(Was) Honda Civic Hybrid goes 100,000 miles with no problems

[peterperkins]

You would be better off with IMAC&C initially to overcome the latest programming changes and manage your battery. Then you can command assist/regen as you want. Combine it with the current mods later when funds permit and you are laughing. 8)

 

[fechter]

Woo hoo! new pack arrived.
I got it from Jeff Bierman at Hybrid Battery Works http://hybridbatteryworks.com/
It turns out that Jeff and I went to the same Junior High and High school in Salt Lake City (but I was a few years ahead of him).
Pack came in 2 days from Salt Lake to San Rafael, which I thought was pretty fast. Install went uneventfully and finally I don't have to look at the stupid IMA and Check Engine lights all the time. So far it works at least as well as the original battery when it was new. I'll do more testing as I drive next week.

Now to think about hacking the current sensors...

 

[mwkeefer]

what if I may ask did that little pack set you back? Weight?

-Mike

 

[Toorbough ULL-Zeveigh]

i'm guestimating $20/cel times 220 cells.
NiMH D cells typically weigh-in at about 170 grams.

say fecht do you know if he sells individual cells or sticks?
i could use a half dozen rocket packs.

as to your current sensor hax, have you ever considered playing with rogowski coils?

 

[fechter]

what if I may ask did that little pack set you back? Weight?

-Mike

$2,000 plus shipping, total was under $2,200. Weight is 76 lbs. The weight limit for UPS is 80 lbs.

say fecht do you know if he sells individual cells or sticks?
i could use a half dozen rocket packs.

as to your current sensor hax, have you ever considered playing with rogowski coils?

I asked him if he would sell raw sticks and he said yes, but it would be nearly the same price. Cells are supposedly manufactured in Japan and assembled in Hong Kong. These are the same sticks used by BumbleBee Battery.

So far the pack works great. The real test is how long it will last. Note to myself: odometer reading 162,000 at install.

Rogowski coil is very interesting but only for AC. The IMA system uses a hall effect sensor on the battery line and 3 more for the motor phase currents. Caveman approach would be to run a fat wire that bypasses the sensor such that 20% (or whatever) of the current goes around the sensor. I think they all need to be within about 1% of each other or else the computer will think there's a short somewhere. Guys on Insight Central have this mostly figured out and you can put a divider on the signal wires to get the same effect.

 

[steveo]

Pack looks good fechter...

welcome to the hybrid modding...

Peter on Insight Central is the Hacking Guru...

I currently have the 30% current mod running in my insight.. with mima...40ah a123 pack... .. my milage over the last 300km is an average of 1.7 liters per 100km or ... about 130ish mpg ... My goal this tank is to hit 1800km+...

my last tank at 35liters used i averaged 2.4 liters per 100km.. however i noticed that one of my sensors on my intake was not connected properly in the engine bay and affecting the fuel consumption of the motor...

I used in the ball park of 75-100kw of energy over my last full tank... or say $6 of energy

-steveo

 

[peterperkins]

The IMA system uses a hall effect sensor on the battery line and 3 more for the motor phase currents. Caveman approach would be to run a fat wire that bypasses the sensor such that 20% (or whatever) of the current goes around the sensor. I think they all need to be within about 1% of each other or else the computer will think there's a short somewhere. Guys on Insight Central have this mostly figured out and you can put a divider on the signal wires to get the same effect.

It actually has five sensors you need to fake/hack.

1 x battery current
1 x motor current
3 x phase current

The sensor bypass wire idea is fine if you can calibrate it properly and they do need to be quite well matched or it will throw an IMA error.
The bypass wire method also needs a physical mod/bypass at each sensor which is a PITA. I prefer tweaking the sensor signals.

The HCH1 seems less capable of tolerating higher currents than the G1 Insight setup and it will throw errors with regen much over + 15%
It can tolerate higher currents for assist as long as the motor is turning at over 1000-1500 rpm perhaps upto about +35%

I'm just finishing a special two stage current hack pcb for the HCH1 and G1 Insight you might want to consider that
Details in the threads on IC.

http://www.insightcentral.net/forums/modifications-technical-issues/16206-inside-g1-mdm-more-power.html
http://www.insightcentral.net/forums/other-honda-hybrids-discussion/24340-hch1-inside-ipu-more-power.html
http://www.insightcentral.net/forums/modifications-technical-issues/25553-parallel-igbt-ipu-more-power.html
http://www.insightcentral.net/forums/other-honda-hybrids-discussion/21887-imac-c-hch1-civic.html

Sounds like you need manual IMA control now as well to get the best out of your new battery!

 

[fechter]

Thanks Peter,

I see 5 sensors on the Insight setup, but only 4 on the HCH. Is there one I'm not seeing?

Yes, if I were to try bypassing the sensor it would be hard to get them all within 1%, but would eliminate the possibility of the sensor saturating. I was thinking about passing current through them with a power supply and using a clamp-on DC meter on the bypass wire and trimming to get within 1%. Resistors on the output does sound a lot easier though.

10-15% boost might be all I want most of the time. Full control with a MIMA setup would be great for me, but maybe not so good for the wifey.

BTW, so far with the new pack, the gas mileage numbers are up to even better than when the car was new even with the lame software update. I'm very happy about that.

 

[peterperkins]

The HCH1 battery sensor is on the battery switch board.
The four other sensors are in the grey metal IPU box.
The sensors won't saturate until the motor or IGBT explodes they are all +/- 200A IIRC

You can turn off the manual IMA control with my IMAC&C if you don't want it active when the wife is driving.

 

[fechter]

Ah, I see the motor current sensor now.

Actually, I'd like to try adding the resistors to divide the signals coming out of the sensors to boost a modest amount, like 20%. I spent hours reading the threads on Insight Central, but haven't exactly sifted out what needs to be done. It's a lot like reading long threads here.

One reason I'm leaning toward the resistor hack is simply cost. I'm pretty broke after buying the new pack. I think it will be easy enough for me to play with once I figure out the resistor values and exactly how to tie them in. The motor and phase sensors all look alike, so I assume they have the same pin configuration. I could do little perf boards on each sensor cable and heat shrink them.

I think someplace you had a schematic of the dividers and some clues as to the wire colors coming out of the sensors. Ground, 5v and signal from what I gather.

After my second day of commuting, my gas mileage is better than I have ever seen it and the pack temperature after the hills is significantly lower.

Here's my motor control unit: 15kW bike controller anyoune?

 

[peterperkins]

Those wanting to try this mod will be able to buy a pcb for the current mod in July.
The PCB I am just finishing can have several configurations depending on how it is built.

1) Permanent current increase as determined by resistor matrix.
2) Temporary current increase as determined by resistor matrix with addition of simple activation switch.
3) Permanent current increase as determined by primary resistor matrix and temporary further current increase as determined by secondary resistors and activation switch e.g. +10 and +30%.

The PCB will interface with the car in the IPU compartment via connections to the MCM & BCM low voltage wiring near the MCM/BCM connectors. It does not require any wires to be cut, but you do have to splice into several wires (preferably with soldered connections for reliability).

I have been working with Matt (Hybrid revolt) on these mods so he will be my US/CAN supplier for these pcb's.

My price will be $25 for the blank pcb.
Matt will either send out blank or built PCB's at additional cost as reqd.

Further details inc schematic etc are available on the IC forum.

The calculated resistor value matrix attached is provided to help those wanting to experiment now.
Please carefully note the warnings in the pdf.

1) Do not use more than +10% current without upgrading the IMA fuse.
2) Any mods at your own risk.
3) Enjoy the extra power of your sleeper car!
4) Best used with a good battery. Steveo and I have Lithium but a MaxIma pack should also give good results.

http://www.insightcentral.net/forums/attachments/modifications-technical-issues/4703d1368132587-inside-g1-mdm-more-power-mcm-current-increase-resistor-matrix.pdf
http://www.insightcentral.net/forums/attachments/modifications-technical-issues/4747d1368777978-inside-g1-mdm-more-power-relay-current-hack-050513.jpg
http://www.insightcentral.net/forums/attachments/modifications-technical-issues/4869d1369563521-inside-g1-mdm-more-power-ima-curent-hack-1-1-.pdf

PS Don't do the resistor mods inside the grey box it's electrically very noisy in there. Do them near the MCM.

 

[fechter]

Awesome! Thanks Peter.
You saved me hours of reading :wink:

Now, off to look for resistors...

BTW, I'm noticing a gradual improvement in my new pack's performance as the cells get cycled during driving. I think this is typical for most new batteries. Now I can actually make it all the way up the big hill before the boost drops out.

 

[fechter]

OK, I found the resistors I needed to make a matrix board. I checked all of them with a good Fluke meter first to make sure they're all within 1% of the desired value. I used 2.2k and 510R for the resistors. I'm going to try it with the stock fuse (but I have a spare one just in case).

I just soldered them on a piece of perf board. About a cheap as it gets.

 

[fechter]

Finally got time to hook up the resistor matrix. I spliced into the wires coming off the motor control unit (MCU). Most of the connections are on connector "C":

I stripped a short section of the wire and soldered on the wire going to the matrix board.

Here's how it looks after making all the connections:

Not having any nice project boxes handy, I made a housing out of electronic "fish paper", which is heat resistant and fairly rugged. The covered matrix board just hangs by the wires.

So, after installing the board, I was eager to give it a try. I decided it would be good to leave the compartment open until it was tested in case there was a problem. When I went to start the engine, the dash warning lights lit up like a Christmas tree and the secondary starter kicked in. Oh crap! I got out my code reader and only saw a P1600. WTF? It turns out I forgot to turn on the main pack switch! Well, at least that was easy to fix...

With the pack turned on, things worked amazingly better. It went through a re-cal on startup, so I assume one of the connectors I unplugged also reset the IMA computer. After the initial recal, it went back to being normal. I only had time for a short spin around the block, but I can definitely notice the change, especially on regen. I'll know more after driving for a week and I can really test it out.

Total cost: $0 (I had everything in the junk pile). If you had to go out and buy everything, maybe $2 plus shipping.

Thanks Peter!!

 

[peterperkins]

Nice. Now you need an IMAC&C to control the power and an OBDIIC&C to see the actual current/battery temp/soc etc etc 8)

 

[steveo]

join the club Richard!!!

looking back at what i've done...

i aquired my insight last march...

since then i've done the following ... yes all of it...

-Refurbished the engine mounts
-put a new 1/2nd gear sync row and coller & bearings..
-new clutch kit
-changed the rear struts with monroe struts... cut chevy aveo rear springs
-installed new interior carpet, new driver side seat cover
-Refurbished the original IMA battery
-Purchased MIMA Plug & Play
-Bcm fooler
-bcm Interceptor
-PP50 Open Source Bms
-Obd2c&c gauge
-Upgraded Ima fuse to 125amp
-Installed 50s2p a123 cells
-30% Current mod
-Installed 172v PSU charger inside the Oem battery case
-Window Tints
-Car Deck
-Rosta Cruise control
-New Bridgestone Potenzas re92

$$$$$$$$$$$$$$$$$$$$$$$

however .. the way i look at it ...

every 2000km.. i save 150$ - 180$ in fuel compared to my old car... yes that much savings...

-steveo

 

[fechter]

Nice. Now you need an IMAC&C to control the power and an OBDIIC&C to see the actual current/battery temp/soc etc etc 8)

Looks like that may be the next step.

As far as I can tell, the resistor matrix has actually decreased my gas mileage slightly. It's hard to tell since there are other factors that changed, like warmer weather and using the AC more or letting the wifey drive the car (that totally kills the mileage). I might try removing the resistor board and see if mileage goes back up.

 

[peterperkins]

The current hack will give you more performance not more fuel economy unless you either use IMAC&C as well to control when you use it or plug it in at night to make up the extra current used.

That extra current used during acceleration has to be paid for by increased background regen

 

[fechter]

Yes, I see how that works now. With the hill profile of my commute, I frequently arrive back home with 100% SOC, so grid charging won't help unless I can discharge more.

I DO notice the extra performance and going uphill on the freeway I have much more pull. I also notice that the additional regen allows me to frequently come to almost a complete stop without using the friction brakes.

 

[peterperkins]

Build an HCH1 IMAC&C on stripboard, download the free demo code, then arrive back home with 0% Soc a smile on your face and see some real mpg gains. :lol:

 

[fechter]

So far, so good. The new battery from Hybrid Battery Works is still performing like new, which is significantly better than the stock battery was when new. No recalibrations have ever happened as far as I can tell. If the IMA computer is expecting a lower capacity, it should keep this pack well within the 20-80% DOD healthy range. So nice to have the car run normally again.

 

[steveo]

So far, so good. The new battery from Hybrid Battery Works is still performing like new, which is significantly better than the stock battery was when new. No recalibrations have ever happened as far as I can tell. If the IMA computer is expecting a lower capacity, it should keep this pack well within the 20-80% DOD healthy range. So nice to have the car run normally again.

Hey Fechter

Do you have a 5 speed or cvt?

Whats you avg fuel economy per tank ? How many miles to you get per tank

Thanks
Steveo

 

[fechter]

Mine has the CVT. Actual mileage (divide gallons put into tank / miles driven) is somewhere around 43mpg.
My commute is 20mi each way with significant hills and some city driving. If the wife drives the car, the mileage plummets to under 40mpg.

The stupid MPG indicator on the dash always reads much higher than the actual mileage. I've gotten that up to 49mpg, but the number is exaggerated.

The tank is 13 gallons, so I could possibly get over 500 mi on a tank. Car now has 165,000 miles on it.

 

[dnmun]

i missed this one at my fav auction 3 weeks ago. 2003, 148k, sold for $2100 and was in good condition:

http://auctions.speedstowingauction.com/2003-Honda-Civic-Hybrid_i16830320

 

[steveo]

Im really in the fence to get a cvt civic to replace my 98 civic si

There is one local to me for just under 4k with 221k

Prob could get him to the 2k mark

But when i think of this gen hybrid all i think about is cvt problems

And thats what really holding me back to grab one

My other option is 2005 prius.. I really like the fact i can run electric mode only

But the car is heavy imo at over 3000 lbs compared to my 1800lb insight!

Have you had any cvt problems?