RT

In order to run boost with the S2000, any and all positive manifold pressure must be hidden from the ECU.
The "traditional" method for doing this (Integra, Civic gang) is to vent any positive pressure via check valves to atmo. This has proven to be only a marginal solution for the S2000.

So, if we could come up with a device that would electronically shunt the boost from the ECU, it would be perhaps a cleaner, more reliable solution.
The problem is, I know only enough about electronics to know that I don't know how to accomplish this task. What?

Here's what I do know so far:

If you have a Helm Manual, the MAP Sensor shows up on page 11-29 in schematic format.

The supply voltage of 5V comes in on the Yellow w/ Red wire (passenger side wire on plug) the Green w/ White wire is Ground (center of connector) and the Signal (0V-5V) returns to the ECU via the Green w/ Red wire (drivers side of plug connector).
Note: Something weird about the Green w/ Red wire, it gets a biasing voltage of 4.8V from the ECU if measured with the plug disconnected. On page 11-29 it shows it

Rasputin314

Here is somthing that will work.

http://www.national.com/ads-cgi/viewer.pl/...ds/LM/LM431.pdf

The TO-98 plastic package, then create the simple Shunt Regulator circuit on page 7. You could take Vref from a stable internal 5V source, then adjust R1 to set the 2.866V.
The whole thing takes no space at all and could easily sit between a harness and a plug. Parts are probably less than $5.

So, do I win a free supercharger? j/k

shaner

RT,

Will my Mugen ECU fix the MAP overboost problem?

It's not installed but I would try it if it will fix the MAP problem...

Thanks!

RT

[QUOTE]Originally posted by Rasputin314
[B]Here is somthing that will work.

http://www.national.com/ads-cgi/viewer.pl/...ds/LM/LM431.pdf

The TO-98 plastic package, then create the simple Shunt Regulator circuit on page 7.

Rasputin314

Before I start rambling, let me say that even though I have a EE degree in fiber optics and robotics, I'm not an expert at cars, or MAP sensor, etc. Also, I dont have the Helm manual so am guessing at some of the system.

The shunt would go exactly between the MAP sensor and the ECU. The two ways to accomplish this are the clean way and the kludgy way. The clean way involves getting another set of plastic harness/connector parts. Honda seems to use a lot of parts from phoenix/amphenol so it may be a stocked part. Then, you can build the circuit that was on page 7 using the NS regulator, some radio shack parts, a solder gun, electrical tape, and seal it with heat shrink tubing.

If you didn't have the extra connector set, you can just cut into the wires before the harness and install it as a blister on top of the wire bundle.

About the 4.8V, you said it was connected to the bus voltage using a bias or 'pull-up' resistor, that is normal. It is probably a safety feature so that if someone pulls the cable out while the engine is running, it will retard the performance.

This is a pretty easy electronic mod to make, but make sure to test it with a power supply/DMM/scope before putting it in your car.

Bobway

[QUOTE]Originally posted by Rasputin314
[B]
About the 4.8V, you said it was connected to the bus voltage using a bias or 'pull-up' resistor, that is normal.

RT

Harry or Ras,
What's up with the Capacitor in dotted lines in the schematic on page 7 (Shunt Regulator)?

Rasputin314

The dotted line capacitor is optional. Like any capacitor between output and ground, it acts as a passive single pole low pass filter. It probably makes sense to use a 1uF to take out spikes as Harry said, but my guess is the MAP sensor can change pretty quickly so better not use anything larger than that.

Your next question is probably going to be what is the unlabeled resistor on the input. Well, it is just to remind you that the output impedance of the MAP sensor will affect the divider network. Use a high R potentiometer for R1 and a high R2 (>10k), and you shouldn't have to worry about it.

The output is limited to Vo which is listed as (1+R1/R2)Vref. There is one thing I did not notice earlier, that is Vref is required to be 2.5V (not 5V). Hopefully you can find 2.5V somewhere in the car. If not, a voltage divider from the 5V line will not work by itself, it would probably have to be followed by an op-amp buffer. This adds a bit of complexity to the circuit.

Too bad you guys are not in Austin. I could help you build and test this thing, and I have been very interested in riding in a S/C S2000...

davepk

Some more food for thought. The MAP sensor is an Absolute pressure sensor not a gauge pressure sensor. In other words changes in atmospheric conditions will result in different WOT threshold values... It seems to me that what is needed is a second map sensor vented to the atmosphere and a circuit to switch between which ever is lower in output.

RT

Dave!
That