battery is charging.

The modification is simple:
Step 1: Cut the WHT/GRN wire that connects to the DCDC converter. See video for wire location.
Step 2: Wrap cut leads with electrical tape, to prevent shorting.
Step 3: There's no step 3.

Bonus: That annoying electrical squeal sound will go away, too... it's caused by the above ECM signal.

Questions, comments anticipated and appreciated. More technical information in the linked video.
 

·
Premium Member
Joined
·
446 Posts
Hi Mudder. I watched the video and the last words were "I think it was a poor decision by Honda to have the DC-DC turn on and off". Sorry to be so ignate but what will this do for me, or what will it prevent. I appreciate your effort to add valuable info to the community.
 

·
Registered
Joined
·
4,007 Posts
Hi Mudder. I watched the video and the last words were "I think it was a poor decision by Honda to have the DC-DC turn on and off". Sorry to be so ignate but what will this do for me, or what will it prevent. I appreciate your effort to add valuable info to the community.
The DC-DC maintains the 12v battery at fairly low state of charge (12.8v?). This isn't great for the battery and it doesn't leave much reserve capacity if you need it... ie headlights on, listening to the radio, whatever.

What this behavior does do is save a small amount of power not used to float charge the 12v.

That pretty much sums it up. I wonder what the 12v in the gen 2 does. Did they change to a more conventional regimen, or does it still do this nickel and dime energy-saving?
 

·
Registered
Joined
·
1,558 Posts
Sounds like a super mod for cold weather as electrical draw increases. As indicated in the video, next time in the back area with the cover off, the wire gets cut.:D
__________________
For measuring the ELD Electronic Load Amps , the obdiiC&C for US cars uses parameter '7F' , 'Eld' .
I did a measurement of the relative amounts of increase in electrical load due to fans and lights, towards winter a few years back. This was recorded via a video linked in a post and that was used to make rough estimates of the electrical load at differing settings.
Thumbnail results showing draw from the lights and fans are made from visual guestimates of averages fluctuating on the obdiicc.
In winter I've monitored the ELD displayed on the obdiicc on account of running accessories like heated seats, full fans for defrost, and having the lights often on -- it would be very nice to have more reserve in the 12v battery as it's cranking with the starter nowadays, and starting in the cold is of course essential.

fwiw:The results of my Eld test were:
Code:
"ELECTRICAL LOAD AMPS
[ELD]"	The approximate contribution 
to ELD from different components and settings. 

Condition	       Eld amps 
Key ON	
eld='6' [primary/sec??] eld=6
Sterio.			+1.5
Engine ON, idle	[fi/ign] 4.5
IMA fan ON slow		negligable.
IMA Fan ON Fast		0.5
HVAC Heater mode	
fan speed 1            	11.0
fan speed 2            	12.5
fan speed 3            	15.5
fan speed 4            	18.5
Auto setting MAX       	22.0
LIGHTS	
Parklight/DRL[Cda]	03.5
Headlights low beam	14.5
Headlights Highbeam	17.5
Flash Highbeams	27.5
missing points for radiator fan, rear window defroster, horn, and windshield washers.	Fans on full, headlights on high, with tunes a blaring runs up an ELD of about 50. Flashing the highbeams pulses it to about [63-67] varies. [fwiw:have since had problems with DRL switch.]
Missed in my readings were the rear defroster [woops!!], windshield wipers, horn and?
Not sure how accurate the measured ELD amps are relative to expected power consumption for the lights for example so did a thumbnail estimate: i.e. assume hypothetical low beam bulbs draw ~55 watts at observed an 13.8 Volts. Given amps = watt/volt = 55/13.8= ~4A draw for our hypothetical low beam bulb x 2, =~8 amp draw for the lowbeams, plus more for the tail lamps etc brings the hypothetical total close to the Eld paramater's reported values. But, though I'm satisfied enough it's in the ballpark, I'm curious how accurate the reported obdiic&c amp measurement is vis-a-vis the scope.
 

·
Linsight Designer
Joined
·
1,778 Posts
Discussion Starter #6
Sorry to be so ignate but what will this do for me, or what will it prevent.
As Sam said, this will force the DCDC converter to always float charge the 12 volt battery whenever the car is started. The OEM behavior is to keep the DCDC outputting around 12.1-12.8 volts much of the time, which is enough to run the various 12 volt loads without discharging the battery, but isn't enough to actually charge the 12 volt battery. Thus, the 12 volt battery remains at whatever state of charge it was at before, causing it to die quickly when you actually need it. The worst part is - from what I can tell - the ECM has no clue how charged the 12 volt battery is, which begs the question: why let the ECM control when the 12 volt battery is charged.
 

·
Registered
Joined
·
41 Posts
Hi I have previously cut ( on the MCM side, GRN/BLK ) pin A27 and added it to a FAS that Darkfish sold me, to keep the DC/DC converter enabled for extended engine off coasting ( http://www.insightcentral.net/forums/modifications-technical-issues/60409-plug-play-fas.html )
I am wanting to do your mod to help the 12 volt battery.
So what I need to know is if these 2 mods will interfere with each other or if they will play nice.
Thanks for your help.
Dennis
 

·
Registered
Joined
·
465 Posts
Hate to be a wet blanket, but in fact there is a logic to why Honda engineered the system that way and it wasn't to save electrons. The battery on the Insight is basically designed to be "floated" all the time because it's rarely called upon to do anything... so Honda engineered a charge regime which is optimal for long-life under float. The on/off or pulse charge cycle is actually shown to be best for long life vs a constant float voltage, and high-end modern backup power supplies (UPS) use a similar regime, they let the voltage drop to a defined value, charge back up to a higher value, and let it drop again etc. etc.
 

·
Premium Member
Joined
·
3,421 Posts
Hate to be a wet blanket, but in fact there is a logic to why Honda engineered the system that way and it wasn't to save electrons. The battery on the Insight is basically designed to be "floated" all the time because it's rarely called upon to do anything... so Honda engineered a charge regime which is optimal for long-life under float. The on/off or pulse charge cycle is actually shown to be best for long life vs a constant float voltage, and high-end modern backup power supplies (UPS) use a similar regime, they let the voltage drop to a defined value, charge back up to a higher value, and let it drop again etc. etc.
You really going to argue with mudder? :)

How do you explain the poor life of the Honda IMA 12V that is "rarely called upon to do anything"?
 

·
Registered
Joined
·
465 Posts
You really going to argue with mudder? :)

How do you explain the poor life of the Honda IMA 12V that is "rarely called upon to do anything"?
I'm only stating what I know from dealing with UPS's that in fact did go to that charge regime. It doesn't mean I know more than mudder about anything else.

The 12 volt battery in my Insight is now almost 7 years old and works perfectly. It was the cheapest one available at that time. Batteries in UPS systems which are constantly float charged have a life expectancy of 3 of 5 years typically. Batteries in typically automotive use also have a life expectancy of 2 to 5 years.
 

·
Linsight Designer
Joined
·
1,778 Posts
Discussion Starter #11 (Edited)
I'm not perfect and I invite anyone to critique anything I write anywhere... academic review is critical for true knowledge, and you have a well laid argument. The last thing I want is for nobody to question what I write, and then write something that is wrong that passes as scripture. Thank you for questioning my statements! You may very well be correct, and thus this warrants further discussion.

Exhibit A:
Continuously float charging lead acid batteries requires a well regulated, temperature compensated charger. If those conditions are met, float charging is the preferred method in most literature I've read. Fortunately, the DCDC converter is well regulated and modifies the output voltage based on engine temperature (which is roughly battery temperature, too). Failure to decrease voltage ~-3 mV per degC while topping is what kills lead acid batteries, particularly at extreme temperatures.

Exhibit B:
Vehicles with alternators always maintain the same voltage setpoint. Some adjust to temperature, but all alternators modulate voltage output by siphoning current into a field winding proportional to the actual current demand and desired voltage. In short, vehicles with alternators do what my modification has done.

Exhibit C:
Keeping the battery maximally charged minimizes sulfation, which accelerates whenever a cell is less than 2.3 volts @ s.t.p (13.8 volt across battery @ s.t.p). The DCDC keeps the battery at ~2.38 V/cell @ s.t.p., which is right in the sweet spot between undercharging the battery and gassing the water out. The battery likely never passes this topping charge region (into true float charging), unless you drive the car thousands of miles at once.

Exhibit D:
Pulse charging to reduce sulfation is maybe what you're talking about. This is a highly contested topic that actually lacks definitive results, hence there are two camps on the issue. However, if a cell is kept at a high enough voltage in the first place, sulfation rates are greatly reduced, hence pulsed discharging isn't required. Maybe this isn't what you're talking about.

It's possible you're talking about a hysteretic charge algorithm, but those are only beneficial when the battery isn't placed under load for extended periods. This method doesn't apply to insight because there's a constant load on the battery whenever the car is off, whereas UPS systems might only activate once per year. The main reason this algorithm is used is in lieu of a sophisticated charge algorithm that reduces float charge voltage after the topping phase is complete. Another reason is to make sure the battery is still working (most UPS systems I've used self test for this reason).

Exhibit E:
Insight's algorithm probably isn't to provide occasional load... based on my observations, the DCDC is disabled nearly every time I let off the gas. If the algorithm were truly as you've stated, then I propose the DCDC would be turned off every N minutes/hours/etc, not based on driving conditions.

Exhibit F:
The modification I propose only changes the DCDC output voltage, but doesn't actually disable the DCDC. Because the voltage is set to 12.x V when the ECM is connected to DCDC, the large load you've proposed would actually be sourced from the DCDC converter, as the battery cell voltage drops pretty quickly as a function of load, whereas the DCDC is going to maintain exactly the same voltage regardless of load*. For example, my car's 12 volt battery quickly drops below 12.5 volts when I place a 10 A load on it... that's an argument against pulsed discharging.

*The DCDC actually increases output voltage under increased load, elbowing around 15 A. For example, all things otherwise equal, the DCDC I tested output:
13.94 V @ 0 A
13.97 V @ 15 A
14.00 V @ 20 A
14.02 V @ 22 A
14.02 V @ 40 A
14.02 V @ 70 A

Exhibit G:
The DCDC isn't disabled when insight is using a lot of current (headlights on, etc). For example, if I drive 2000 miles with the fan on full blast, the ECM will never toggle the DCDC to output 12.x V.

I do believe Honda's only reason was efficiency... the DCDC converter consumes 10 W when it's on and not outputting any current. Further, with a 2 amp load, it's only 62% efficient. Maybe Honda really just wanted to eek out every bit of juice from the IMA battery? I admit 10 W is inconsequential in the grand scheme, but I'm just not convinced they were trying to preserve 12 volt battery life.

I'm open to further discussion. I live for discussion, even when I'm wrong.
 

·
Linsight Designer
Joined
·
1,778 Posts
Discussion Starter #13
I suspect we need to further define battery charging states.

Topping is the second stage, and requires a higher voltage than float charging. I propose that automotive batteries rarely enter float charge mode due to drive cycle... car is off and under load. Car turns on and pulls current until DCDC turns on (after car starts). Car is then charged at constant voltage, never transistioning from topping to float charge.

Good battery tenders enter a float charge mode to prevent boiling the electrolyte out the battery... I propose that this isn't an issue while driving because the battery doesn't transition from topping to float state during most driving conditions.

I think if Honda were truly implementing a float charge algorithm, then they would lower the voltage to 13.4 volts, not 12.x. At 13.4 volts, you can charge a lead acid battery forever.

Constantly charging/discharging the battery - as the OEM system does - likely accelerates wear much faster than the possibility of entering the float region on a VERY long trip, with constant low DC load.

Maybe the jury is out, but I'm keeping my wire cut for now ;).
 

·
Linsight Designer
Joined
·
1,778 Posts
Discussion Starter #14
What about if you replaced the lead acid battery with cells from the hybrid pack like eq1 did?
If I remember correctly, eq1 was using 12 cells in series? That's 14.52 volts at 1.05 V/cell. Since the DCDC is outputting in constant voltage mode, the output current is unregulated, unless you hit the DCDC's maximum output current limit, at which point it'll decrease voltage to shed load demand. In short, CV charging NiMH continuously is never a good idea, as the recombination process that occurs inside each full cell is heat-limited... continuing to push current faster than the cell can shed heat is a recipe for disaster. To properly float a NiMH cell, you need to provide ~5% nominal cell capacity or less (~250 mA or less in insight).

Even without my mod, replacing the lead acid with a NiMH battery could be problematic, as the DCDC could quickly overheat the NiMH cells. My mod would make this more likely to occur, although the problem exists regardless... insight wasn't designed to use a NiMH battery.

For example, if eq1 drove 1000 miles with the AC running, both the OEM and my proposed modification would always keep the DCDC converter outputting a voltage that could potentially overheat the NiMH cells.
 

·
Linsight Designer
Joined
·
1,778 Posts

·
Registered
Joined
·
972 Posts

·
Administrator
Joined
·
10,698 Posts
Guess which wire i'm cutting today?

I've always hated the way the insight abuses the 12v battery.
but couldn't be bothered to do anything about it.. till now..
 

·
vote 4 mawah
Joined
·
275 Posts
12v

I just commented about eq1's 12V mod in my other post up a minute ago. IIRC like, I believe eq1'a mod is using 10 cells of the 12 cell stick pair. I also believe that this configuration would make the stick pair more vulnerable to the overcharging potential, { volts in #=! cell capacity }, mutter is sighting in this thread.

I'm surprised that the age and condition of the battery being discussed is not taken in account in most of the calculations and formulas I've read.
Are we supposed to take foregranted that the battery being discussed is new?
And/Or how much different would a battery that is 10 years old,
for the sake of argument,
behave under the same charge/discharge conditions.

I for one treat my 14 year old lead acid a lot different than I treat my 5 year old lead acid, especially during the cold winter months.
 

·
Registered
Joined
·
6,004 Posts
^10 cells. They never get overcharged, they get undercharged - maintained at a low capacity, probably like 1 to 2 Ah... 1.38V per cell isn't high enough to charge enough...

[edit: my latest calculations and tests put it at about 4-4.5 Ah, with decent sticks... I've done this a number of times and I'm now quite confident in this value.]
 
1 - 20 of 108 Posts
Top