Why Prius ECU can't reduce battery temperature while Dr Prius app can?

When that white plug in the back rear right quarter get corroded the fan certainly does not work as expected and when they're corroded and broken it doesn't come on at all . It forces you back there to fix it jumper it something I now in the four twos here the fans seem to work as expected . I'm in the 100 not blowing batteries . We also not hard on car I other cars to beat on.

 

@amitt are you getting any codes?

Repair Manual excerpt attached:

 

No codes, i was just actively monitoring temps on dr prius on a hot day. I had suspected this fan for a couple of years now, tried to clean twice, and put some oil on the splinder of the motor.

But when i saw the texhstream mode 4/5 attempt with no fan response, i knew that something is wrong with that auto mode PWM.

Tombukt, that symprom will happen only in auto mode , but work fine in manual active test mode or via dr prius?

 

Both screenshots you attached there are showing VMF = 9.4 V for control mode 4. Is that the VMF value you're saying is wrong, or the one you'd consider right?

 

Correct, 9.4v is not a good feedback for speed #4 , how do i know besides trusting my ear that speed #4 was not on?

Well... look at the attched screenshot i took with " active test mode" video, at speed 4, the voltage should be much below 7.4v , ( just toggled through the speeds fast without letting the feedback settle) at speed 6, the voltage goes down to ~ 2.5v

BTW Mendel, thx for the pdf, i found the needed page regarding PWM signal ( attached) it appears that when the signal down time increase, the fan blows faster. This mean that if there is noise there ( signal high) the fan will not blow properly because the duty cycle will be too short, or it will not blow at all.

Now i need to get my old back to open those trims again and probe it. I was involved in a car accident recently and im in really bad pain, i dont see myself doing this before i get better , honestly. I was hoping it will not be a physical fix.

I live in the bay area, if you're a diagnostic freak and feel like you want to help me solve this for the good of this great community, im open for help i got the scope and everything... just no muscle power , and herniated disc

 

Just for fun, to continue the previous message, this is the image of feedback voltage reading i got with speed #6, the feedback is at 2.8v.
This means that as the actual fan speed goes higher, this voltage goes lower.

I wonder why toyota decided to let this go out of control and not monitor whether the fan is not spinning at the correct speed that the ECU order it to go, and instead just rely on the temperature sensor to shoot a fault , only when the battery is very, very hot.

This is definitely a problem with this design, since i doubt that even toyota technicians know how to read this feedback, they would probably just toggle the speeds and determin that everything works properly in active test mode, but its obviously not ok when you switch to auto mode.

Chatgpt claims that in later models, toyota added an encoder to monitor actual fan speeds.

I wonder how many gen 2 prius owners slowly cooked their hv battery until it died this way.

I bet that there is noise on that signal, which interrupts the duty cycle and the "low" time of that PWM signal, which makes it only blow at very low speeds, or not at all, which is exactly my problem.
Why the PWM is not wrong in manual mode? Maybe because the signal that causes noise ( due to corrosion or whatever) is disabled in manual mode.

My problem seems to happen to other prople online that wondered why their fan is not able to cool down their battery.
Some people responded to them that toyota engineers knows what they are doing and this is how it should work even with these high temps, but !, the fact that the ECU iteslf, that toyota engineers designed, orders it to go to speed 4/5 and it does nothing or spin very slow, is a clear problem with this design, and a problem with the lack of processing this feedback voltage, to at the very least, log a warning in the ECU.

That is just too sad, those gen 2 prius are solid otherwise.

Please let me know if you live in the bay area ( im in SJ) and can help me as a prius diagnostics freak... in any case this forum is so fun, i like to talk about problems with my old beloved beater prius, with other knowledgeable people . Much appreciated !

 

The repair manual Mendel linked is pretty straightforward relative to testing and basically says if it passes active tests start looking at the ductwork and then change parts.

I have driven a gen2 and a gen3 for roughly 450k miles and never had issues with battery cooling. However I haven't kept a Prius that was 19 years old. I did change the gen3 battery around 200k miles to restore capacity.

Honestly it could be the rather simple voltage controlled battery blower or the pwm to voltage blower controller assuming the battery itself is not bad and simply overheating.

The fact they pass the active tests makes the battery suspect but I could see bad bushings on the blower or pin corrosion as Tom points out.

I tested my gen3 today for 1.5 hours with mixed driving up to 80 mph and could not exceed 105f on the center temperature sensor and 100f on the other two. This was at 90f ambient and 85f battery intake. It appears regen at freeway speeds caused the most temperature fluctuation. 95% of the time the fan was about 40% and always inaudible.

Regardless, gen2 blowers are $40 on ebay with the controller around $30. The logic and hardware of the battery ecm seems unlikely but even those are available for less than $50. I would at least have a blower and controller on hand along with a can of DeoxIT D5S-6 spray when your helper tears it down.

 

Thanks rjparker,

The fact that you could not exceed 105f is reassuring that the control system in my car is not working properly, i can easily get 125f-130f with very low speed of the fan, or sometimes it does not work at all.

But, when i let Dr prius regulate the fan speed, or if i turn speeds 1-6 from techstream active test, the fan works perfectly fine.

Thanks for the thought about the blower controller, it appears that it receives the PWM from the ECU, and turn it into DC voltage to control the fan speed.
It is still not clear why it works well i manual mode only, there may be noise that does not allow the modulated signal to stay low for long enough , even for low speed interpretation by the controller.
Another though is a slightly different FW functionality in manual mode, which has a more robust signal integrity.

I'll try to get to it when my medical condition is better, i dont know if ill have the bandwidth to probe everything before the controller with Oscope, or just go over the connectors to check for that famous corrosion.
From what is see here on priuschat, the corrosion is after the controller, so at this point it is DC voltage and not PWM, very strange.

In any case I'll drive with Dr prius on for now, its able to keep it cool to about 105f-110f when i set the override temp to 92f.
Not sure if its because my battery is old so it needs a lower temp setting in dr prius to achieve lower than 110f , but thats besides the fan signal clear abnormalities, since the ECU is sending it to high speed, ans its only going low, or not going at all.

I added a computer 12v fan at the very end of the duct near the 12v battery, to pull air out instead of pushing it like the blower fan does - i have to say that it does not do almost anything at lowering the temps, even though it is clearly working hard and i see the flap taking air out of the car. I also feel that it is pulling air into the car through a small crack in my driver window, however, it is not cooling the battery much.
strange... i guess that pushing air into the ducts is far more afficient than pulling air our of the ducts. See attached pic of where i placed the computer fan, its pulling a lot of air directly from the end of the duct and out of the car body, but not able to cool the battery, probably because it pulling air through the many holes in the ductwork and the battery case, oh well, at least i tried.

Thank you

 

Maybe not conclusive, but if the system requested fan and didn't get a response, I'd think there would be codes. Keep in mind, your concerns hinge on your estimation of what's acceptable temps.

Are you always monitoring the fan via an app, or sometimes just listening for it? If the latter, fwiw: I've never heard ours, even though scangauge has shown it to be running.

It's been years since I've monitored with scangauge (it apparently was the cause of some comunication errors, I finally disconneced), but when I was, for another data point:

on a cool evening, light traffic, coastal highway at around 80 kmh (50 mph), middle temp sensor was around 35C (95F), and fan speed 2. As usual, I could not hear the fan.

 

I think amitt has reported seeing substantially different VMF readings when selecting fan speed 4 manually compared to when fan speed 4 is automatically selected. That's a puzzling finding in itself. I certainly don't have any story for it. Maybe we'll know more when amitt gets back out there with a 'scope.

 

A fan that is dragging with bad bushings or brushes can work at reduced rpm or intermittently not at all. In this case the voltage feedback is simply the applied dc voltage regulated by the battery fan controller with no actual rpm status.

 

Agreed. Which makes it even more puzzling to see different values for the same targeted speed.

I'll look forward to what amitt finds with the 'scope.

 

Axial fan? Axial computer fans usually won't work against (or create) much of a pressure differential. They are fine for moving warm air from inside a case to outside a case because the pressure difference in that application is quite small. That's why the car uses a centrifugal blower for the pack (and the cabin). The pressure can be quite different across a fan in that sort of application, and it needs to be able to move air regardless.

Edit: it is a pity that the fan part of this blower is permanently attached to the motor. I disassembled a fan from the junkyard as far as I could, and did get access to the back of the motor. So it was possible to add a bit of oil there. Unfortunately the cage won't come off, so replacing the brushes or lubing the front bearing isn't possible. (Given the crap ton of powdered graphite that fell out during disassembly, I am pretty confident this unit has graphite brushes.) New replacement fans are stupid expensive for what they are.

 

I beg to differ, my assumptions hinge mostly on the ECU reported fan mode that it is trying to set , i can clearly see in on techstream as in my previous image, and the fact that the fan is either not turning at that speed, or not turning at all -is the issue.

This is a clear indication that the ECU 'admits' that it detected a high enough temp for speed 4 , for instance as in the pic, but based on both my hearing, and the feedback voltage, it is 100% not turning at that speed.

In one of my early images you can see auto mode, fan mode 4 ( as determined bu ECU) and temp of about 120f. This was an already reduced temp after i screwed aroung with the fan speed after it reached 135f while driving in auto mode, with absolutely no sound from the fan, as i put my ear right at the duct.

Following that, when i set the speed to 4 via active test on techstream, the fan starts to work fine and is very loud at speed 4, plus, the feedback voltage goes down as it should, and temp finally goes down.
Switching back out of active test either turns the fan off, or turns it on at much lower speed than 4, even though the ECU is still trying to set it to mode 4 per techstream.

So... i dont think it can be any more clear than that ,after driving a few days in hot weather and crazy santa cruze downhill regen charging, dr prius is able to keep the battery cool , which meams the fan is fine, the system that controls it isnt. I am suspecting the PWM signal and some noise that does not allow the low duty cycle to reach minimum pulse width...

I hope to feel better soon and probe this for you guys, in the meantime , the dr prius auto temp and fan speed regulation is working well, for some weird reason , since in theory it uses same PWM controller that turns into DC voltage for the fan

Hope the issue is more clear after the above summary
Thx!

 

I still wonder, if the cars not getting what it asked for, why no code(s). Unless it’s gone HAL.

there “could” be a reason, but something to consider.

 

In post #20 it's suggested "per chatgpt" that the VMF feedback signal is routed to the ECU so it can be reported to a scan tool for diagnostic purposes, but isn't used internally for closed-loop control or for triggering codes.

To me, seeing anything "per chatgpt" in a technical PriusChat post screams ELIZA Effect ... modern LLMs are certainly a lot more sophisticated than the original ELIZA, but I still shake my head at seeing them expected to answer a specific technical question about how a specific car works ... especially when the car's technical docs are paywalled as Toyota's are, and so probably all excluded from the masses of internet content that got slurped in to train the LLMs.

So I might not be strongly convinced that claim about how VMF is used is necessarily right. But maybe it is,

 

Hi Mendel, the reason why there was no code, is because im kind of machines freak, and i am very aware of what the car does, on a super hot day, i often monitor the battery temps, and i noticed 135f going back home whixh is abnormal for the 12 years i own the car, while the fan is completely silent with my ear right on top of the duct.

So you could say that there was no fault because i believe i caught it on time, or because my drive was just short enough, If my drive was longer, or more of a stop/start drive, i dont see how the temp would not reach the maximum possible, because the fan was completely not working - 100% , trust me...

Chapman, I'm with you on chatGPT, however, from the pdf that Mendel kindly provided, the VFM is only mentioned twice,just some basic voltage range , and voltage at speed # 1, which is very similar to the voltage i was getting when my ECU ordered it to go to speed 4 and it was about speed 0 , or 1, which match this documentation perfectly.

Perhaps there is monitoring of this Voltage , but it is not clear how well it is done, and how to validate the threshold or delta from target.
Even though VFM is not mentioned with any DTC, P0A82 or P0A81 may suggest the usage of some voltage , even though VFM is not specifically mentioned, and no clear guidelines of target voltage range for each speed is specified.

It would be interesting to disconnect the controller and see if it even throws a code when the ECU is sending PWM, and the VFM is not reaching back to the ECU. It definitely doesnt work right in my car though- per the VFM diff between manual and auto modes.
Maybe it just didnt hit a threshold, or the connectors corrosion is affecting something with this feedback
Thx