TPS relay trick (REALLY LONG)

Raffe,

I have some questions...
Your instruction mentioned "connec the red/black from ECU one to one side of the relay and the other red/black from TPS to the other side." Does it matter which side it is of the relay? (is polarity important?)

Also, on the relay, the wire to tap into for +12v (PCS) is a red wire (A-20) on ECU right? What wire color/pin location for the pressure switch?? Thx.

BTW, Great job!!!

As far as I know, the JR relay and the one for Rafe's trick are the same.

Eliminating the TA trick will cause CE problems unless you relocate the TA sensor to upstream of the blower.

Greg

I've kept my IAT probe from the intake stream since I installed the slater 3" short ram. I've cut down on the amount of knock (which was already minimal) by half. I connected a piece of 3/8" rubber hose and ran it down close to a fresh air source below the engine bay (i.e. next to the shift lever mechanism).

Notice that my intake doesn't have ports for the valve cover or IAT.
http://albums.photopoint.com/j/View?...nce=0&res=high

Combined w/ Raffe's trick I'm sure most of you will cut down the amount of tip-in w/ the IAT removed from sensing the actual intake temps.


------------------
Mods

beekayITR,

I'm not entirely sure about this, but I think you don't want the J&S armed until the engine is under load. Otherwise mechanical noises present in an unloaded engine (such as piston slap) can be misinterpretted as knock.

Holden,

I agree, Radio Shack should hire a few of us to build them an e-commerce site that actually works. For a company of it's size, I'm amazed that they can't do better.

randomwalk101,

1) "Polarity" is very important here. After cutting the red/black wire, the half that runs back to the ECU must be connected to the center (common) pole on the relay so the ECU always receives signal regardless of the relay's state. The other half, the one that runs to the TPS center terminal, should run to the relay pole to which the center relay pole connects when the relay coil is energized, thereby connecting the true TPS output to the ECU when the pressure switch is closed (no boost).

2) PCS power source: I'd wire this under the hood along side the enrichment relay that's part of the JR system. The pressure switch itself is mounted on the back (firewall) side of the JR manifold near the center. Don't confuse with the IAT sensor that's over toward the brake master cylinder.


[This message has been edited by rjaffey (edited October 24, 2000).]

Which wire is best for wiring all this...cant find it on the radio shack site for some reason..they ask too many questions..lol

rjaffey,
Is it possible to use the JR relay, and connect the red/black wire to the poles (haven;t checked to see if there are any unused poles)

It seems to be doing the same thing ie: switching between true TPS output and constant 4.5v (WOT)when 1psi of boost is reached.

Thanks

teamfx:

The relay socket has a decent length of wire coming off it. If you need more, just use any decent wire. Maybe a fatter (14 gauge or larger) for the 12VDC+ and ground wires for the relay, but the other wires carry very little current. 18 gauge or better should do it.

94Integra:

I don't think you can use the JR Relay. A "dual relay", if one even exists, would be needed to perform both the TPS and TA sensor tricks at the same time. Why would JR use a relay like this, which would cost almost twice as much as a "single relay"?

Greg

Greg,

You're right. I was actually thinking bout removing the TA fuel enrichment and just using the TPS trick so I wouldn;t be overly rich at part throttle.

Guess that might cause a problem with the TA getting high temps and causing the ECU to freak out.

But it is the same kind of relay, isn;t it?

[This message has been edited by 94Integra (edited October 24, 2000).]

Batoutahell,

Is the 4.8 volts what's listed in the manual, or did you take a measurement while holding the throttle wide open?

Be sure to take your measurements with everything hooked up. I jerry rigged an extension harness with pins that fit the TPS connector on the wiring harness and sockets that fit the pins in the TPS sensor itself, so I could take measurements and perform testing without butchering the real harness unnecessarily. You will modify the real harness when you're ready to do the mod itself.

All voltage measurements must be taken with the ignition in the ON position, but be sure to switch the ignition off each time you change the wiring around as you make tests. Be sure nothing shorts and nothing touches ground or anything else.

With the real harness disconnected from the TPS, check the resistance of the TPS sensor itself between the top and bottom terminals (not the sweeper which moves between the two as the throttle position changes). Mine was 5K Ohms.

Hook everything up with the test harness in place and ignition switched on, and check the voltage between the top and bottom TPS terminals. Mine was about 5 volts.

Check between the TPS ground and center terminals with the throttle held all the way open and take note of that voltage. I'll refer to this voltage Vw in the text below. Mine was about 4.5 volts.

To duplicate the WOT voltage between the resistors use the following formula:

Let R1 be the resistor connected to the top TPS terminal (Vcc, +5 volts on my car).

Let R2 be the resistor connected to the bottom TPS terminal (ground).

Let Vw be the desired (WOT) voltage between the two resistors.

Let Vcc be the supply voltage to the top TPS sensor terminal.

So Vw = ( R2 / ( R1 + R2 ) ) * Vcc

Using this formula, you want to select resistors such that R1 + R2 is significantly larger than the resistance of the TPS sensor as you don't want to increase the load on the Vcc supply any more than necessary. However, it's also important that R1 and R2 not be too large lest they not supply enough current back to the ECU on the TPS signal (what is normally connected to the TPS center terminal). From my testing, I determined that R1 + R2 could be as low as 5K (plus the load from the 5K TPS sensor) without pulling Vcc down below 5v. This would therefore be the absolute lower bound for R1 + R2 on my car. However, as it's probably a good idea not to be anywhere near the lower bound, I took a stab and tried an R1 + R2 that was well above it. As the ECU's TPS signal input probably has a fairly high impedance, the resistors can likely be even larger than those I selected.

After calculating the proper resistors, hook them into the test harness, hook everything up and verify that you see the desired Vw between the point between the resistors and ground, regardless of throttle position. Also verify that Vcc is still 5v.

For the final test, cut the TPS signal (center) lead in the test harness and connect the Vw (point between the two resistors) to the TPS signal wire going to the ECU. Then verify that Vw is the same as it was in the previous test. If it isn't, R1 + R2 is too large and the Vw supply has insufficient current to drive the TPS signal input on the ECU.

Assuming you're seeing Vcc = 5v and Vw = 4.8v my initial guess at what you want for resistors would be 22K to ground and 915 Ohms to Vcc. As 915 Ohms will be difficult or impossible to get I'd go with 22K and 1K, that will give you just under 4.8v. Remember, the resistors have a 5% tolerance. That means they can vary up to 5% from their specified values, so you could wind up a little above or below your target anyway. I think you should be OK if you're close though. Be sure you verify both Vcc and Vw before selecting resistors.

Another thing to consider is that the ECU may switch out of closed loop operation before the throttle is totally wide open. This may vary across models and years, so it should still be tested. I rigged another test using potentiometers to determine at what voltage my ECU switched out of closed loop. I don't have that number handy, but I know it was below 4.5 volts. I just figured I'd try to get as close to the real Vw as possible. What does this mean? It means you should try to get your target Vw, but it's probably OK if you miss it by a few percent either way.

Hopefully I haven't babbled too much.

Good luck.

[This message has been edited by rjaffey (edited October 26, 2000).]

Radio Shack shipped my stuff yesterday, so I'll be an OBD-II guinea pig next week.

Greg