Fuel Rail High Pressure Sensor Upgrade

[06Speed6]

JJ: Your right, it has the same thread but the sealing is different. The connector is different but it uses the same min and max voltage and the same three wire setup. I think the one I listed is a racing version.

The 0265005303 sensor is available here:

All European

Part Number: 0025400917
Description: Pressure Switch for ESP Unit
Brand: Bosch
Alternate Part Number 0265005303
List Price: $210.00
Sale Price: $122.51

[SSinstaller]

Originally Posted by bmorrisj ha i answered a question before ssinstaller... can i be in the msf smartypants club? keep that brain juice flowing

LOL. For the next twenty-four hours, you are an unofficial honorary member of the smartypants club..

[bnoon]

I like reading the hell out of these threads. Subbed.

[SilverDemon]

In SUMark's case, I am almost certain that it was the dash hawk reporting that high of pressure...but maybe someone should invite him in here and let him tell us... hahaha

[NJSPEED3]

Lord my brain is now swollen i hate wiring, and voltage crap.

Someone dumb this down 5 levels.

I thought the sb was tricking the sensor like Dustin was mentioning are you guys stating that it will be pegged off no matter what once the check valve is in place?

[djuosnteisn]

Potentially, yes.

You can only trick a functional sensor... but once it goes flat, your trickery goes flat as well.

[SSinstaller]

Originally Posted by NJSPEED3 Lord my brain is now swollen i hate wiring, and voltage crap. Someone dumb this down 5 levels. I thought the sb was tricking the sensor like Dustin was mentioning are you guys stating that it will be pegged off no matter what once the check valve is in place?

The SB tricks the ECM, but it's still limited by the range of the stock sensor...

For example lets say the sensor voltage is 4.5V or ~2080psi.. If you tried to add 20% more pressure with the SB the voltage the ECM gets will drop to 3.6... It will try and increase the actual FP via the spill valve until it's gets back to the 4.5V it wants to see...

The problem is in order for the computer to ever see the 4.5v it wants(which is already the actual sensor voltage pre-SB), the rail sensor output would need to hit 5.4v which can never happen... it will peg at 4.8v, so the computer will just continue to request more pressure until something fails...

[Lex]

Has anyone sustained injection pressures above 2100 or 2200? I am very curious how the injectors would handle those pressures.

Knowing how automotive control software is written, there is a table or scaling factor in the ECU that correlates sensor output to actual pressure so it's certainly doable from that standpoint.

[Erich]

I have had sustained pressures over 2074 and both WOT and cruise. As my ECU pegs at 2074 I can't say specifically what they were but my 1800 map would hit and hold 2074 at times so the 1900 and 2000 maps are probably at least over 2100.

[djuosnteisn]

Can you post some logs? For all you know, you could be hitting like 2300.

[Erich]

Valve cracks at 2150. So I don't think I am sustaining anything over that. Logs are posted here.

[SSinstaller]

Originally Posted by Erich I have had sustained pressures over 2074 and both WOT and cruise. As my ECU pegs at 2074 I can't say specifically what they were but my 1800 map would hit and hold 2074 at times so the 1900 and 2000 maps are probably at least over 2100.

I'd be very surprised to see a log of pressure holding above the RV crack pressure..

I spike as high as 2200, but it just oscillates back and forth between 2100+ and the crack pressure..

If your logs peg at 2074 I bet your doing the same, but the flat line at 2074 makes it look like it's holding that pressure..

[Erich]

Originally Posted by SSinstaller I'd be very surprised to see a log of pressure holding above the RV crack pressure.. I spike as high as 2200, but it just oscillates back and forth between 2100+ and the crack pressure.. If your logs peg at 2074 I bet your doing the same, but the flat line at 2074 makes it look like it's holding that pressure..

This is what I suspect as well. oscillating between 2074+ and 2150+

[djuosnteisn]

I've heard of valves being overwhelmed in the past. Not saying yours was, but that it has happened. At one point, back in the day, there was even talk about actually porting the relief valve, hahaha.


Okay, jus checked your logs... and welp... i'd venture to say that 1), your quite possibly overwhelming the RV, and 2), you've successfully stopped the inj's from opening from fuel pressure. Too bad we don't know what pressure that is.

[Erich]

Been running the 1800 map since running the tests. My plan is to slowly adjust the the FP tables until I am getting 2050-2070 where I want maximum pressure then stick with that until it is possible to read higher then that on the ECU.

I guess it is possible that I was overwhelming the RV on the 2100 map. The ECU only ever thinks it is seeing 2074, so it never reaches its target and keeps on sending more fuel. Wonder if I would get similar results setting the FP tables to something like 2070 or 2065. Even on the lower pressure maps I get extremely rich high in the RPM range (can smell raw fuel in the exhaust.)

[06Speed6]

The injector opening is stopped by the net pressure differential between the rail and cylinder, it has nothing to do with rail pressure alone.

Rail pressures above ~1800psi probably only need to be targeted at boost pressures above 18psi, and above ~2000psi for say 22psi, and above ~2200psi for say 26psi. Ide have to refer back to the dynamic compression ratio thread to get projected cyl pressure increases for specific boost pressures. It would also probably be more accurate to reference off of maf/g for a more static cfm rating since its corrected for atmospheric conditions.

EDIT: I believe the inj driver can shut down the injectors if rail pressure exceeds ~2500psi and missfires can begin around 2300psi but I have no idea what the conditions were. Missfires can be logged, just not with a DH or AP or really anything else except the dealer diag tool. I know Versatune is workin on adding it in.

[FreeFlyFreak]

I dont think it is mentioned specifically in this thread, and it could cause some confusion for others reading this, it confused me for a second........but I am presuming from the conversation that Erich is currently running an aftermarket relief valve? Correct?

EDIT: Oopps mean to post this in the Car does not like 2100 desired at all thread

[djuosnteisn]

Is that true? Erich, are you running an aftermarket valve?

[Erich]

Correct.

[djuosnteisn]

Originally Posted by 06Speed6 EDIT: I believe the inj driver can shut down the injectors if rail pressure exceeds ~2500psi ...

But the ecu would never know when that occurs, cause the sensor is pegged by 2047.


+1 on your remarks about cylinder compression. Very good info to keep in mind.

Originally Posted by Erich Correct.

Um.... someone's got to say it...




Are you sure it's opening? And are you sure it relieves pressure at least as well as the stock valve?

[06Speed6]

The sensor can be overdriven to around 5.5v before its damaged, so it goes to around 2610psi.

[djuosnteisn]

That's from the supply voltage IIRC. When i tested my sensor it would go buggy above 6 volts.

[FADE2GRY]

Just in case this info is useful to the discussion...

Our vehicles use PID 23(hex) to report fuel rail pressure via OBD-II. This PID uses 2 bytes and thus has a range of 0 to 65535 with a scaling of 10kPa per bit. This means that the PID can report 0 to 655350 kPa or 0 to 95050 psi.

So... the limitation to 2063-ish psi isn't related to the AP or DashHawk or any OBD-II compliant device. The ECU is just not reporting anything higher.

Cheers!

[06Speed6]

Originally Posted by FADE2GRY Just in case this info is useful to the discussion... Our vehicles use PID 23(hex) to report fuel rail pressure via OBD-II. This PID uses 2 bytes and thus has a range of 0 to 65535 with a scaling of 10kPa per bit. This means that the PID can report 0 to 655350 kPa or 0 to 95050 psi. So... the limitation to 2063-ish psi isn't related to the AP or DashHawk or any OBD-II compliant device. The ECU is just not reporting anything higher. Cheers!

You sir need to post more!

Thanks for the infos

Dustin: The feed voltage is 15v with a max of 16v so it should be basically alternator current.

[djuosnteisn]

Originally Posted by 06Speed6 Dustin: The feed voltage is 15v with a max of 16v so it should be basically alternator current.

I'm gonna go out on a limb and guess that "feed voltage" is the term for the pull up voltage for the signal line. There's conditioning circuitry in these sensor, and the last stage is likely an op amp with an open collector, so it requires a pull up.

In the application schematic the datasheet shows, they're pulling the signal line upto the supply rail (5V). But whether or not that's what the mazda engineers did is unknown. Someone would need to disconnect the sensor connector and measure pin 2 with the car on (engine off) and see if it reads 5V or 12V.

But regardless, i don't think your going to be able to squeak any additional voltage out of it.

[SilverDemon]

We need to try a test in a real world application. I have an extra fuel pressure sensor that I can give up for the cause. I could machine a block with one side with the sensor threads cut in one end, add a port on the top for a hydraulic pressure gauge, and at the other end add a port for a porta power hand held hydraulic power unit. Plug the sensor into the stock wiring harness, turn the car on and data log fuel pressure on the dash hawk, and compare it to the analog hydraulic gauge. Just a thought....

[Fobio]

Here's SUMark's thread:

HELP! Car won't hold idle....

[06Speed6]

Off of the pdf:

Nominal range is .2v to 4.8v
140bar or 2030psi is ~5v
Overdriven range is 5v +-.25v
Max overdriven read is 180bar or 2610psi


But actually we are both wrong lol, I did the math.

5.00v measured to the hundreth = 140bar...

140/500 = .28bar per .01v

.28*550 = 154bar... so boo on 5.5v being 180bar lol

180/.28 = ~642 or 6.42v

Basically if 5v is 140bar and the overdriven range is 5v +-.25v or 5.25v then

.28*525 = 147bar or 2132psi

[SSinstaller]

Originally Posted by 06Speed6 Off of the pdf: Nominal range is .2v to 4.8v 140bar or 2030psi is ~5v Overdriven range is 5v +-.25v Max overdriven read is 180bar or 2610psi But actually we are both wrong lol, I did the math. 5.00v measured to the hundreth = 140bar... 140/500 = .28bar per .01v .28*550 = 154bar... so boo on 5.5v being 180bar lol 180/.28 = ~642 or 6.42v Basically if 5v is 140bar and the overdriven range is 5v +-.25v or 5.25v then .28*525 = 147bar or 2132psi

The data says otherwise...



I'm still saying 5v= ~2350psi, and it looks like 4.5v would line up nicely with the ~2074psi peak that all are seeing with the AP...

[SSinstaller]

A different point on the same log shows 4.5v = 2089psi according to the SB.

[06Speed6]

That doesnt match the data points in the graphs Bosch provided for the sensor.

If 5.00v = 2350psi or 162bar then
.01v = 4.7psi or .32bar
Then if we bump it to the over driven limit of 180bar then:
180bar/.32 = 562 or 5.62v
or 140bar
140bar/.32 = 437 or 4.37v

Which is closer to the graph that Bosch provides than the points on the data sheet that they say are the limits of the sensor.

On the Graph they have 140bar being 4.5v
140bar/450 = .31bar per .01v
So 162bar should be
162/.31 = 522 or 5.22v
and 180bar should be
180/.31 = 580 or 5.8bar

[06Speed6]

If 4.50v is 2089psi or 144bar, then:
144bar/450 = .32bar per .01v
then
140bar/.32 = 437 or 4.37v
180bar/.32 = 562 or 5.62v

That matches exactly to .01v the previous graph you posted.

If someone else has some sb rail pressure and rail v logs please post them so we can see if there is a variance between sensors. There may be a correction factor of some kind that is related to baro or temps or something else.

[djuosnteisn]

Marcos, what do you think the over drive limit is? I'm confused at what your saying basically. You can't really "over drive it" afaik. There's a supply voltage for the analog circuitry, and if you up the rail voltage, the output voltage will scale with it, even for a fixed pressure... kinda like a resistive voltage divider. And the driven voltage is simply the pull up voltage for the open collector out put of the signal line. You can pull it up to 16V, and all that will happen is the signal line (op amp) will simply sink the excess current.

[SSinstaller]

Ok just did some math and it looks like the scale on the SB is off..

if .5v = 0 bar and 4.5v = 140 bar then (pressure in bar)= 35(Vout)-17.5

then 4.8v = 151 bar 2219psi, and 5v = 2315psi.

I guess we're all wrong... LOL

I'd bet the actual sensor output will be usable up to about 4.9v, but we'll just have to see what the sensor pegs at... but it's definitely not 2074 or 2350... LOL

Al this really tells me is that the max RV crack pressure I'd want using the OEM rail sensor is still going to be about 2300psi...

[06Speed6]

The data sheet claims that it can be "over driven" to 180bar or 5.5v but I did notice that it never claims that 180bar is 5.5v.

The graph goes to 4.5v at which point the graph flat lines. Graph itself has a 4.8v limit after which is a "alarm" area, but it doesnt say the sensor actually quits functioning.

Originally Posted by SSinstaller Ok just did some math and it looks like the scale on the SB is off.. if .5v = 0 bar and 4.5v = 140 bar then (pressure in bar)= 35(Vout)-17.5 then 4.8v = 151 bar 2219psi, and 5v = 2315psi. I guess we're all wrong... LOL I'd bet the actual sensor output will be usable up to about 4.9v, but we'll just have to see what the sensor pegs at... but it's definitely not 2074 or 2350... LOL Al this really tells me is that the max RV crack pressure I'd want using the OEM rail sensor is still going to be about 2300psi...

0bar is 0v... you can look at a DH with the engine off and it still can read ~12-14psi of rail pressure.

[djuosnteisn]

Maybe it's uncalibrated at that point...

I like Scott's idea for testing the sensor.

[06Speed6]

+1 get on it scott! =)

[SSinstaller]

Originally Posted by 06Speed6 0bar is 0v... you can look at a DH with the engine off and it still can read ~12-14psi of rail pressure.

Just because the engine is off does not mean the ECM is powered down.. The 5v power output for the engine sensors is powered anytime the ECM is awake..


This log was posted in another thread by someone also using the SB to increase the FP



Based off my logs the sb FP equation is, (FP psi)= 540(Vin)-341.... So assuming his SB software is using the same calibration his 2274psi means the OEM sensor is capable of outputting at least 4.84v/2233psi

[djuosnteisn]

I'm just gonna throw this out there, and maybe it's rubbish, but if the SB ground is different that the ECU's ground or even the sensor's ground, you can get different voltages (much like the crank pos sensor issue chris).


But it certainly seems like the sensor reads quite a bit over 4.5V.. which is a good thing. I'd just like to verify it.

[forcedinduktion]

anyone else feel stupid reading this ?