CTSC 9 lb./High Boost Question

[emvanderpol]

Just wanting to verify the injector size for this kit, as well as the pulley diameter for the 3.0L engine.

Also, any good/bad/ugly on this setup would be greatly appreciated.

I already did a search.

Thanks,
Gene

 

[BryanZublin]

I believe the 9 psi Comptech SC kit uses fuel injectors from a Japan domestic market (JDM) Honda Prelude. Stock NSX injector flow rate is around 240 cc/min and the JDM Prelude injector flow rate is around 280 cc/min? I'm sure someone here has the real numbers.

The good: more power.
The bad: Still no ignition retard?

 

[emvanderpol]

Brian,

Thanks for the reply. However, regarding the ignition timing, the ESM when adjusted properly, will compensate. I unforntunately found out the hard way. But it can be done.

Thanks again,
Gene

 

[len3.8]

Gene, I can measure the outside diameter of the pulley that I have. I have the 9lb kit ,I don't have it on the car yet though.
I will measure and post for comparison.
Len

 

[matteni]

However, regarding the ignition timing, the ESM when adjusted properly, will compensate. I unforntunately found out the hard way. But it can be done.

Can you post some of the details of this? What happened? What did you do to adjust the ESM correctly?

 

[emvanderpol]

Can you post some of the details of this? What happened? What did you do to adjust the ESM correctly?

No Problem!!!

Do a search on my username in the Forced Induction section, and you'll find out the long story (dyno charts, A/F ratios, Injector stuff, etc.etc.). The short story was that I got a used CTSC, which had one of the first 50 manifolds/air inlet's that Comptech made. It was also the manifold that fit the OBDI engine, i.e. no air injection. After some time I found that the car was detonatiing, tried everything to fix it, even went and bought a brand new blower to see if the air injection would fix it. It didn't. It turned out that the ESM was not adjusted properly. Just a few turns of the screw and running perfect.

I was told by Comptech that the original maifold/air inlet generated more power due to being hand made, and that they allowed more CF/M through the engine. The dyno number didn't lie. Same 3.0L engine/6 psi system/ambient conditions/blower:

Old manifold/inlet = 360 hp & 272 tq
New manifold/inlet = 340 hp & 250 tq

I kept the old manifold/inlet, but new blower. I had to experiment with the ESM to find the "sweet point." According to Comptech factory ESM settings are at 2.92 volts, but I found 2/100th of a volt not only will elimintate detonation, but add/subtract 10-15 hp. I believe that the precise voltage may be vehicle dependant. Here is what I found with the new manifold/inlet (lower hp/tq).

Right now my ESM is at 2.950, and running perfect under all kinds of loading conditions.


NOTE: These are just my findings. To find your "sweet point" go to a dyno and spend alot of time bending Shad's ear. Don't just arbitrarily adjusting your ESM.

 

[BryanZublin]

Gene,
If I am not mistaken, the "ESM" is an electronic voltage "clamp" that installs between the stock MAP sensor and the ECU. It prevents the ECU from seeing MAP voltages higher than what it expects to see in an NA NSX. This is required so that the ECU does not flag an error (and shut down) during boost. Please correct me if I am wrong.

The circuit I described above will not directly affect ignition timing when under boost. However, if it is adjusted to too low a voltage, the ECU will never see "wide open throttle" and will not provide the largest programmed injector duty cycle & probably also slightly less ignition retard (ie. more ignition advance).

I designed a similar circuit called the SmartMAP that was used on the early BBSC design (when it used a raising rate fuel pressure regulator like the Comptech SC). The clamp voltage is not adjustable and is fixed at 2.75V -0/+0.15V. This set point was determined from measurements of the MAP voltage on a stock NA 1992 NSX during wide open throttle (Edit: or more accurately, with the engine stopped, ie. 0 manifold vaccum). Split Second has a similar product.

You mentioned that a 0.02V change in the set point (from 2.92V to 2.95V, actually 0.03V difference) had a significant impact on measured power and detonation presence. I find it incredible that your setup is that sensitive to the MAP voltage. 0.03V out of 2.95V is only a 1% change. This is probably less than the accuracy / repeatability of the MAP sensor itself from one vehicle to the next. If your detonation is now gone, that's great. But I have a feeling that you are still on the edge of detonation, and any shift in input conditions such as slightly higher intake air temp or slightly lower octane fuel will result in detonation once again.

 

[sjs]

Indeed, Bryan's SmartMAP solved a very similar problem on my Bell TT (before I got the AEM). I was using a Hobbs switch and solenoid valve to cut off the vacuum line to the map sensor when MAP hit atmospheric. At least theoretically. After much frustration with detonation, especially on fast up-shifts, I concluded that during rapid transitions the long lines to/from the solenoid were trapping a bit of vacuum. That meant the MAP sensor didn't see atmospheric and therefore sent too low a voltage to the ECU, which in turn didn't think I was WOT. I was looking for a valve with another pole so I could switch the MAP sensor side open to atmospheric when I discovered the SmartMAP (for 50% of the CT price) After months of pure frustration ,a few minutes to install it and the problem was gone.

As Bryan said, your fix is actually letting the engine know that you really are at high engine load so it will supply fuel accordingly. However, I suspect that since it thinks the load is less than max it may also not be backing off the timing, which I believe it probably does at high RPM and max load.

My suggestion for setting it is to keep increasing the voltage until your ECU trips into limp mode at max boost. That means it saw some boost, which it does not expect, and of course that fact is the reason for the ESM to begin with. Once you hit that point, back it down slightly until it never trips under any load. That should be the ideal set point.

 

[spdntckt]

Somewhat off topic.. but with the 9PSI CTSC Setup, do folks find they they still are able to pass smog fine here in CA? Wondering if the larger injectors will cause a rich condition during normal idling, light load (during smog sniff tests)..

anyone try this?

 

[emvanderpol]

Thanks for the replies.

Brian:

To clarify, take a look at my chart. The hand written numbers represent the ESM voltage. There also is the question of output votage that the ECU sees, which is about 0.2 volts less than the 2.92; kind of odd. You can see it is the 0.02 volts that made the knock go away (2.90 to 2.92) for the new style manifold/inlet. I had to go 0.03 volts over the 2.92 with the older manifold (higher hp & tq, more CFM) to find a balance. Also, take a look at the lower section. I'd like to get your input/interpretation.

SJS:

My suggestion for setting it is to keep increasing the voltage until your ECU trips into limp mode at max boost. That means it saw some boost, which it does not expect, and of course that fact is the reason for the ESM to begin with. Once you hit that point, back it down slightly until it never trips under any load. That should be the ideal set point.

I have to respectfully disagree with increasing the voltage to limp mode, then backing off a little. In the research that I did regarding this issue, most people thought that there is a max load fuel map with Acura pre-set fuel cells, and a comensurate timing chart. Not according to the data. It seems "SEEMS!!" that the ECU is making non-linear compensations in open loop at a very rapid frequency. In looking at the timing & short term fuel trims with the Acura scan tool under full boost (don't ask), you can see how quick this cycling occurs, and it is very dynamic (+/- 10 degrees/sec with short term fuel trims bouncing 0.72-0.90 and non-linear). It would be similar to a 2.900 ESM voltage under max load putting out proper injector pulses, but be on the border of too much timing = Detonation. Whereas a 3.050 ESM voltage, would be pulling the correct timing, but the injector pulses would be too little = Detonation. I found this out the nasty empirical way, which led me to believe there was a "sweet point," at which you can have some variances, yet not generate detonation, i.e. [(2.90 + 3.05)/2] = 2.975.

Of course this is all my interpretation of the data. Until Acura/Honda starts to give us some of there programing, it's the best I/we can do.

 

[sjs]

I don't doubt that there is more going on than we know and it's good to hear that you have been able to use a scan tool while investigating. However, I'm a bit perplexed by your example. Higher voltage results from higher manifold absolute pressure (MAP), and for a stock engine that amounts to essentially atmospheric pressure (zero vacuum, zero boost). The ECU is designed to choke if it sees a voltage significantly higher than the one associated with that MAP.

From that I assume when the MAP is at or below that voltage (pressure) it looks up the appropriate cell in the main fuel table (MAP vs RPM), applies any short and/or long term trims and other sensor trims, then pulses the injector accordingly. Likewise the timing from the main timing map, trims, and sensors (including knock sensors).

If that's true, then I would expect to get the greatest injector pulse from the highest tolerated MAP/voltage. I can't think why a higher voltage at a given RPM would ever yield less fuel as you seem to suggest. Timing on the other hand goes the opposite direction, reducing advance as RPM and load increase.

I think that we have always assumed that in both cases with the standard CTSC we want the ECU to act "stock" up to the point were MAP passes into boost, and from that point on we want the injectors pulsing as if we were at that max point in a non-boosted car. Then fuel pressure is increased with boost to provide the additional fuel at a constant (under boost) pulse width, and that makes for a relatively simple and predictable fuel delivery curve.

If on the other hand you clamp the MAP voltage (via the ESM) at a value below what the ECU understands as maximum demand, then you never run your injectors at their intended max duty cycle. That's no big deal if you can simple up the pressure to compensate, but less than max load (MAP) also means you aren't using the upper corner of the timing map and therefore have too much advance, which can surely be disastrous.

However, if in fact the stock timing maps actually increase ignition timing advance towards the cells where max load and RPM intersect, then indeed I can see where a slightly lower voltage would be in order. But I think that would be highly irregular. At least from my experience, if timing is adjusted at all under high loads it is towards less advance, not more. Perhaps the ECU will tolerate a slightly higher voltage than “atmospheric” but the fuel and/or timing maps don’t cover that value and it drops off the map rather than being treated as the max value. That would be lame, but I don’t think it works that way because falling off the map is probably what throws the code. I wish I had a scanner and stock NSX for testing.

I just re-read your post and still don't understand what you think does determine injector pulse and timing if not a fixed base map of MAP vs RPM, plus trims and sensors. What else is there once in open loop?

 

[spdntckt]

Logic to me would seem to be somewhat the opposite...

in a non-boosted car, more throttle = more load = MORE VACUUM (not the other way around).

On a boosted car, more throttle, more load, more boost.

 

[sjs]

Logic to me would seem to be somewhat the opposite...

in a non-boosted car, more throttle = more load = MORE VACUUM (not the other way around).

On a boosted car, more throttle, more load, more boost.

Well, more suction at the front of the intake path (ahead of the throttle plate) yes, but not manifold pressure. When the throttle plate is nearly closed, such as at idle, the engine is trying to pull the air it needs to fill the cylinders through a very small "straw". Imagine sucking hard through a small straw. Your cheeks will be sucked in as your lungs demand more than they can pull through the straw. That's vacuum from the perspective of inside your mouth. Now swap the straw for a large tube and you can inhale fast and easy. A vacuum gauge tapped into your cheek will read zero pressure/vacuum.

The MAP sensor reads Manifold Absolute Pressure downstream of the throttle plate restriction, so at idle you should see 20+ inches of vacuum in a normal street engine. Open the throttle plates and the vacuum drops. By WOT and high revs it approaches zero if your intake is not restrictive. If it doesn't, then you are losing power.

 

[spdntckt]

Ahh good point.. I guess reading prime and trying to work through logic while trying to pay attention in a meeting doesn't pay off - lol... thx - i completely forgot that its in the manifold, not ahead of the throttle plates!

 

[emvanderpol]

I'll have to respond the the previous post tonight:


Len,

I'm wondering if you had time to measure the 9 psi pulley diameter. Also wanting to see if the upgraded injectors read Keihin, and have a blue mark on them. 760-931-9100

Thanks,
Gene

 

[len3.8]

Gene , I will give you a call this evening, sorry I got them out of storage today.
Len

 

[BryanZublin]

Originally posted by emvanderpolBrian: To clarify, take a look at my chart. The hand written numbers represent the ESM voltage. There also is the question of output votage that the ECU sees, which is about 0.2 volts less than the 2.92; kind of odd. You can see it is the 0.02 volts that made the knock go away (2.90 to 2.92) for the new style manifold/inlet. I had to go 0.03 volts over the 2.92 with the older manifold (higher hp & tq, more CFM) to find a balance.

All 5 plots on the dyno chart that you posted look very similar to each other, so it is not obvious to me which one had detonation and which didn't - I'll take your word for it . If there was no other annotation, I would have guessed that the variation was only a result of measurement variability.

How did you verify the setting of the ESM? The dyno notations indicate a fine degree of resolution (about 0.05 V between runs), so I assume you have an accurate method to measure the setting. An offset voltage of 0.2V between ESM input to output is quite large and not what I would expect from this type of device. Here are 3 tests you can perform to verify that your ESM is working properly:

1. At idle, the voltage from the MAP should be relatively low (less than 1V). Measure the MAP voltage into the ESM and the output voltage from the ESM. The difference should be very small (less than 0.01V).

2. The next test requires that the engine is stopped so that the MAP sees atmospheric pressure (ie. no vacuum or boost). Measure the MAP voltage into the ESM and the output voltage from the ESM. The difference should be very small (less than 0.01V). The MAP voltage into the ESM should be around 2.7V.

3. The next test requires that the engine is stopped and the MAP sensor is disconnected. Apply a voltage of 3V or more to the input of the ESM and measure the output voltage of the ESM (with it still connected to the ECU). The output voltage will indicate the "clamp" voltage setting that you desire.

To better illustrate how I think the ESM operates, take a look at Figure 1 in the SmartMAP installation manual at:
http://www.daliracing.com/v666-5/download/smartMAP_install.pdf

Also, take a look at the lower section. I'd like to get your input/interpretation.

I'm not sure what you are referring to here.

 

[len3.8]

I'll have to respond the the previous post tonight:


Len,

I'm wondering if you had time to measure the 9 psi pulley diameter. Also wanting to see if the upgraded injectors read Keihin, and have a blue mark on them. 760-931-9100

Thanks,
Gene

Tried to call you a minute ago Gene. It would seem that we do indeed have the same injectors for the 9lb/high boost kit.
I have the same markings. The tips are green in color.
I sent you a private male on the other measurements.

 

[sjs]

Something I've noticed is that the ESM typically gets mounted in the engine compartment nearer the MAP sensor than the ECU, whereas the SmartMAP instructions make a point of suggesting that it be as close to the ECU as possible (presumably) to reduce the potential of any interference. I wonder if that has caused some problems.

 

[emvanderpol]

Man, you guys sure write alot.

Brian, I was just noting the lower part of my response.

Essentially, the problem that led me to believe that the ESM voltages modified the ECU fuel trim/timing was the detonation I was getting. Between approximately 2.75 and 2.90, I was getting detonation. It gradually decreased as it approached 2.90. Between approximately 2.90 and 2.98, no detonation. Between 2.98 and 3.10 detonation gradually increased. I finally through a code at 3.17. From what I was previously told, once the ESM sees 2.92 volts it goes into max load conditions. Therefore, in theory a higher voltage should do nothing, until of course the ESM thinks there is too much boost in the NA engine and throughs a code. However, the power in the higher ESM voltages 2.98-3.10 went way south. The odd part of this whole series of tests was that the a/f ratios were nearly the same for all test conditions 2.75-3.10, which would make me think that the timing may the the predominate variable.

Further, during some of the dyno tests, and road tests, we measured the ESM's output voltage, and they were pretty consistent at -0.2 volts when compared to the ESM's tab voltage.

Given these data points for a specific engine/blower, there must be a point which would yield max hp/tq, and still not detonate.

What do you guys think.

 

[Onsoku]

Man, you guys sure write alot.

Brian, I was just noting the lower part of my response.

Essentially, the problem that led me to believe that the ESM voltages modified the ECU fuel trim/timing was the detonation I was getting. Between approximately 2.75 and 2.90, I was getting detonation. It gradually decreased as it approached 2.90. Between approximately 2.90 and 2.98, no detonation. Between 2.98 and 3.10 detonation gradually increased. I finally through a code at 3.17. From what I was previously told, once the ESM sees 2.92 volts it goes into max load conditions. Therefore, in theory a higher voltage should do nothing, until of course the ESM thinks there is too much boost in the NA engine and throughs a code. However, the power in the higher ESM voltages 2.98-3.10 went way south. The odd part of this whole series of tests was that the a/f ratios were nearly the same for all test conditions 2.75-3.10, which would make me think that the timing may the the predominate variable.

Further, during some of the dyno tests, and road tests, we measured the ESM's output voltage, and they were pretty consistent at -0.2 volts when compared to the ESM's tab voltage.

Given these data points for a specific engine/blower, there must be a point which would yield max hp/tq, and still not detonate.

What do you guys think.

EmVanderpol,
I just got my CTSC installed It feels strong, but I have no dyno numbers or gauges installed. I think I'm still getting the same CEL that I was getting when I took it in; 52? Also I think I hear detonation, only at night. Daytime is fine. Do you know of a place here in San Diego that can "git er done" right, and quickly?

 

[ken sampson]

Pulley size on mine is 3-5/8" I also have a custom 3" for higher PSI.

 

[emvanderpol]

Give Nick at Applied Motorsports a call. He knows all I know about that damn ESM. His number is 760-439-0959.

Gene

 

[Onsoku]

Give Nick at Applied Motorsports a call. He knows all I know about that damn ESM. His number is 760-439-0959.

Gene

Gene,
Thanks for the call; I'll figure this thing out...:biggrin: