another thermistor hack

[ken1784]

Warning: This modification may damage your cold engine.

Hello all,

Bob Wilson is doing a thermistor hack to reduce warming up fuel on his NHW-11 Prius.
http://hiwaay.net/~bzwilson/prius/pri_temp.html

I would like to report another solution to hack the current NHW-20 Prius thermistor, the THW sensor for engine coolant.
Please refer to attached picture.

Left picture: the original THW engine coolant thermistor sensor connection to the pin E4-19 of engine ECU.
The resistance value of the thermistor is approximately 1085 ohm at 40C, 430 ohm at 70C and 240 ohm at 90C.

Middle picture: a simple solution. You do not need to cut any wire. Just connect a switch, 1k ohm variable resistance and 240 ohm fixed resistance.
When the switch is off, the thermistor works as original, no modification.
When the switch is on, the resistance value to the E4-19 pin becomes combined value with the original thermistor and manual resistance value.
We only can reduce the combined resistance value on this simple solution.

We know Prius coolant temp needs to be above 40C to move to S2.
So, if we could make the combined resistance value to 1085 ohm or smaller, Prius thinks it doesn't need initial warming up and moves to S2 directly.
If we could make the combined resistance value to 430 ohm or smaller, Prius thinks it already finishes warming up and moves to S4 directly.
This simple solution only reduces the combined resistance value. We can't increase combined resistance value, which means we can't fake it is cooler than the real coolant temp.

Right picture: a complex solution. You have to cut the original THW wire.
However, You can fake any coolant temp as you like.
When switch to true position, the thermistor works as original, no modification.
When switch to fake position, you can let the ECU know any temp you like.

Enjoy,
Ken@Japan

[gazz]

<div class='quotetop'>QUOTE(ken1784 @ Aug 15 2007, 10:15 AM) [snapback]496123[/snapback]</div>

Warning: This modification may damage your cold engine.

Hello all,

Bob Wilson is doing a thermistor hack to reduce warming up fuel on his NHW-11 Prius.
http://hiwaay.net/~bzwilson/prius/pri_temp.html

I would like to report another solution to hack the current NHW-20 Prius thermistor, the THW sensor for engine coolant.
Please refer to attached picture.

Left picture: the original THW engine coolant thermistor sensor connection to the pin E4-19 of engine ECU.
The resistance value of the thermistor is approximately 1085 ohm at 40C, 430 ohm at 70C and 240 ohm at 90C.

Middle picture: a simple solution. You do not need to cut any wire. Just connect a switch, 1k ohm variable resistance and 240 ohm fixed resistance.
When the switch is off, the thermistor works as original, no modification.
When the switch is on, the resistance value to the E4-19 pin becomes combined value with the original thermistor and manual resistance value.
We only can reduce the combined resistance value on this simple solution.

We know Prius coolant temp needs to be above 40C to move to S2.
So, if we could make the combined resistance value to 1085 ohm or smaller, Prius thinks it doesn't need initial warming up and moves to S2 directly.
If we could make the combined resistance value to 430 ohm or smaller, Prius thinks it already finishes warming up and moves to S4 directly.
This simple solution only reduces the combined resistance value. We can't increase combined resistance value, which means we can't fake it is cooler than the real coolant temp.

Right picture: a complex solution. You have to cut the original THW wire.
However, You can fake any coolant temp as you like.
When switch to true position, the thermistor works as original, no modification.
When switch to fake position, you can let the ECU know any temp you like.

Enjoy,
Ken@Japan



[/b]

Thanks for that I might give it a go.

[bestmapman]

Originally Posted by gazz

<div class='quotetop'>QUOTE(ken1784 @ Aug 15 2007, 10:15 AM) [snapback]496123[/snapback]</div>


Thanks for that I might give it a go.

So, did you try this? If yes, how does it work?

[gazz]

Originally Posted by bestmapman

So, did you try this? If yes, how does it work?

No not as yet, just posted another thread on first 1 mile, in which using "N" during the warm up stage (negative ign angle phase) seems to have made a positive difference.

[douglas001001]

Are you using this to be able to initiate glide between 35-40 mph before s3/s4 temp is reached instead of only having ev glide available below 34?

[ken1784]

You can glide between 35 and 40 mph in s2 too.

Ken@Japan

[Presto]

Very nice! I was hoping a hack would come out for this. Great work, Ken and Bob!

[hobbit]

Wow, I must have missed this thread the first time around [but
I did see all the discussion over on prius_technical_stuff, and
all the fun Bob was having..] The funny thing is while developing
the THW "linearizer" for the gauge, I wound up paralleling in
a test pot and running it a bit too far down, telling the car
that the coolant temp was WAY over 100C. Fans roared to life,
and the ToD popped on... but all that went right off again when
I brought the fake temp down to a reasonable level. The fans
seemed to kick on around 105C, if I remember.
.
Since I just use EV mode to work around various non-shutdown
issues in the early stages, I haven't played with THW fakeouts..
.
_H*

[bwilson4web]

Originally Posted by ken1784

Warning: This modification may damage your cold engine.

Hello all,

Bob Wilson is doing a thermistor hack to reduce warming up fuel on his NHW-11 Prius.
http://hiwaay.net/~bzwilson/prius/pri_temp.html

. . .
Enjoy,
Ken@Japan

I would like to suggest one important and one "nice to have" modification to the middle circuit:

• forward bias signal diode with 240 ohm resistor
• low temperature disabler on "fake" (parallel mode)

Forward bias diode

One of the Prius Technical Group posters had suggested using a forward biased diode in the resistor network. The nominal 0.65V forward bias threshold means that when the thermistor reaches about 80C, the voltage across the diode will take out the adjustment resistance network. This is completely automatic so any high temperature events will be accurately detected.

Low temperature disabler

Originally suggested by Cor, the forward bias diode could be replaced by a pair of transistors so one keeps the resistor network transistor off until the thermistor reaches 30-40C. This would replace the mechanical switch and with the built-in, Si forward bias, fully automate operation. It should look something like this:


If someone has a working copy of SPICE, they could quickly optimize the design. Note that once the bias circuit enables, it also latches the enabler transistor, T1. I cheated by using a microprocessor to software control the thermistor bias and record the performance. It also allows me to software control disabling the bias when the ICE cools off, which this circuit lacks, and deal with initial states.

When the car is parked in "READY" facing a strong, cold wind, the ICE and coolant will get colder and colder. Once it reaches 60C, my 2003 Prius ICE turns on until it is warmed to 70C. With the microprocessor solution, it monitors the real ICE temperature and stops adjusting to allow the ICE to come on. I don't know how this dual-transistor circuit behaves during "READY" cooling.

Having a lower temperature, cut-off threshold may not be a problem. As the catalytic converter cools, it MAY on its own trigger ICE operation to keep them in an operational range but I have not done any testing with this circuit. The problem is if the ICE cools below 30C, I've seen unreliable operation including stalls.

One also has to be sensitive to startup voltage transients. You don't want T1 to 'turn ON' before voltages have settled. I suspect there is not a problem but a cap on the reference voltage network, lower resistor, might be a good idea. My software solution has a delay before it starts operating.

The simpler circuit may be a better answer but it needs more engineering and testing than I've been willing to invest. If someone has SPICE and an interest, I would recommend giving it a try and seeing how well it works.

Bob Wilson

[Dan.]

Fantastic stuff guys! I few questions for the laymen in the group. The splice suggested in the "middle" diagram would require a splice "upstream" of the actual sensor. Is there an "easy" way to get to that part of the circuit. A possible simpler solution would be a "downstream" splice. My hope is that the sensor line goes into the ECU under the dash (not sure if this is true). Now if you can splice the line downstream under the dash, then you could hook it up to a reostat control (possibly next to the EV button). The reostat would add a positive delta to the "true" temp. Then you just dial the delta up or down as desired.

The "driving experience" I see is as follows.

1. Turn on NHW-20 then switch into EV.
2. While in EV adjust the reostat until the temp on SGII reads 85c (185f).
3. Drop out of EV and when the ICE lights it should jump strait to S4.
4. At the end of each pulse when the ICE is off, you dial the reostat to a "final" temp of around around 75c (165f)
5. During each glide you could also dial down the "delta dial" to zero to get a reading of the "true" temp.

Here are some strange S1 behaviors I've seen that might be of interest to this mod:

• IGN +6 in S1? New Prius Hack?
• New S1 Veto Prius II technique

11011011