Honda Insight Forum banner
561 - 580 of 977 Posts

·
Linsight Designer
Joined
·
2,647 Posts
Discussion Starter · #561 ·
The Nissan CIMA is the JDM version of a Infiniti q70 hybrid that has been sold stateside since 2014.

According to the Infiniti First Responders Guide, it has:
Li-ion battery voltage 346V (400V max.)
Number of Li-ion battery modules in the pack 12
Li-ion battery module voltage 28.8V each
Li-ion battery dimensions 33.35 x 17.83 x 15.43 in. (847 x 453 x 392 mm) 121.28 lbs (55 kg)

The next step is to find one for parts!

Edit: the Infiniti M35h used the same battery beginning in 2012. Also according to the Infinity First Responders Guide.
Double Edit: so did the Q50.
Excellent research, particularly for a new member (welcome!)!
I looked into the available donor Infiniti vehicles. Unfortunately, there are only a handful available and they're expensive! Looks like I'm back to contemplating the 4S1P Leaf option.
 

·
Linsight Designer
Joined
·
2,647 Posts
Discussion Starter · #562 ·
Personally beyond proof of concept testing (which i admire) i don't see any point having a lithium pack of equivalent size to the OEM one.

If you have Linsight you will have the ability like IMAC&C to manually control the motor and grid charge the pack.

Surely the idea is to put at least 10-30ah into the car to make use of cheap electricity and manual ima control for high mpg.

I doubt an oem size lithium pack and Linsight to manage it would be cheaper than a 3 year warrantied nimh pack from one of our sellers on here.
I'm hoping to find a lithium option that is cheaper than NiMH. If we can procure a lithium pack for less than $1095, then Linsight will be cheaper than NiMH.

Yes, Linsight as I've proposed it provides manual IMA control, which is still useful for OEM-sized packs, and would still be useful with lithium, too. My major goal is to sell Linsight PCBs (surprise), which means I need to be a better option in all ways - including price - compared to NiMH solutions. The addressable market for huge lithium packs is much smaller because the battery cost is too high for the average owner.

Fortunately, Linsight supports any CCCV rechargeable cell chemistry between 1 and 5 volts, which is nearly every modern battery technology I know of.
 

·
Super Moderator
Joined
·
9,229 Posts
I looked at the video again trying to visualize how the bus bars would be attached. A question arises, are the remaining copper "stubs," which will need drilling, spot welded to the prismatic cell flap terminals, or is that the arrangement?

I assume one would have to use very small screws and nuts to attach the required bus bars?
 

·
Premium Member
Joined
·
2,949 Posts
Fortunately, Linsight supports any CCCV rechargeable cell chemistry between 1 and 5 volts, which is nearly every modern battery technology I know of.
Personally I think this is the elephant in the room .. What battery is chosen as the 'default' .. is just the cherry / sprinkles on top.
 

·
Linsight Designer
Joined
·
2,647 Posts
Discussion Starter · #566 ·
Can we assume that Linsight will work with the OEM NiMH battery pack?
I hadn't thought of using Linsight with the OEM pack... hmm, the BMS won't work at all because the maximum input voltage the LTC6804 supports is 5 volts between taps. Since the OEM NiMH doesn't have voltage taps at each cell, the input voltage at the stick ends is too high to measure. Sure, I could make Linsight 'work' with the OEM NiMH cells, but it wouldn't be able to measure stick voltage. It can still measure overall pack voltage, but that's not great. I'm really designing Linsight for lithium cells and assume anyone buying it will switch over to a lithium pack.
 

·
Registered
Joined
·
163 Posts
Wow, I'm a bit late to the party here but I'm definitely interested in buying one of these setups. If we could get the entire conversion to cost in the $2500 range that would be awesome.
 

·
Premium Member
Joined
·
636 Posts
What'd ya think - 6 of these?:
Turnigy Graphene Professional 10000mAh 6S 15C LiPo Pack w/5.5mm Bullet Connector

6S, 22.2V nominal, 10Ah, 150A constant discharge, 50A charge, $129 each pack X 6=$774... 36 cells total, 133.2V nominal...

They also sell a 16Ah one for like $190...
I'd worry about some of their claims, but if anyone want to try them, let us know how good they perform.

"Maintains higher pack capacity even after hundreds of cycles" not thousands?
"Longer Cycle Life 600+" That's not a lot of cycles, specially for our cars
"Fast charge capable, up to 15C on some batteries." Some batteries means inconsistency on their batteries.
 

·
Registered
Joined
·
7,843 Posts
I'd worry about some of their claims....
"Maintains higher pack capacity even after hundreds of cycles" not thousands?
"Longer Cycle Life 600+" That's not a lot of cycles, specially for our cars
600+ cycles seems about right; from what I recall, the Insight stock NiMH cells are 'only' about 500-1000. I'd still worry about cycle life, but I don't think the 600+ metric in itself isn't something to worry about.

"Fast charge capable, up to 15C on some batteries." Some batteries means inconsistency on their batteries.
I think that statement is just marketing-speak for batteries (i.e. packs) that have parallel cells. Non-paralleled packs supposedly do 5C. What's the real-world figure? Not sure. My guess is the packs can do 5C charge burst without going up in flames. But you wouldn't want to subject them to a constant 5C charge all the time...

Overall, I'm leery of the charge performance of any of the 'Lipo' cells, whether the C rating would seem to cover Insight needs or not. Of all the specs for hobby-type packs I've seen, though, these are getting closer... Think about it: 10Ah, 5C charge max, that's 50 amps, OK?. The Insight in stock form maxes out at about 50 amps every so often - but that's with a pack of 120 cells that have at minimum something like 3mO IR, a total of 360mO. With these Turnigy cells having an IR of say 1.5mO (the spec says 'as low as 1.2'), we'd be looking at only 54mO... So, it seems like we'd be easily able to get stock charge performance at a much lower current rate. And we'd be able to adjust the max current anyway, with Linsight, right?...

And, overall, I worry about 'Lipos' in general, the chemistry and potential for catastrophic failure...

Not sure how I feel about the packaging and the need to build the pack/connections. You'd probably have to build a hard case of some sort, to protect the cells/modules, unlike with the Leaf cells, for example. You'd have to figure out what to do with connections - and you'd have to either trust that the stock wiring and connections are good enough or take it all apart and redo them - which is a major hassle (at least for me - I've taken one 4 cell Lipo battery apart and tried to work with one cell, it was quite a mess)...

What's attractive is that I could plunk down not-too-much $ and have new cells at my door, unlike the Leaf option...
 

·
Registered
Joined
·
958 Posts
Queen Green

Check out the green color of the Queen's outfit (Queen of England) in the news today. This bright fluorescent green would be a good color for the ink on the Linsight sticker.
 

·
Linsight Designer
Joined
·
2,647 Posts
Discussion Starter · #573 ·
Periodic Update: I've gotten quite busy lately consulting on a few projects. I've hardly touched Linsight and unfortunately it's on hold for now, until things settle down. Linsight is still happening, just not anytime soon. Gotta pay the bills ;).
 

·
Linsight Designer
Joined
·
2,647 Posts
Discussion Starter · #574 · (Edited)
tl;dr: Linsight can now send all the IMA error codes the OEM system can send... they'll show up on the OBDII port and are useful for troubleshooting down the road.

...

Update to my previous MOTFSB post:
Bitbanged the MOTFSB line to determine the bitstream that causes the ECU to display each Pxxx error code. If anyone is interested in the MOTFSB bus, here's what I've deduced:
MOTFSA is a 24 Hz clock.
MOTFSB is a single ended 0:5V data line.

The MCM sends packets on MOTFSB to the ECU. The ECU can only listen. To successfully send a packet, you must repeat each packet twice in a row; sending only one packet results in a check engine light.

The ECU only looks for VERY specific packet types, taking the following form:
-(S) A high start bit, which must occur at least 10 clock edges after the last high bit; you must hold the line low for at least 10 clock edges before broadcasting a packet or the ECU will generate an error.
-(DDDDD) five data bits, indicating the Pxxxx code
-(MMM) mask, which is always '0b110'; ECU ignores all other packets (no CEL).
-(C) a low stop bit. You must hold the line low for at least 10 clock edges between packets or the ECU will generate an error.

Using the above (arbitrarily chosen) lettering, an error packet must be as follows:
0bSDDDDDMMMC

As mentioned above, you must have at least 10x zeros before and after each packet, and you must send each packet twice, so the actual stream is:
0b0000000000SDDDDDMMMC0000000000SDDDDDMMMC

More specifically, all valid packets that result in a check engine light take the following form:
0b1DDDDD1100

Moreover, no check engine light occurs if ANY other packet is broadcast at least once every 5 seconds. Thus, you can prevent a check engine light by sending any packet (twice) as long as it doesn't follow the above syntax. For example, no check engine light occurs with the following stream:
0b11111111100000000001111111110000000000

Now for the Pxxxx codes:
---------------------------
0bDDDDD - Pxxxx code
---------------------------
00010 P1449
00011 P1647
00100 P1438
00110 P1584
00111 P1565
01000 P1635
01010 P1586
01011 P1580
01101 P1568
01110 P1582
01111 P1440
10010 P1448
10011 P1581
10100 P1445
10110 P1583
10111 P1572
11000 P1439
11001 P1573 (no CEL)
11010 P1585
11011 P1576
11100 P1577
11101 P1648
11110 P1649 (no CEL)

...

Peter probably figured this out a long time ago.
 

·
Registered
2000 insight mt
Joined
·
185 Posts
Mudder,

Thanks for figuring this out and sharing. I also appreciate your forethought with have error messages sent through the obd. I am sure that will serve me well down the line.

I am looking forward to updating my aging beast with a new ev soul! I'm saving my pennies for linsight.

Keep up the good work!
 

·
Linsight Designer
Joined
·
2,647 Posts
Discussion Starter · #576 · (Edited)
Today I revisited how I plan on installing Leaf modules into the IMA battery bay. We only need 9 modules now that I've figured out how to rewire the Leaf modules 4S1P. I'm in talks with another IC.net member to sell modified Leaf modules as shown in that video, and then assembled into a 36S1P "Linsight Leaf Pack" (QTY9 Leaf modules rewired as 4S1P creates a 36S1P pack). The LLP is 4.5 kWh (4.0 kWh usable), which is four times more capacity than a brand new NiMH pack. Not bad for a solution that weighs the same as OEM.

The LLP fits nicely in the IMA bay and leaves enough room to reuse the existing junction board. This is super awesome, as it means Linsight is SUPER easy to install. I've created a proof of concept video showing roughly what a Linsight IMA battery bay looks like:
Direct link for the non-Flash crew:
https://youtu.be/ACQRUa22DOQ

I haven't created the mounting bracket everything attaches to yet, but it will look something like this:
Line Diagram Ceiling Transparency Rectangle
The junction board mounts to the bracket at the bottom and the OEM IMA battery bay cover at the top (via the 8mm bolt near the circuit breaker). I'm hoping I can get Scott to make this bracket once I get final specifications. If not, I'll make them myself, but it's just one more thing preventing me from releasing Linsight.

Summarizing the project thus far, Linsight kits will come with the following items:
-QTY1 Linsight PCB
-QTY2 low profile fans (and mounting hardware)
-QTY4 temperature sensors (to monitor batteries)
-QTY1 200W constant current LED driver (a.k.a. "grid charger)
-QTY4 high current cables (connect to junction board)
-QTY1 Low profile USB cable (for firmware updates)
-QTY1 Linsight User Interface module (to control Linsight)
-QTY1 mounting bracket (secures batteries and junction board to frame, see picture above)

Note: QTY9 Nissan leaf lithium battery modules aren't included with the Linsight kit; you'll need to source them elsewhere and they'll need to be rewired 4S1P. As mentioned above, I'm in discussions with an IC.net member to provide complete "Linsight-ready" battery modules. If nobody takes this subproject on, then I'll do it myself, but I'd rather have someone else do this part due to my general lack of time.

If you're converting your own Leaf modules to 4S1P, you'll need the following parts (I'll offer a 'DIY battery' ordering option for an additional fee):
-QTY36 spacers to properly space cells
-QTY8 low profile sex bolts
-QTY4 all thread pieces
-QTY9 'A'-type copper bus bars
-QTY9 'B'-type copper bus bars
-QTY27 BMS leads with eyelets
-QTY2 compression end plates (from the Leaf)

...

You could forego modifying Leaf modules to 4S1P, but then you'd need 18 modules (see post 1) and you'll run out of room in the IMA battery bay for everything else... you'll also be adding 75 pounds to the vehicle, whereas the LLP configuration I proposed above doesn't add any weight at all.

...

The junction board still fits with a 48S1P pack (12 Leaf modules, rotated 90 degrees). The four high current battery leads are much longer in this configuration, but that doesn't cause any problems. I'll add alternate battery mounting tabs to the same mounting bracket (allowing the same bracket to mount either a 36S or 48S pack).

...

Parting comment: Once I finish Linsight, I'll do a complete video walkthrough - start to finish - with every single step required to convert an OEM insight to Linsight.
 

·
Super Moderator
Joined
·
9,229 Posts
Very nice update. Looking forward to seeing some H/W in Indianapolis. Now I just have to study up some more on what I have to do myself to build the 12 module version.;)

It would help the diy folks if you could walk through a module modification at Indy.
 

·
Registered
Joined
·
739 Posts
What is the operating voltage range of this proposed pack?
136.8v nominal
vmax is about 148v (slightly above that can cause damage or at a minimum rapid wear)
Vmin is around 108v but realistically could sag as low as 90v without damage though there is no reason to risk that area of the discharge curve.

On this topic since I remember brief discussion of minimum and max voltage that could go into an insight now that Mudder has graciously replaced certain control modules out of the car,
what were they?

I have a 96v LiFe pack and a 192v NiMH pack but I think both would be inop under the old controls, the insight has something like a 400v IGBI which seems like it could handle 200-220v and obviously without the stock control systems in the way should be able to handle any lower voltage.

Does Mudders device open up the top and bottom ends to pretty much anything or are we still in a very narrow voltage range? (if the voltage could be anything I forsee a 240ahr 48v pack going in my car)

(Sorry if I am being dense, I have held off on repackaging my NiMH cells into my project insight to wait for mudders device to see if I can get the whole enchalada)
 
561 - 580 of 977 Posts
Top