Honda Insight Forum banner

241 - 260 of 268 Posts

·
Registered
Joined
·
7,165 Posts
Discussion Starter #241 (Edited)
Please understand I'm not trying to be argumentative, but rather just contributing to the knowledge base.
I totally understand, no problem here with your follow-up questions...

I'm not sure what you mean about "pre-mature empty or full signals". Wouldn't this imply an imbalance, if it's at the tap level?
There would be an imbalance, but the imbalance itself is not what triggers the code - rather, most often it would be a cell being truly empty when the rest are well-charged, near '75%'.

In your case, with a P1449, probably a P1449-78, it'd go like this: at least 1 cell is nearing empty, that cell's voltage slope is tanking fast, the BCM measures that slope and sees that a cell is empty - so it considers the whole pack empty. It does a 'neg recal' (BAT gauge plummets to bottom). It then starts a forced-charge - but since the other cells are near full, the BCM can only charge the pack a small amount before those taps reach the 'full' (75%) threshold/voltage. If it can't charge more than 10% of the nominal capacity (10% of 6500mAh) you get that P1449-78 (I've actually seen closer to 20%, where the charge was near 20% and I still got the code)...

So I guess you can look at this either way - as 'pre-mature empty' or as 'pre-mature full'. It's "premature" because normally you'd want and expect all cells to be empty and full at about the same time, so you can use the full capacity range.

You can operate that pack 'forever' with say a tap voltage imbalance up to perhaps 1.2V, but it will only cause a problem, the car will only throw a code, when that voltage imbalance reflects a truly capacity-imbalanced pack of stick-pairs - which can only happen when at least one cell is found to be truly empty... Or the other extreme cases I mentioned...

I just realized one tricky point: realizing there's a difference, a distinction, between capacity imbalance and voltage imbalance. Pretty major tap voltage imbalances are fine*; it's only when voltage imbalance reflects a major major capacity imbalance that the car raises a fuss.

* "fine" in terms of not causing the car to throw codes and the like. Very small voltage variations can reflect major imbalance. A single cell can be near empty, say 1.25V at rest, while all the others are say 75% charged, each with a voltage of around 1.35V. The tap with a near empty cell will have a voltage only 0.1V lower than the other taps - yet the pack as a whole will have problems, barely functional, throw a P1449-78...

Honda says an imbalance of 0.61v should throw a code, but while observing the cell taps as I drive, I never see that much of an imbalance.
I'm not aware of that dictum... Plus, it contradicts what Honda says in some of the troubleshooting sheets. For example, here's an excerpt from the P1568-66 code (emphasis added): "The high voltage side resistance of the voltage sensor is divided into two (2 MΩ and 1 MΩ) to measure the voltage that is divided at adjacent channels (individual voltage) when the circuit is open. If the combination of batteries is normal, the voltage hardly varies (about 1 V at the maximum)."

So here Honda says up to 1V max variation is "normal."
 

·
Registered
Joined
·
337 Posts
Isn't the slope dropping or rising reflected in the voltage of the stick pair tap? Wouldn't this create an imbalance of the voltage level of this stick pair tap as compared to the other taps?

I guess what troubles me, is I'm not seeing any major voltage imbalances as I drive...usually +/- a tenth of a volt between all 10 taps. Under load, maybe +/- two tenths....no where near 1.2v.

Also, I was wrong about the 0.61v throwing a code. My memory was incorrect about what I had read before:

The advanced DTC troubleshooting explains DTC P1449 (74) as:
"The voltage difference between channels (12 cells/channel) is 0.61 V at the maximum within the allowable capacity range (15%).
However, if one cell in a channel is abnormal, the voltage difference is 1.2 V. Also, the voltage detection tolerance for the BCM (battery condition monitor) module is specified as ±0.12 V at 32 - 140° F (0 - 60°C). Therefore, if the difference between channels is 1.2 V or more, at least one cell is considered abnormal. If there is an offset problem or a gain problem in the voltage detection system, and the voltage detection error is out of the tolerance specified, a malfunction is detected and a DTC is stored."

Reading this again, the 0.61v maximum voltage difference between channels doesn't create a DTC until the voltage difference is 1.2v or more. I stand corrected!
 

·
Registered
Joined
·
7,165 Posts
Discussion Starter #243
Isn't the slope dropping or rising reflected in the voltage of the stick pair tap? Wouldn't this create an imbalance of the voltage level of this stick pair tap as compared to the other taps?
Not sure I understand your questions exactly... If a single cell is near empty, the voltage is gonna fall fast, i.e. steep slope, and that slope will be reflected in the tap voltage slope, i.e. the tap voltage will have a quick drop, too. So I think the answer to your first question is 'yes'. And the answer to the second one would be, generally speaking, yes it would...

I guess what troubles me, is I'm not seeing any major voltage imbalances as I drive...usually +/- a tenth of a volt between all 10 taps. Under load, maybe +/- two tenths....no where near 1.2v.
And why does that trouble you?... If your pack is good, cells normal, you shouldn't see much variation among tap voltages.

Also, I was wrong about the 0.61v throwing a code. My memory was incorrect about what I had read before:
The advanced DTC troubleshooting explains DTC P1449 (74) as...
I've never been able to understand exactly what they're saying in this DTC text, particularly the first sentence.

"The voltage difference between channels (12 cells/channel) is 0.61 V at the maximum within the allowable capacity range (15%) "

To paraphrase, 'tap voltages have a max variation of 0.61V within the allowable capacity range (15%).'

I'm not sure what they mean in the italicized part. It sounds like they're talking about the minimum allowed capacity, like for the P1449-78 when they say it's 10%. So, are they saying that when the pack drops to 15% capacity the max voltage difference among taps is 0.61V? Yeah, I don't get it...
 

·
Registered
Joined
·
337 Posts
And why does that trouble you?... If your pack is good, cells normal, you shouldn't see much variation among tap voltages.
It's troubling because my monitored CVT pack is not good, and I would think I should be seeing a significant voltage drop in a tap value to reflect the lower pack capacity, and I'm not. Without remediation to the pack, I don't think this voltage drop would be a one time event. It would be repeatable on every heavy current demand.

Looking back over my charts of tap voltages, I see that I've driven with some pretty extreme tap voltage imbalances. I can now conclude that tap voltage imbalance alone is not an issue for the BCM. You can drive with major tap voltage differences and the BCM will not bat an eye. OK, maybe it bats an eye, but it doesn't trigger any codes or the like.

--- and ---

I should probably add that, yes, more extreme voltage differences can matter. For example, we know that one of the trouble codes will trigger if a tap's voltage deviates from the others by more than 1.2V for something like 24 seconds or longer within a narrow usage current range (like between -10 amps and +20 amps). But that would signal a major major cell failure. It looks like anything less than that, though, doesn't really matter, the BCM won't make a fuss...
I think the BCM voltage differential may trigger a code for the condition of decreased pack capacity under certain conditions (see below), since I threw a code without the resistor matrix, but when I add the resistor matrix, I'm able to drive much further/longer without an IMA light in both my MT and CVT cars.

I'm also troubled with my crappy CVT pack not seeing a 0.61v differential or a 1.2v difference between taps while getting an IMA light trigger. (Although, it may be that I need to focus more on voltages under load).

In trying to unravel Honda's statement: Is it possible that a 0.61v drop under load is what determines the pack capacity? From what Honda's statement seems to imply is that the 0.61v tap imbalance under load helps determine a trigger to lower battery capacity, and a 1.2v tap differential is low enough to trigger a DTC code. (I'm interpreting Honda's "...within the allowable capacity range (15%)" to mean under load, since Honda also specifies a current range condition of +11 amps to -20 amps for the DTC code)

I must say that Honda's statement is confusing, especially because they go on to say the 1.2v differential requires a duration of 25.4 seconds. Maybe the 1.2v duration greater than 25.4 seconds is indicating a shorted cell or a heavily imbalanced pack, but the 0.61v differential is what determines or helps determine pack capacity, since it's not likely the car remains under load for 25.4 seconds that often.

For my driving style that would be never!!

I hope this makes sense...
 

·
Registered
Joined
·
7,165 Posts
Discussion Starter #245
It's troubling because my monitored CVT pack is not good, and I would think I should be seeing a significant voltage drop in a tap value to reflect the lower pack capacity, and I'm not. Without remediation to the pack, I don't think this voltage drop would be a one time event. It would be repeatable on every heavy current demand.
hmm... When you say the pack isn't good, what exactly is happening that makes you come to that conclusion?

You can have major capacity imbalance without seeing major voltage variation, at least, within most of the charge state range except the very bottom and very top...

I've measured that, seen that, when I had a stick with 1 fast self-discharging cell. The voltages for that cell were roughly the same as all the others - but the cell was operating in the 0-1000mAh range, while others operated in something like the 3000-4000mAh range (on scale of 0-6500mAh). Major capacity imbalance, but not a major voltage imbalance.

A relatively larger voltage drop, one tap vs. another, that you'd see on "every heavy current demand," would likely be a different kind of deterioration, such as 'voltage depression', 'high IR', and the like. That's not really the same thing as bonafide 'capacity imbalance' or degraded capacity.

Strictly speaking, when we talk about "capacity," it should be restricted to amp-hour capacity, i.e. capacity measured in amp-hours. You can have two packs, both discharge and charge the full range - 0-6500mAh - but one might have 'high IR' or 'voltage depression' and not discharge and charge at the right voltages. It can 'take in' and 'put out' about the right amp-hours, but it can only do it at low current... Charge voltages would be higher, discharge voltages lower, so 'usable capacity' has effectively shrunk - because it can't provide full performance within the device's programmed usable voltage range. But, subject it to lower current charge and discharge and you'd generally see the rest of the capacity, measured in amp-hours...


I think the BCM voltage differential may trigger a code for the condition of decreased pack capacity under certain conditions (see below), since I threw a code without the resistor matrix, but when I add the resistor matrix, I'm able to drive much further/longer without an IMA light in both my MT and CVT cars.
I can see that this is a very hard concept to grasp... In some sense it may seem like semantics, but it's not. Let me try again.

It's not the voltage imbalance that triggers the code. True, the BCM needs the proper voltage readings to make the 'decision', so if you have a 'balancing matrix' installed you would not get a code. But the nature of the imbalanced voltages reflect a true, underlying difference - an empty cell. It's not the imbalance that triggers the code, or triggers empty or whatever - it's a genuinely empty cell. The voltage imbalance, or the steep slope, or whatever, signals that a cell is empty.

hmm... Maybe it is a bit 'semantical'... Cuz, it's not strictly 'voltage imbalance' at play. When I wrote what I wrote some days ago, what you've quoted above in your post, the bold part, I was saying that you can have major voltage variation among taps, like up 1.2V, and not have problems - the BCM won't throw a code.

IF however the voltage variation includes a relatively quick drop - steep slope - that's a different thing. In that case you can have much less voltage variation and still have problems - because now the BCM has detected an empty cell. If that happens when other taps/cells are still pretty full, you could end up with the P1449-78, you have major capacity imbalance...

So, there's really two types of tap voltage variation we're talking about:

1) the overall range of variation, where one tap is at say 15.12V and the others are at say 16.08V, under charge or discharge load, or rest, and the BCM doesn't care (doesn't throw a code - it might 'care' and do some things differently, I just don't know about it).

2) the speed of voltage change under some algorithmically-based conditions. I haven't had the ability to measure in real-time, but the most cut and dried circumstances are something like, humming along with a 20 amp discharge load, tap voltages holding fairly steady at say 14.5V, all of a sudden, one tap drops from 14.5V to 14.4V, then to 14.3, 14, 13.8... etc., within a handful of seconds. A cell is near empty.

One minute all tap voltages are about equal, the next, one is dropping fast. So, that's a different 'type' of voltage change, 'imbalance', than I was addressing in the bolded, general statement...

Your resistor matrix hides an empty cell. Before matrix - a cell was probably empty, neg recal triggered, the car tried to charge, couldn't charge more than ~10% before tap-full voltage, P1449-78. After - the BCM doesn't have the data to calculate the correct slope - so it keeps discharging. Perhaps most cells in that tap are operating at around 1.25V (?), that empty one is probably driven into reversal - and probably extreme reversal...

Let's see, the BCM would have to detect a rapid change in voltage of 1 cell, a drop of at least 1.2V, probably more like at least 1.7V, with -0.5V of reversal -- in a string of 120 cells rather than only 12. So the amount of change in the former is only a small fraction of the total when compared to the amount of change vs. the total in the latter:

-1.7V / (120 X ~1.25)=1.13%; -1.7V / (12 X ~1.25V)=11.3%

I guess this simply means the slope would have to be 10 times steeper in order to be detected.

There's other problems/codes you might have gotten or could get - like the ones I posted for extreme imbalance. But what I describe above would be the most typical, I'd think...


I'm also troubled with my crappy CVT pack not seeing a 0.61v differential or a 1.2v difference between taps while getting an IMA light trigger. (Although, it may be that I need to focus more on voltages under load).

In trying to unravel Honda's statement: Is it possible that a 0.61v drop under load is what determines the pack capacity? ...
Yeah, I don't know what they're saying. I've always taken this trouble code to indicate major cell failure (or computer problem), because the enabling conditions are so narrow, while the error values are so extreme. It'd take near-utter failure to see a cell that basically sits at zero V for that long, within those conditions. I don't recall seeing anyone report the code, or it's at least rare. There's a handful of other codes that are much more likely to pop up first - like a P1568-66...
 

·
Registered
Joined
·
7,165 Posts
Discussion Starter #246
Wanted to say a few words, add a couple things, to a couple of the 'management thresholds' I listed earlier. Past couple days I've been driving and focusing on watching these things (watching OBDIIC&C data)...

~17.4V resting - regen limiting invoked, time limits and usage behavior come into play. For example, the first time 17.4V is measured, regen limiting will be enabled and last for some timed duration, or until you use assist and drop voltage.
I'm less and less convinced that 'steep slope' plays a role at the very top, mostly that it almost certainly does not play the major role. It really wouldn't make sense for that management strategy, anyway. It doesn't fit the nature of NiMH voltage at the top.

Steep slope works at the bottom because an empty cell usually has a very distinct discharge curve when it's empty. Although the charge slope does get steeper at the top, when a cell approaches full, it's not nearly as cut and dried as the slope at empty.

At the top, what seems to play the most important role is simply resting voltage. Picture trying to fill a bucket with water, using a hose. If the flow is heavy, it's hard to get the bucket totally full - because water keeps splashing out. If you turn the flow down, much easier, less water splashes out. Now, think of two strategies: You can use the heavy flow and 'cycle' the water on and off - and you'd be able to get more water in the bucket. Spray into the bucket, some water stays in, some splashes out, turn the water off and let the waves settle, try it again. Each time you do this you're likely to get a bit more water in the bucket.

The BCM is measuring voltage - the amount of water in the bucket - when you have your hose OFF, i.e. resting voltage. When the voltage reaches a threshold, about 17.4V - it's considered full, or full enough.

If you try to restrict current/regen as low as possible, you're able to cram a bit more into the pack. You've effectively turned the hose down and can now trickle some more water into the bucket without spilling, until the bucket is nearly full to the brim...

Point is, the 'cyclic' approach makes a lot more sense than a 'slope detection' approach, simply based on how NiMH cells behave when they're close to full and being charged. And this is generally what I see trying to stuff my pack. The BCM or MCM triggers a regen limit at what looks like the very first and quite fleeting occurrence of 17.4V resting. I can drain a bit and repeat the process. If I modulate the current rate, low as possible, I get more into the pack before regen limit triggers.

As I've mentioned before, there's actually multiple things happening at or near the top. This is just one of them.


~13.2V - generally the minimum sustained voltage allowed under assist load.

12V - absolute minimum voltage under assist load. As soon as 12V is measured, assist is throttled rapidly to bring voltage back up to about 14.4V. After that, if assist continues, voltage will be throttled every time it dips to 13.2V. Initially, perhaps at higher charge states, the 'throttle' voltage is a bit higher than 13.2V. Some of this is a bit variable, I think, as it depends on how much assist you're requesting. For example, the above scenario would be more 'text-book' for a full throttle full assist, which can last up to about 4 seconds. But if you invoke full throttle full assist in say 4th or 5th gear on the MT transmission, the BCM/MCM seems to allow minimum voltages below 13.2V but higher than 12V...
There must be a lot of complicated things going on 'around 13.2V'. 13.2V seems less and less of an absolute threshold the more I try to test it. I see it so few times as an absolute threshold that I can hardly claim it to be a threshold. I'm OK with 12V, but 13.2V - not really sure what role that plays.

In general, I want to say that something else takes precedence whenever you're invoking full throttle assist in any gear. As long as you're going full throttle, it looks like ~12V is the absolute low-end threshold. Perhaps anything less than full throttle you're likely to get something around the 13.2V mark - but I don't really see "13.2V" exactly, for example I don't see assist limit flags at 13.2V...
 

·
Registered
Joined
·
337 Posts
FYI... I've got the CVT crappy pack on the simple cycler now, and it's finishing up a fairly deep discharge down to 60v. (No worries, it's an expendable pack LOL) I want to see if the conditioning makes a difference to the IMA light appearing or not with the resistor matrix in place on the CVT (it was throwing an IMA light while the MT hasn't so far). When it's finished I'll take it for a drive and try to record the voltage levels under load.

1) the overall range of variation, where one tap is at say 15.12V and the others are at say 16.08V, under charge or discharge load, or rest, and the BCM doesn't care (doesn't throw a code - it might 'care' and do some things differently, I just don't know about it).
From your discussion above when you say "under charge or discharge load, or rest, and the BCM doesn't care..." I wasn't viewing a one amp discharge as much of a discharge load. I was thinking more like 10+ amps of a discharge load... and the BCM would care then, which is really your point "2) the speed of voltage change. This was what I was referring to that the BCM does care about voltage "imbalance" when under a significant load and the tap voltage drops relative to others in order to detect reduced capacity, high IR, etc. At the end of the day I think we are agreeing, just the terminology was/is being spoken of or viewed differently.
.
I'm pretty confident that my crappy pack has reduced capacity. At rest, after 24hrs, the taps are all within a tenth of a volt. I've measured this with my "flying cap voltage board" and confirmed the voltages with a DVM. I'm pretty sure the IMA light is coming from your point 2) "the speed of voltage change" imbalance on a tap that the BCM is detecting. I'm just trying to capture that value and have an idea what voltage percentage or threshold it's set at by Honda.

My real focus for this is trying to make my "flying cap voltage board" successfully monitor a second pack, and getting the parameters set correctly for that. That's why I'm so keen on what's happening with my crappy CVT pack.

Thank you so much for taking the time to discuss this in finer detail...You are very much appreciated!!!
 

·
Registered
Joined
·
7,165 Posts
Discussion Starter #248 (Edited)
^ No, thank you, the back and forth helps me develop the ideas, think about things (hopefully) more clearly...

From your discussion above when you say "under charge or discharge load, or rest, and the BCM doesn't care..." I wasn't viewing a one amp discharge as much of a discharge load. I was thinking more like 10+ amps of a discharge load... and the BCM would care then, which is really your second point "2) the speed of voltage change." This was what I was referring to, that the BCM does care about voltage "imbalance" when under a significant load and the tap voltage drops relative to others in order to detect reduced capacity, high IR, etc. At the end of the day I think we are agreeing, just the terminology was/is being spoken of or viewed differently.
I still think we're not quite on the same page. It's really hard to capture all the specifics in ...language, at an internet forum...

This is tricky. There's a difference between a fast voltage drop caused by say mashing the throttle, and the 'steep slope' voltage drop that occurs when a cell is near empty. That's why I described a specific scenario of when the empty cell voltage drop would happen and would matter.

When I say "the speed of the voltage drop," I'm really saying the speed of the voltage drop under some kind of conditions that the BCM or MCM imposes on the measurement. I guess I did say that - "algorithmically-based conditions." The conditions have to be something like, '4 seconds after assist is invoked', 'assist held steady for 5 seconds or more - or whatever', 'moving average current within a range', etc etc. Otherwise, 'steep slope' would be detected all the time, under all sorts of driving situations with a normal, operational pack...

Additionally, I'm pretty sure that, when it comes to tap voltage imbalances and/or the speed of voltage change, uneven resistances or conditions of cells from tap to tap doesn't even matter that much.

You can have a 'tap' with such 'high IR' that when you invoke assist, one tap drops quickly to say 12V and the others drop only to say 14.3V. Say they started at 16.1V: the slope for 16.1V minus 12V (-4.1V) is a lot steeper than the slope for 16.1V minus 14.3V (-1.8V) - but it doesn't trigger a code. The voltage variation from tap to tap at peak of assist current is 2.3V (14.3V vs. 12V) - the 'high IR' tap or what-have you drops 2.3V lower than the others. BCM doesn't care, it doesn't throw a code. It will just throttle assist (or the MCM will, based on what the BCM reports)...

So, it's not merely 'the speed of voltage change'; you have to include the conditions.

Certainly, you could have extreme voltage imbalances that would trigger a code - like 4V difference, triggering a P1568-66 - due to, not necessarily an empty cell, but to higher IR on one tap vs. another, at high discharge load (>20, 30 amps perhaps). But that's different from what I've been crowing about, one of the really extreme conditions (though it doesn't take too much to actually end up seeing such a code)...

In a nut shell, the BCM only cares about tap voltage imbalances to the extent that they either:
A - signal major amp-hour capacity imbalance, or
B - signal major cell or equipment faults.
 

·
Registered
Joined
·
337 Posts
I agree with A and B for sure. I'm not so sure about your example where you have a 2.3v differential under load. I think the BCM would store that as a fault or maybe trigger an IMA.

I just took a drive with my CVT and applied a healthy amount of throttle, providing around 10 bars of assist, and I only saw a 0.2v differential between taps when the voltage sample was taken. Keep in mind there are a few milliseconds delay between when tap 1 is sampled and tap 10 is sampled...which could cause an even greater variance in readings. Also, when not under load and just cruising with a steady throttle (no bars), the tap voltages are within 0.1v of each other and frequently all match perfectly.

On my tap monitoring board I have a trigger for cell imbalance set at 2.5% of the average tap voltage obtained. So a sample reading might look like:
162 163 164 163 162
163 164 162 162 163
The numbers above are an example of how my display reads after a sampling. (162 is 16.2v - I just leave out the decimal point on the display to save space, and since there are 10 taps, it's a good indicator of total pack voltage as well).

The Arduino totals all the values and determines the average. In this case it would be 162.8 It then multiplies the average times 0.975 which gives 158.73 as a cutoff voltage and is rounded to 159. The Arduino then compares each voltage to the cutoff number. In this example, all the voltage taps are above the cutoff (which is a 0.4v differential....16.3-15.9 = 0.4v). If any voltage reading was below the cutoff, the display shows which voltage tap(s) were below the cutoff voltage and lights an LED as a warning. I'm pretty sure a shorted cell would easily be detected.

This is how I intend to monitor the second pack of a parallel HV battery set up. What do you think of this scheme for "basic" pack monitoring for the purpose of finding serious faults in the second pack? I'm trying to stay on the conservative side for safety.
 

·
Registered
Joined
·
337 Posts
I forgot to tell you... I did one full throttle run and the max differential I saw was 0.5 volts between taps. It did, however, give me an immediate 12v battery idiot light and the "Brake" indicator light on the dash came on..but no IMA light. Turning the ignition off and restarting the engine cleared all of the warnings. Not sure what to make of it. I think it might have been from the dc/dc converter? I kinda remember weird stuff like that when I had the HV battery switch off in the past
 

·
Registered
Joined
·
7,165 Posts
Discussion Starter #251
^ I thought I understood that kind of error at one time, but I forget now. In general people have most often thought that happens because pack voltage drops too low for an instant and the DCDC gets disabled. But, I don't know... I used to get that a lot with my first pack, which was really crappy. It had a lot of imbalance and high IR...

I'm not so sure about your example where you have a 2.3v differential under load. I think the BCM would store that as a fault or maybe trigger an IMA.
I'm like 99.8% sure it would just throttle assist. Take a look at this graph, in the first page of this thread: The quintessential Insight NiMH voltage thread

It's something Eli at BB made a long time ago, tap voltages under load for a full-capacity duration assist...

Looking at that again, now, I see my "12V" minimum voltage threshold might be wrong - the lowest tap in this graph falls to ~10.5V...

I'll have to look at the rest of your post later.
 

·
Registered
Joined
·
337 Posts
I think one big difference between our writings is that you are referring to shorter term readings, where I'm referring to sustained voltage differences. When you say, "the BCM doesn't care, it doesn't throw a code" regarding a large voltage differential, isn't exactly correct. It does care about the imbalance and takes action by limiting the assist current which brings the voltage differential within "safe" values. If the voltage differential remained at the 2.4v variance, the BCM would care and surely throw a code. I think Honda refers to this value as a 1.2v differential sustained over 25.4 seconds, or it will register an IMA fault.

I'm sure when I'm driving and a voltage tap sample is taken with my monitoring board, I'm typically seeing a snapshot of the sustained voltage reading of all the taps. It may be difficult for me to obtain a specific short term snapshot reading with my setup.

I agree that the DC/DC being disabled is the likely cause of the brake light and battery symbol being illuminated. That was my thought as well.
 

·
Registered
Joined
·
337 Posts
I just came back from a drive and have some real voltage readings for you. I was able to record them on my phone:

cutoff=465
582 541 561 410 430
440 470 460 440 440

The above numbers are from a full throttle reading. Looks like I caught the readings as the BCM was going to cut back on current (like the beginning of Eli's graph). Notice that the readings trend lower because of the brief time delay from reading tap #1 to reading tap #10. I think the readings are two decimal places in this case: e.g. (582 = 5.82v). The average is 4.77v and my cutoff differential of 97.5% calculated out to 4.65v in this case (a differential of 0.12v max allowed from average). Taps 4,5,6,8,9,10 lit up LEDs on my board based on my 97.5% differential rule. The Honda BCM did not light the IMA, but the DC/DC fell out and lit the battery and brake lights on the dash. Surely this a worse case type scenario. I also want to mention that the battery tap resistor matrix is still in place, so the BCM is seeing balanced voltage taps, where as the voltage numbers listed above are read from the battery by my voltage tap board.

This is why I refer to this pack as a "crappy" one. Also, when cycling it on the charger/discharger, I only calculate about 4 amps total when discharging it (my rough calculation of mA load vs time)

Here's another set of real data from the same drive with less assist:

cutoff=118
124 122 123 120 120
121 122 121 121 121

In this case, the average was 12.15v and my 97.5% cut off was 11.8v (a differential of 0.35v max allowed), so no LED was lit on my board.

Right now I'm feeling pretty good about my 97.5% of tap average cutoff rule for secondary pack monitoring. It will still need some adjusting when it is in a parallel set up, since the current demand is cut in half.
 

·
Registered
Joined
·
7,165 Posts
Discussion Starter #254
I think one big difference between our writings is that you are referring to shorter term readings, where I'm referring to sustained voltage differences. When you say, "the BCM doesn't care, it doesn't throw a code" regarding a large voltage differential, isn't exactly correct. It does care about the imbalance and takes action by limiting the assist current which brings the voltage differential within "safe" values. If the voltage differential remained at the 2.4v variance, the BCM would care and surely throw a code. I think Honda refers to this value as a 1.2v differential sustained over 25.4 seconds, or it will register an IMA fault.
Well, I've pointed out numerous times that there are extreme cases that do make a difference - like the very case you cite...

When you say, "It does care about the imbalance and takes action by limiting the assist current which brings the voltage differential within 'safe' values," in general, no, that's not true. It cares about the voltage falling too low, it doesn't care about the imbalance from tap to tap...

Granted, there's only a fairly narrow range in which tap voltages can be imbalanced without any tap being too low. For example, decent, charged taps might simmer around 15.6V at a 20 amp load for a good number of seconds, like maybe 10, while one tap might have messed-up cells or be charged lower or whatever - and its voltage might fall to 12V immediately upon 20 amp discharge. The difference is 3.6V. That would probably be just about the max, real-world voltage imbalance possible without causing problems, because the 12V tap is at the very edge of being too low, while taps won't hold 15.6V very long. The variation itself is not the problem, though, in the BCM's eyes, it's the low voltage tap.
 

·
Registered
Joined
·
7,165 Posts
Discussion Starter #255 (Edited)
I just came back from a drive and have some real voltage readings for you. I was able to record them on my phone:

cutoff=465
582 541 561 410 430
440 470 460 440 440
Holy crap, I had no clue what readings these were at first - but these are actual tap voltages?? 4.40V?? No wonder your DCDC is disabling - the pack voltage IS too low.** I'm a bit surprised the BCM or MCM doesn't respond fast enough to throttle, but then again, this probably happens in a split second...

You should try a tap-level ultra-deep discharge. I've had good luck with that, 'cleaning up' the garbage that makes cells basically sh*t their pants like yours are...

** This is actually great data. I had been just thinking about the DCDC disable thing, trying to recall what Peter P. wrote some days ago or so. I think he said the minimum DCDC operational voltage is something like 75V. So, when people are seeing the 12V battery warning light and brake warning light, it's probably because their pack voltages are dropping low like yours - below 75V. That actually makes a ton of sense, because Insight NiMH cells tend to uphold voltage right around 0.7V per cell (due to chemistry), so 120 X 0.7V = 84V... If pack voltages are dropping lower than that, those cells are really really bad... Even dropping that low is really bad.
 

·
Registered
Joined
·
337 Posts
^ "Those cells are really really bad..." Yes sir! I told you they are crappy!! But with a CVT, it's not a big deal because you have a near infinite gear ratio. I had just cycled those cells down to 60v to get them to last a few more days. They're actually pretty well balanced now, and I usually don't mash the gas to full throttle!

Anyways, the Escape pack is going to replace them real soon. I just finished installing the monitoring wiring in both my MT and the CVT. An LCD display sits on top of the rear view mirror and a small LED hood sits on the steering column, similar to MIMA.
Well, I've pointed out numerous times that there are extreme cases that do make a difference - like the very case you cite...
I'm sure you have...I really need to read this whole thread.
When you say, "It does care about the imbalance and takes action by limiting the assist current which brings the voltage differential within 'safe' values," in general, no, that's not true. It cares about the voltage falling too low, it doesn't care about the imbalance from tap to tap...
Well it certainly was the voltage falling low in this case, since the matrix was sending perfectly balanced tap signals to the BCM. And I agree about the BCM caring about the voltage being too low as well, I just believe Honda's data about the sustained 1.2v condition would trigger a code. It just makes sense to me that the BCM would be sensitive to that voltage for a shorted cell. But that's just my crazy way of thinking, and you were right, my verbiage was off when I called it an imbalance when it was really the cell voltage dropping.
 

·
Registered
Joined
·
7,165 Posts
Discussion Starter #257
...And I agree about the BCM caring about the voltage being too low as well, I just believe Honda's data about the sustained 1.2v condition would trigger a code.
It would, no argument there.

It just makes sense to me that the BCM would be sensitive to that voltage for a shorted cell...
No one's arguing otherwise. Your inclusion of this code, these points, now makes it seem like you're trying to generalize the "1.2V imbalance" thing to other situations, where it doesn't apply. You have to take into account Honda's very strict enabling criteria for that code, it's not just 1.2V imbalance, it's 1.2V plus:

-Tap voltage above 14.4V
-Minimum pack temp 77F
-input/output current between -20 amps and 11 amps
-and it has to happen for 25.4 seconds or longer

In other words, conditions have to be near-perfect, the load only modest, and then, if there's a deviation of 1.2V or more for more the 25.4 seconds - BCM will throw a code at you...

This code, this 1.2V deviation thing, is all about uncovering a truly defective/failed cell.
 

·
Registered
Joined
·
337 Posts
Maybe my focus is different because of my intent to monitor a parallel pack and a failed cell can be a little more catastrophic than in a non-paralleled pack? I'm sorry if I offended you.
 
241 - 260 of 268 Posts
Top