Electronic Switch for Heater Core Bypass Mod

[ChrisH]

This electronic switch is a modification of Tony Wa’s Heater Core Bypass. (Original thread at Heater Core Bypass Mod (Tony Wa) [How To] )

This electronic switch modification uses the signal from the temperature control knob in the cabin to determine when the control is at the cold setting, and activates the bypass of the heater core when the control is set to cold. Hence, no mechanical switch is needed (neither the base rocker switch or the optional push/pull potentiometer from Tony that activates the bypass when it is pulled). The switch unit is completely contained in the same forward compartment as the blower, so if you are installing the bypass for the first time, there is no need to run wires into the dash or cabin area. In my case, I also didn’t want to remove the radio head-unit again because it was very hard to install last time (after-market head-unit with iPod interface made for a crowded mass of wires in the dash behind the head unit).

I wanted to create an electronic solid-state switch rather than a relay so as to avoid mechanical reliability problems that can occur with relays. I also wanted the bypass to be activated only when the temperature control was really set to cold, which means all the way counter-clockwise, plus a click or two on the knob back clockwise for some margin. I measured the voltage on the control line at its connector with the flap actuator in the blower compartment, with the flap actuator connector unplugged, and determined that a level of 1 Volt on the control line would be a good point to switch the bypass. Note that this voltage might change if the connector was plugged into the actuator because of the current draw from the actuator. However, I didn’t measure that case. I simply made sure the flaps were set for full heat airflow, and then disconnected the plug from the actuator. I had to move the flaps some by hand because my flap mechanism is binding (hence the need for the bypass kit). My measurements of control line voltage with the flap actuator unplugged are as follows:
Engine running:
All the way cold: 0.13 V
1 click up from cold: 0.78 V
2 clicks up 1.34 V
3 clicks up 1.86 V
4 clicks up 2.38 V
5 clicks up 2.93 V
6 clicks up 3.44 V
7 clicks up 3.93 V

To create a voltage sensing switch, I wanted to use a voltage comparator (or Operation Amplifier operating as a comparator) to switch when the temperature control voltage reached a set level (in this case, about 1 Volt). The comparator would then drive higher current transistors to feed electrical current to the vacuum switch. I initially looked at finding the individual comparator and transistors to create the circuit, but then realized that a simple single chip integrated circuit should be available to do this since it is a pretty common need. And, sure enough, I found some. I decided to use the ST Microelectronics TDE1798, which was available at Digi-Key (for $7.23). This part has sufficient current capacity, a broad temperature range, and protection for a variety of real-word problems, such as over heating, short circuits and built-in demagnetization energy clamping for solenoids or motors that it is driving (the vacuum switch is a solenoid). I determined the current needed by measuring the resistance of the vacuum switch (38 Ohms), and assuming a 14.5 V level from the alternator when the engine is on. This yields 0.38 Amps. However, the transistors in the switch have a breakdown voltage of just over 1 V, which means the voltage to the vacuum switch is more like 13.5 V, and the current is more like 0.36 A. The TDE1798 is rated at 0.5 A. It is an 8-pin mini-DIP package, and I used a socket as well (in case I managed to fry it). (You can get an 8 pin mini-DIP socket at Digi-Key for $0.48 if you don't have one - Digi-Key part number ED3108-ND.)

The basic circuit diagram is below. The 13.5 K Ohm and 1 K Ohm resistors in series create a voltage divider that sets the approximately 1 Volt reference level. This was based on a typical 14.5 V operating voltage, not 12 V as shown, when the car is running. (Note, I didn’t have a 13.5 K Ohm resistor in my electronic junk box at home, so I simply put a 10 K Ohm resistor and a 3 K Ohm resistor in series to create 13 K Ohm, which was close enough as precision is not needed.)

I connected the reset input to the switch via a 20 K Ohm resistor. This will reset it in case it ever has a problem detected that causes a shut-down.

I used a circuit prototyping board from Radio Shack (part number 276-168, about $3), and cut it down to a size that would fit in my enclosure. I used a small rugged aluminum enclosure to contain the circuit, also bought from Digi-Key (part number HM570-ND, $8.34). I mounted the circuit board with short standoffs, slightly too short so as to press the chip against the top of the enclosure in order to help keep the chip cool. (I made the standoffs by simply cutting some stiff plastic tubing - from a liquid medicine dropper tube.) I used 0.75 inch long #6 pan head screws with nyloc nuts. The nuts were on the outside due to space issues, and I used nyloc because I didn’t want to torque them down so tight that it stressed the circuit board or plastic stand-offs, but still kept the switch part pressed against the aluminum enclosure. A picture is shown below. Please note that since that picture was taken, I swapped the connections to pins 2 and 3 because it turned out I had the polarity reversed originally (a hot setting was creating cold, and vice versa).

For the +12 V connection to the car, I used the clamp on connector Tony provided in his kit on the passenger side window wire on the back of the fuse box (fuse 4). However, I substituted the crimp on blade connector Tony provided with a fully insulated blade connector. I used fully insulated connectors throughout the installation. I did use the 3 Amp clamp-on inline fuse that Tony provided.

For the temperature control signal, I simply spliced in a wire into the yellow/green temperature control signal wire going to the flap actuator of the blower. It is in a group of 3 wires that go to the actuator, and is visible just above the heater core.

I didn’t hard mount the enclosure. I simply tucked it into the space and used a zip tie to the large wiring harness to keep it from floating around too much (same way the vacuum switch was mounted). I tucked the wires appropriately to keep them straying into hot areas or other undesirable places.

The result is, it works.

It is more work than simply using the switch in Tony’s kit. However, its operation is now transparent to the user – no rocker switch or knob to pull, and, I didn't have to pull my head-unit out again. In my testing, the bypass was active when the temperature knob is at its coldest setting and 1 click up, but is not active (that is, there is heat) when it is 2 clicks up from cold and above, exactly as intended. By-the-way, it does take almost a minute after going from hot to cold for the heater core to cool down with the blower set at high. It heats up much faster.

Now, I just need to use it for a while to make sure it survives.

[doug_porsche]

way cool, or hot depending on the voltage!

one for the winter project file.

[dasmoofler]

No bypass cap?

[ChrisH]

Originally Posted by dasmoofler

No bypass cap?

Not needed in this application. A little ripple doesn't matter, and there is already plenty of bypass capacitance elsewhere in the car's electrical system.

[JustAndyC]

I plan on installing the core bypass mod this winter so your write up here was perfectly timed- thanks a bunch!

[dasmoofler]

Originally Posted by ChrisH

Not needed in this application. A little ripple doesn't matter, and there is already plenty of bypass capacitance elsewhere in the car's electrical system.

I was thinking more in terms of noise protection rather than ripple. A .1uF cap is cheap insurance.

[ChrisH]

Cheap, easy to install insurance is generally a good thing, and bypass caps are pretty standard practice. However, I just didn't think it was needed in this application because of there is already a good amount of bypass capacitance in the car's electrical system anyway, and this part is good up to 50 Volts. Something pretty serious would have to happen to get a voltage transient over 50 Volts on the car's electrical system, and I think a lot more things would go bad if that happened. If I am wrong, well, it is an inexpensive part.

For anyone who is interested, here is a link to the data sheet for the intelligent switch.
http://www.st.com/stonline/products/.../tde1798dp.pdf