Triangle of death and scangauge

I may be totally wrong but aren't you quoting the Gen II HV voltage? The Gen I is quite a bit higher, something less than 100 more volts. BW knows all those numbers cold.

 

These look OK although 13.8-13.9 would be closer to my tastes.

The 212 V traction voltage seems off but I've got my Auto Enginuity with me this evening. This is one of those cases where it could be the instrument, the car but the traction battery seems improbable. I've never really studied traction battery voltage except to calculate energy flows. I'll check some of my recorded Graham Miniscanner data files later tonight.

Bob Wilson

 

First, to clarify ignition-on. There are 5 positions on the key cylinder:

1) LOCK - to lock steering wheel
2) OFF - key can be removed
3) ACC - Accessory mode
4) ON - Ignition-on mode, Ready light is off, engine cannot be started
5) START (momentarily) - Ready light is on, engine can start anytime

In a conventional ICE-only car, ignition-on would mean engine is running. But for a hybrid car, engine can stop and start as required, hence it is important to note the status of Ready light.

When you switched to Accessory mode, the SMR will disconnect the HV battery and Ready light will be off. You can read about SMR here:

http://www.autoshop101.com/forms/Hybrid03.pdf

As what you 've mentioned in another post, the 212V is likely due to transient. I've observed varying voltages too. Now, if you pressed the Home button to wake up SG and check btV again (you need to cycle the button to refresh btV) , it should be blank.

When the car is Ready-on, the voltage reading of the 12V battery is actually measuring the DC-DC converter output that is charging it. Mine is about 14.0V. When it is Ready-off, the HV Battery is disconnected so no output from DC-DC converter. What you see now is the actual output from the Aux. battery.

The ic1-ic5 are newly discovered XGauge so it was not in the tables. It was posted on the same thread at post #67.

Here are the suspected INF. code:

	Column 1	Column 2	Column 3	Column 4	Column 5	Column 6	Column 7
0	XGauge	TXD	RXF	RXD	MTH	NAME	Notes
1	INFO. Code 1	8216F101AA	0316044105AA	2810	000100010000	ic1	XXX
2	INFO. Code 2	8216F101B1	0316044105B1	2810	000100010000	ic2	XXX
3	INFO. Code 3	8216F101B7	0316044105B7	2810	000100010000	ic3	XXX
4	INFO. Code 4	8216F101BD	0316044105BD	2810	000100010000	ic4	XXX
5	INFO. Code 5	8216F101DF	0316044105DF	2810	000100010000	ic5	XXX

I've tried to induce an error but I would need more data to confirm.

 

I think it is possible to query the various ECU for DTC. I've not tried it since my car has not logged any yet.

Here are some new Xgauges (not tested):

	Column 1	Column 2	Column 3	Column 4	Column 5	Column 6	Column 7
0	XGauge	TXD	RXF	RXD	MTH	NAME	Notes
1	MIL illumination	8210F10101	031024410501	2801	000100010000	MIL	OFF / ON
2	No. of Engine DTCs	8210F10101	031004410501	2907	000100010000	#EC	XXX
3	MIL illumination	8216F10101	031624410501	2801	000100010000	MIL	OFF / ON
4	No. of HV DTCs	8216F10101	031604410501	2907	000100010000	#HC	XXX
5	MIL illumination	82D5F10101	03D524410501	2801	000100010000	MIL	OFF / ON
6	No. of Battery DTCs	82D5F10101	03D504410501	2907	000100010000	#BC	XXX
7	Engine DTC 1	8110F103	03101443	2010	000100010000	ec1	XXXX hex.
8	Engine DTC 2	8110F103	03101443	3010	000100010000	ec2	XXXX hex.
9	Engine DTC 3	8110F103	03101443	4010	000100010000	ec3	XXXX hex.
10	HV DTC 1	8116F103	03161443	2010	000100010000	hc1	XXXX hex.
11	HV DTC 2	8116F103	03161443	3010	000100010000	hc2	XXXX hex.
12	HV DTC 3	8116F103	03161443	4010	000100010000	hc3	XXXX hex.
13	Battery DTC 1	81D5F103	03D51443	2010	000100010000	bc1	XXXX hex.
14	Battery DTC 2	81D5F103	03D51443	3010	000100010000	bc2	XXXX hex.
15	Battery DTC 3	81D5F103	03D51443	4010	000100010000	bc3	XXXX hex.

The DTC are Bit-encoded. See here:

OBD-II PIDs - Wikipedia, the free encyclopedia

 

Using Auto Enginuity, we may have a clue about the 215 V. traction battery:

I configured Auto Enginuity to record the HV battery ECU data. The first surprise is it has no single voltage metric. However, I could record all 19 modules. So I used a spreadsheet to sum all modules and discovered that turning off the car gave only a partial number of modules and the final sum was ~215 V (see chart.)

I'm going to speculate that the battery ECU may only report the 19 module voltages and the HV ECU may do the sum to report the total traction battery voltage. During vehicle shutdown, some controllers may turn off sooner, say the battery ECU, delivering only a partial count of module voltages. The HV ECU can only sum the available data. But the SG may not 'clear' the last value. So during shutdown there may be a race condition that leads to abnormally low, traction battery voltage.

I also recorded the minimum and maximum voltages as well as the module identifier:

Off hand, I don't see a problem as the weakest modules are spread throughout the pack. There is one 'super set' that could be discharge trimmed down but nothing worth spending a lot time.

Here is the map of min/max voltages and current:

I was surprised to see the minimum/maximum battery voltages flip in the data. This suggests transients can significantly impact how these are reported.

Bob Wilson

 

I agree with the outliers in data points being caused by transients, and I filter them out in Excel.
Below is a graph done not too long ago, of an 03Prius Original Traction Battery hi and low block voltages sitting overnight just before start, and then captured the voltage sag from the start itself. The charge cycle I believe starts immediately afterwards..

 

And here's a plot of a short drive while recording hi/lo block voltages. You can see the initial start sag, then during the drive, regen braking can be seen as the dominant charge factor.

 

In the last plot the dip to 14V on the low block is from a hard acceleration. Looks like that module drops during accels more than any other in the pack. Perhaps this signals the end of this 8 year old pack is near.

 

Could you capture the block numbers of the low and high blocks? Something like this:

Granted, we don't have a properly instrument, pre-failure battery survey. Still, the failure records usually show just one module in a pair fails first. I would expect that module-pair to show a higher frequency of low-voltage hits in graph.

One surprise, I found P3002 in the battery ECU. I had induced some codes to test the other scanners. Now I need to go back, read all the codes and retest with the CEN-TECH and Graham Miniscanner. FYI, nice write-up here.

Bob Wilson

 

Those are supposedly the Infomation Code or subcode. If you got 0, that means no subcode has been logged. You don't have to keep monitoring these, only when you suspect something is not right, then check what is there to give you some clue.

Mine are mostly 21 mOhms with a few at 20 and 22 mOhms. So your values look alright.

My Vlo and Vhi are also within 0.2V~0.3V. I can't remember the exact Blo and Bhi but I know some of the blocks tends to appear only at Blo and some at Bhi. I don't think it is a problem if your Blo is always the same block unless you are also seeing more than 1.2V difference for Vlo and Vhi.

Your SOC is higher than mine. Do you have many hills in your commute?

Normal. My stock Aux battery is even lower at 10+ V. I've a portable jump start kit just in case.

So far, all your data seems normal to me. I suggest keep yourself familiarized with all other Xgauges and their normal values. If the triangle come back again, you can find out what has changed.

 

According to the RM, INF. Code 123 means HV Battery Malfunction - Input of abnormal signal from battery ECU (HV battery system malfunction).

Together with your previous P3002 - No serial signals transmitted from HV ECU, it is highly likely that the communication bus line has intermittent open/short circuit. If checked OK, then your HV ECU could be sending wrong signals or not sending the right signals to the battery ECU.

 

The battery connectors are at the rear on the driver side:


The HV connectors are under the passenger side, upper foot rest.

FYI, I also got a P3002 recently but have no idea how. I've cleared it and will wait for it to come back.

Bob Wilson

 

Last night as I was going to sleep, I realized that you should be able to fault-isolate a P3009 using a modern, high-impedance, DVM, not a 20,000 ohm per volt, mechanical meter. The usual practice is to use a MEGGER that uses a high voltage to measure high resistance shorts but the traction battery already provides significant voltages.

After appropriate safety precautions, remove the cover and starting from each end, measure the voltage between the case and the end terminal. If you get a steady voltage, you are measuring the voltage loop from the probe, down the battery pack, the resistance trace to ground, and ground. The offset should be:

measured_V / 7.2 ~= offset_to_leak
​

The safety precautions should include removing the interlock but for good measure, disconnecting the 12 V battery ground to ensure nothing can even attempt to enable the main power relays. In this mode, the traction battery consists of:

• 20 cells, ~144 VDC
• 18 cells, ~130 VDC
• 0 VDC at leaking module terminal

Having fault-isolated the leaking module, there are two alternatives:

• remove, module(s), clean and put back (UGH!)
• replace module(s), tricky because the charge needs to be normalized and we'll never know the Ahr capacity versus the other modules (UGH!)

So you have options but thinking about the P3009, it looks like we can fault isolate to a specific module with just a DVM.

Bob Wilson

 

Your car missed you and felt neglected and betrayed, thought you were out test driving another car. Watch out, your car may show up at a few murder sites.

 

"-when the car is first started, there is either some capacitance or hysteresis that causes the computer to initially read the voltages from when the car was last running"
Yes the sys will retain the last values before last shut down, or until it gets new data. Say this sure sounds like a low 12v battery. If you let a gen1 sit for ten days that battery would be too low, and don't forget all of the monitoring systems in the prius depend on 12v reference being stable, at least at startup before the inverter takes over. just me 2c