Discussion in 'Gen II Prius Modifications' started by ken1784, Aug 15, 2007.
Can this hack be mod'ed to go immediately to S4 ??
If not, then why?
My auto temp hack project has stalled at the moment. Moving into a new house and I have a trip to Hybridfest in July coming up shortly. After things start to settle down I will try to get the ball rolling again. In the mean time I'm still just using the basic POT resistor combo.
You might want to install a manual EV button since you have a short commute. You can go into EV at startup bypassing the warm up stage. If you do that you still may want to let the car go through S1 every other day or two just to keep things lubed up.
No immediately. For best relults the actual engine temp must be 100F+ before you crank it up to 159F. If you dont let the engine get to 100F it will start to stutter if you fake the temp. You can however get into S4 after you go through S1. You could get into S4 in as little as about 2 minutes. Thats just my guess. I never have the opertunity to try it out on my commute. Plus I have no need for S4 on my commute with the temp hack. I think I mostly stay in S3.
I do have the EV mod from coastal tech, but it will display "can't change to EV mode now" at startup - seems I have to get to 125 F before it will engage. So my engine runs for about 2 minutes and then I'm electric. I'm still getting great mileage, just was wondering if I could bypass the gas engine with the short trips when there is plenty of hymotion battery.
Ok, so its over a year now that I have been driving with the hack and: No obvious damage so far to my engine, and way better mileage. I have forgotten it it's a hack, and think of it as part of how the car behaves. I love it but obviously Iâ€™m biased.
Answers to questions:
1) Why canâ€™t you just jump right to S4?
The system tricks the computerâ€™s temperature sensor. BUT Temperature is used for things other than the hybrid warmup cycle. Specifically engine timing and air fuel mix. If you fool the car into thinking it is warm when it is in fact too cold, the engine will run progressively rougher and eventually stall. I imagine you would eventually also damage the engine. Also, if you spoof too early the computer will notice, it will throw an engine error light and it will flag your temp gauge as broken. You will need a scan tool to reset the light and tell the engine the temp gauge is fixed.
That said, by adjusting the "start spoofing" potentiometer, you CAN do it, I tried it, but only once.
2) What transistors? They are just standard general purpose silicon transistors, almost anything would do, but two are NPN, and two are PNP, you must get that part right. The actual ones I used were: NPN: 2N4124 and PNP: 2N3906.
3) What is the not gate?
Note that the thick dots on the diagram are just connection points. At the point pointed to by the "not gate" arrow, the voltage will be high when the temperature THW pin is low, and the voltage will be low when the THW temperature pin is high. Thus at that point the voltage is the inverse (or "NOT" function) of the carâ€™s temperature signal.
4) Will it work on a prius?
Obviously it works on a highlander, I have read the prius factory repair manual, and the temp sensor circuitry is pretty much exactly the same as far as I could see. In fact this circuit should work in probably any car. The only difference is the pinouts for a prius and a highlander are different. You must connect your circuit to the right points.
If you do try to build this, do it on a breadboard, and play with it, understand it, make sure itâ€™s working, before installing it in your car. Use a resistor to simulate "CAR" and a potentiometer to simulate the engine temp sensor.
I know itâ€™s a complexish circuit, but it really isnâ€™t that hard to build. The majority of the complexity is due to my wish to not have the real car temp signal compromised anywhere but in the region I wanted it to be spoofed. I tried a lot of simpler circuits, and found my cooling system trying to max itself out before my car was even at temp. Unwanted voltages created by simpler circuits can have unwanted effects on your car as the temperature sensor of the car is very sensitive.
Not shown in the diagram, but in the picture, I used a single pull double throw switch so that both the points at which the circuit connects to the car can be disconnected at the push of a button.
Is your circuit posted somewhere? My take on this is it really needs to be a microcontroller which reads actual engine coolant and oil temps, and then spoofs the car. And it would have selections for like PHEV , Charge Sustaining and Short Trip modes of operation.
The circuit is in the "attached thumnails" of my original post, back on page 2.
I had thought of doing this the microcontroller way, but this seemed simpler. Had this not worked, I was going to do just that. Ultimately, even the microcontroller is going to turn on and off a reference voltage, all I did was use some transistors to do the switching, since the overall logic is so simple:
Engine below T1--> no spoofing.
Engine above T1 but below T2 --> Spoof T2.
Engine above T2 --> no spoofing.
All the circuit really is, is two reference voltages, and transistors that switch on or off depending on the engine's temp relative to T1 and T2 thus enabling or disabling the simple spoof circuit.
However with a microcontroller, and maybe a bit more info from Toyota, you could do some pretty cool spoofing.
One major improvement that might be possible: Maybe you could only spoof when the ICE was OFF, or the throtle being released. IE turn the spoofing off when the engine is supposed to be running, and only use it to keep the engine off during coasts and stops. Thus the fuel mix and timing would not be affected by the spoofing.
Even better, a microcontroller might be able to just send the appropriate CAN message to the car's computer that warm up was complete.
What I'd really LOVE to do, is fool around with the ECU code and tweek it so that the car does what I want it to do withough having to fool it, but without Toyota giving me accsess to the code, and documentation, (not a chance) I this will remain a pipe dream.
I'm wanting to build your design but I'm not sure on what some of the stuff is in your schematic.
Can you tell me what the transistor values are? And can you tell me what the "NOT" gate is? Is it just a connection or is there an actual NOT gate there?
Ohm's Law for Dummies.
OK... think I'm about ready to do the hack. Had a few "Ohms Law" type questions.
If the Prius uses a [ame="http://en.wikipedia.org/wiki/Thermistor"]thermistor[/ame], why is the data I'm reading correlating certain temperatures to certain voltages, instead of correlating temperature to a certain resistance, as I would think most thermistors would? With that said, does anyone know the resistance across the thermistor at 40Â°C verses the resistance across the thermistor at 70Â°C and 90Â°C. Also does the thermistor's resistance raise with an increase in temperature, or lower? Since this hack has the resistors in parallel, I would think the resistance must drop as temperature increases.
I know others have suggested a 240Ω resistor with a 1K POT in series with a switch, but aren't there some POTs with a built in switch, turn it all the way counter clockwise and a bit further to click it off? Would a really big logarithmic POT (1M) with a built in switch work? Turn it on and it's at 1MΩ with almost no effect on the original circuit, then as you dial down the resistance on the POT, the effect becomes more substantial.
Anyway... Thanks to everyone for coming up with this hack!
Re: Ohm's Law for Dummies.
OK, sleep cleared it all in my head. I looked at Bob's page again and saw the other resistors behind PIN 19
So effectively what you have (sorry no art skills here) is:
R1 = Internal resistor
R2 = Thermistor
a = Sample Point
b = Sample Point
Now the ΔV between "a" and "b" is I*R2, and 5V=I*R1 + I*R2, so I = 5V / (R1 + R2)
so the ΔV between "a" and "b" is 5V * [R2 / (R1 + R2)] or
5V * α where α = [R2 / (R1 + R2)]. This makes since since as R2 approaches ∞, α approaches 1, and as R2 approaches 0, α approaches 0.
Anyway, I got the range of the Thermistor (R2) from Hobbit's graphic as between 79Ω (@+140Â°C) and 156KΩ (@-40Â°C), but I don't know what it's "operational" range is at temps between 40C and 95C. I also have no clue what value R1 is either.
Re: Ohm's Law for Dummies.
My experiment tells me the R1 value is approx 2.6kΩ.
Re: Ohm's Law for Dummies.
My complete and WAG puts R1 between 7K-10KΩ. That would make the current range between 0.03mA and 0.70mA. Does this sound right for this type of curcuit? Seems kinda low for curent.
Re: Ohm's Law for Dummies.
First guess was way off. From hobbit's page on the subject sounds like R1 is known to be 2500Ω and R2 is VERY non-linear. So by plugging some numbers, I'm thinking a 5K POT in series with a 500Ω resistor may be a better way to move between 40Â°C 70Â°C. If my math is right, at 40Â°C if you turn on the circuit you will blink 5500Ω in parallel. This should pop your temp up to about 46Â°C (1.30V). Now if you dial the POT all the way clockwise you should be at 75Â°C (0.60V). On the other end... when the thermistor is registering 70Â°C and your POT is dialed as close to off as you can get it, your will be down to 500Ω of resistance, giving you an "adjusted" temp of about 72Â°C (0.64V).
As a side note, looks like you could dial in 70Â°C in as soon as ~27Â°C (but you really should wait for 40Â°C).
Re: Ohm's Law for Dummies.
Thanks Ken... sorry I missed your reply.
Dan a 1k linear POT and 250ohm resistor works just fine. If you want a little safety in the winter use a 520ohm resistor. I find that the 520ohm resistor will give you about 160F when the actual coolant temp is around 100F. This will tell you if you need to start the engine you might want to turn off the fake temp so you dont run the chance of an engine stalling or getting error codes. If you want to stay in S4 in the winter for long periods of time like me since I have the extra battery the 240ohm will keep you there for ever. But if your actual temp drops way below 100F and you try to start the engine with the fake temp you might run in to trouble.
For my commute 240ohm would not help since when I get off the highway I'm traveling below 35MPH and dont really need S4.
Very cool. Thanks,
Think I'm gonna go with a 1k linear POT with a 240 or 520 resistor. The idea of going with a 5k POT looked bad once I graphed it. Got real non-linear. Solution would be a 5k reverse-logarithmic POT (left side of the dial has a bigger effect than the right side). Finding a 5k reverse-log POT is not as easy as getting a 1k linear, so decision made.
I'm also probably gonna get a POT with a built in switch. Just one component to install.
As a after thought, I was considering putting an LED or something in the circuit, but since the current is so dang low, I doubt the thing would glow at all. Neat idea though.
I had a hard time finding a 1k POT with switch. I just checked and the place I bought mine no longer has them.
This place looks to have them. Dont know how long the shaft is but you can always cut it if its too long. Plus you will need a knob if you dont want it to look like a hack job.
You have to buy $10.00 worth before shipping.
If you look at their catalog you might be able to get everything you need and come up with $10.00. I had to do that with the other place and ended up buying a few extra to sell.
I've been in contact with jstcd on his automatic design and I under stand it more now. I have parts on order to make a few and I should have a protype built hopefully in the next few weeks.
I also found that the internal ECU resistor is about 2.5K.
For my fakeout I went with 470 ohm in series with a 5K linear pot,
which gives me a little more range but "safeties" the thing so it
doesn't drive the temp way up into fan-running territory. And a
separate toggle switch to cut the whole thing out once things are
genuinely warmed up. It'll get a mention in the next biggish
maintenance writeup I'm working on.
I'm including the latest post I just made on the Coolant Temperature Hack with some updated information.
Re: thermistor hack
Some updates to my temp hack
1) now been using it trouble free for quite some time, no engine explosions, of course legal warning, that may not be your experience.
2) The capacitors on the diagram should be 66 not 660
3) If you want to put in a switch (I did to turn it off, but in practice I always leave it on) I used a single throw double pull swith to diconect both the red and purple lines from the "car". You could get away with just switching the conection between the purple line and TWH
4) This was built using 1/4 w resistors
5) Had some conversations with a guy who built one of these for a phev. This circuit will prevent itself from spoofing until the car reaches minTemp. (If it turns on too early the engin really runs rough, then stalls and throws codes etc, not good) However once spoofing, it will continue until the car is turned off, which resets the circuit. In a phev, the engine can "recool" making this particular circuit a poor choice. In a normal hybrid, the engine will tend to stay at temp once it gets there.
6) This circuit will boost your apperent temp only if the temp is lower than that. If the temp is higher, you will seee the true temp.
7) To adjust the potentiometers:
The set temp: usingh a scan guage or your car's temp guage, turn car on, but engine off. adjust tmin pot until temp system trips and begins spoofing. Now adjust the settemp pot until the car't temp appears to be where you want it. (Which should be just under normal running temp, but over the temp you need to get to hybrid mode3)
Next set the min temp pot, again car on, engine off, adjust so that system fires when temp guage just comes off its peg. (once fired you must turn off car to reset)
Now run car, if spoof turns on to soon, (ie engine begins to run rough) add more minT, if not try taking some off.
Temp setting works best in conjuction with a scan guage that can read the car's apparent temp more accurately than the car's guage, and you can add a pot accross ground to TWX (20K should do) to simulate car temps and ensure operation before actually turning onthe engine.
8) When ideally set the spoof allows you to skip mode #2, getting to hybrid operation sooner.
the cold car should:
run as per normal in mode#1, and when warm enough enter mode #2
Soon after entering into mode #2 the spoofer should activate, the suddenly warm car is allowed into mode #3.
The car will switch to mode #4, the most fuel efficient mode, the next time you stop for over 10s (ie red light)
Mode #3 is the worse fuel economy mode, so it is good to make your brief stop once you see the needle jump.
9) This hack is infinitly superior to a simple manually adjusted pot, as said hacks need to be trimmed all the time, but there are now even better once based on microcontrollers out there.
Separate names with a comma.