According to one of their presentations, the voltage can be as much as 700 V if not higher.
ScanGauge II Extended PIDS: Capacity and Specific Energy
- Thread starter AnthonyW
- Start date
- Replies 59
- Views 18K
According to one of their presentations, the voltage can be as much as 700 V if not higher.
Adding on to this, this report. It analyzes the previous Accord hybrid (a little of the BMW i3), including motor and inverter efficiencies. Various pictures including inside the inverter. It is an Oak Ridge Labs analysis.
https://www.google.com/url?sa=t&sou...FjAAegQIAxAB&usg=AOvVaw2zOZSRgZx_-Rcus9B1XQ54
AnthonyW
Well-Known Member
FINALLY heard back from James at ScanGauge (it has been about 6 weeks). He said he took some time off and then was catching up when he got back. He said he has been studying the data from the first data logger I returned to him. To move forward he said he is going to send me another ScanGauge with updated software and another data logger that will be configured differently. So hopefully we will have this show back in the road in the next few weeks.
Sent from my iPhone using Inside EVs
Sent from my iPhone using Inside EVs
AnthonyW
Well-Known Member
So I have some more info that might help settle the debate of the buffer at the top and bottom of the battery (or add to the confusion). @Viking79 has held firmly that the battery only charges to 85% of its total capacity. In other words, when HondaLink shows 100% it is really on 85%. @Viking79 may be right...
Those of you have followed my Scangauge experiments already know that there is an internal SOC called MXS that is different from the HondaLink SOC. MXS ranges from 96% to 20% which corresponds to HondaLink's 100% to 10%. For a while, I was thinking MXS represented the entire battery pack SOC. Now I am beginning to wonder. I believe that MXS represents what Honda has assigned as the usable capacity of the battery in its current state.
So for the past few months I have been taking open circuit voltage (OCV) readings of the voltage of the battery and the associated MXS % of charge. By OCV, I mean voltage measurements where no charge or discharge has been applied to the battery for at least 2 hours. I translated the voltage of the entire battery pack (352) to that of a single cell (4.20) by dividing by 84 since our battery is 84s2p. I graphed the relationship, applied a trend line, and determined the slope of the curve equation (order of 5, polynomial equation). Here it is:
Inferring from the trend line, it shows that the cut off voltage is 2.75 volts (which is spot on for a NMC battery), a 20% buffer at the bottom and a 14% buffer at the top. If this is true then it would suggest that of the entire pack the true SOC available to us is around and about 86% to 20%. I should probably rescale the percentages but for now this is close enough.
Those of you have followed my Scangauge experiments already know that there is an internal SOC called MXS that is different from the HondaLink SOC. MXS ranges from 96% to 20% which corresponds to HondaLink's 100% to 10%. For a while, I was thinking MXS represented the entire battery pack SOC. Now I am beginning to wonder. I believe that MXS represents what Honda has assigned as the usable capacity of the battery in its current state.
So for the past few months I have been taking open circuit voltage (OCV) readings of the voltage of the battery and the associated MXS % of charge. By OCV, I mean voltage measurements where no charge or discharge has been applied to the battery for at least 2 hours. I translated the voltage of the entire battery pack (352) to that of a single cell (4.20) by dividing by 84 since our battery is 84s2p. I graphed the relationship, applied a trend line, and determined the slope of the curve equation (order of 5, polynomial equation). Here it is:
Inferring from the trend line, it shows that the cut off voltage is 2.75 volts (which is spot on for a NMC battery), a 20% buffer at the bottom and a 14% buffer at the top. If this is true then it would suggest that of the entire pack the true SOC available to us is around and about 86% to 20%. I should probably rescale the percentages but for now this is close enough.
GrayRed
Member
Wow, great work!
I agree with your estimate above.
Don't forget, many State laws require Honda to warranty the battery for 15 years or 150K miles.
Unless Honda is planning to go bankrupt ;-), Honda has to plan and to be sure that that at ~15years/150K, the battery failure rate has to be below "~1 sigma" of the approximate bell-curve life/failure rate.
I agree with your estimate above.
Don't forget, many State laws require Honda to warranty the battery for 15 years or 150K miles.
Unless Honda is planning to go bankrupt ;-), Honda has to plan and to be sure that that at ~15years/150K, the battery failure rate has to be below "~1 sigma" of the approximate bell-curve life/failure rate.
I appreciate the work that is being done here. I certainly don't understand a lot of it, but what I do understand is that we are trying to gain some important information about how our Clarity's operate, and hopefully eventually information about battery health and longevity. I have a Nissan Leaf, and important information about battery health was learned from an OBD ii app called LeafSpy.
Ray B
Active Member
Hi AnthonyW - just curious if you have any update on the ScanGauge II capabilities. I am considering getting one, and I am curious what data it can access on the Clarity PHEV. Also, does it log the data in any way, or is it just displayed. If it is logged, how do you access it afterwards?
Unless the latest firmware upgrades have added this ability, I've never seen any data logged - it's instantaneous. But for those who'd like to know things like coolant temp, RPM, etc. it's a great device. Seems like it's taking them a long time to get Clarity x-gauges up and running, though. Having one, I'm kind of bummed since it's pretty much dead in the water until they create some x-gauges that work.
Ray B
Active Member
A search of the SAE site for 'honda clarity' came up with this link for 'diagnostic communication protocol' stuff.
https://www.sae.org/standards/content/j2809_201710/
Not certain it is useful, as there is no direct mention of the Clarity in the information shown on the page, and it says it only applies to some Honda models.
https://www.sae.org/standards/content/j2809_201710/
Not certain it is useful, as there is no direct mention of the Clarity in the information shown on the page, and it says it only applies to some Honda models.
Clarity_Newbie
Active Member
@AnthonyW
I ran a EV 0 test yesterday. The SGll numbers are below
HBV 342 volts full charge
HBV 292 volts when kicked on ICE
MXS 96.4% after full charge
STC 99.7% after full charge
Notes:
STC fluctuated Essentially between 8% and 10%...appeared to me the ICE was throwing charge to battery as often as it could to maintain ~10% SOC
HBV fluctuated ~3% Essentially between 290v and 300v
Hondalink showed 8% charge remaining versus STC 8.65% at end of test.
If you want to see 5 pics of SGll during test run https://bdacomputer.homestead.com/
Hope this helps.
ClarityBill
Active Member
Not sure where to post this, but I have been using a Blue Driver scan tool, and recording engine data. Seemed like I got the 'angry bees' when the engine RPM went to 4,000. First time, I took it out of HV mode, and drove in EV... Second time, I was going to be on the highway too far, and needed the HV range. RPM's came down, and seemed to respond well to the hills after that. Was hoping to get some discussion on these charts. Seems like RPM's jump up too quickly.
MrFixit
Well-Known Member
ClarityBill:
Thanks for the data ! I love data !
Some thoughts and lots of questions -
1. I assume the X-axis is minutes in both cases? (one plot states that on the title, the other blank).
2. It sure looks like this is the bees... Plot1 is nice because it shows the entire event (the bees only last for a few seconds though). In Plot2, the bees are already swarming at the start. Would you have considered the bees to be swarming during only the first minute (~4000), or would you have called it bees during the much longer stretch ~3000 RPM? Do you have any Plot2 data just before the event in order to see more about the condition that triggered this?
3. What is "Battery"? Is this the state of charge for the traction battery (% units)? In Plot1, this goes from 80% to 50% in only 10 minutes. This seems implausible. Could there be a scale factor issue? Plot2 is steady, presumably because you are in HV mode.
4. What is 'Temp'? Plot1 is 30-40, Plot2 is 10... Is this Degrees C? It doesn't look like engine temperature based on what was going on.
5. The RPM's in Plot1 make sense to me. In Plot2, the RPM's are much more erratic than I would have expected. The 'high' RPM's last for a very long time here. Was your charge really at a steady 40% as implied on this plot?
6. With Plot 2, you were running at ~74 MPH for 28 minutes (35 miles?)... Was this mostly a flat stretch? There are two segments where direct drive may have been engaged (11.6-14 minutes, and 19-27 minutes). I assume you were on cruise control through most of this? I might have expected more direct drive, but maybe it couldn't do it at 74 MPH. I wonder if the two segments where it was direct were downhill grades?
Related to your setup:
1. Is your scan tool "Clarity Aware", or did you choose a different vehicle to get this far?
2. Did you have to define any custom PID's in order to get 'Battery', and 'Temp'? I have been using Torque, and it knows nothing about the Clarity. I could get engine RPM's, but nothing about the battery (other than the 12V battery). Temperature was also non functional. I have been trying to define some custom PID's based on some work that AnthonyW has done with ScanGauge, but have not been successful yet.
3. Can you obtain the PID parameters for 'Battery', and 'Temp' from your tool so I can try them in Torque?
4. Can you access other 'hybrid' parameters? In particular, traction battery voltage, current, power consumption?
Ray B
Active Member
MrFixIt: My bluetooth OBD scanner provides EV battery SoC% as standard on Torque Pro without adding any custom PIDs (https://www.amazon.com/BAFX-Products-Bluetooth-Diagnostic-Scanner/dp/B005NLQAHS/
Also a 30% SoC drop in 10 minutes could be real. If we take ~38 miles as a normal winter range, then 30% of that would be ~11 miles, which is about what would be covered in 10 minutes given the higher speeds at the beginning; also not sure about the use of heat or other factors.
MrFixit
Well-Known Member
Ray B: Thanks. I do have SOC in the menu, but I am pretty sure it wasn't working for me. I'll try again and maybe upgrade to Pro in case it is an issue with the Lite version.
Do you get any other EV-unique stuff? I would like to see traction battery voltage / current...
interesting observation about the SOC. I am still thinking excessive though. The 30% decline occurs between 0 and 7 minutes. Even if the average speed is 60 during that time (it is probably lower) then that would be only 7 miles. At 30% depletion, that represents only a 23 mile total range. Not out of the question, but seems unlikely.
Last edited:
Ray B
Active Member
EV battery SoC% (actually called "Hybrid Battery Charge (%)" ) is all that I can see on the list that would be different than normal ICE car parameters.
I have been hesitant to add any other extra PIDs manually yet. (I'm referring to the ones mentioned in this http://www.cleanmpg.com/community/index.php?threads/50474/ that was linked in this thread, https://www.insideevsforum.com/comm...auge-ii-obdii-extended-pids-temperature.4435/). Let me know if you have attempted to enter any of those extra PIDs into Torque...
Last edited:
ClarityBill
Active Member
AnthonyW
Well-Known Member
@Clarity_Newbie
This does help as it is very similar to mine. 342v (which corresponds to 4.10 volts) is the max I get. MXS ranges between ~96.0% and ~20.0% and STC from 100.0% and ~10.0% so we are in line. My ICE kicks on between 295v to 285v (3.51v to 3.39v). Thanks for the info!
AnthonyW
Well-Known Member
Below shows the mapping of the amps. This comes from the CUR/HVC PIDS and the readings are instantaneous. For each tick mark the amps could be +/- 3 amps for each number shown. I believe the 6th tick mark is 150 amps but I have not had a chance to hold it on that tick mark long enough to get a good reading. In my experience there is no defined relationship between a set number of amps and the green tick marks. It is variable based on conditions. Lastly the mapping that is shown is the same in all modes, including Sport. We are working on a watts/kilowatts PID but so far not much luck.
