Jump to content
SAU Community

Recommended Posts

Hey guys, so after hearing some horror stories of the lift pump in the main tank failing and running the surge tank dry causing the engine to fail due to it leaning out i have decided i want to try and monitor the level in my surge tank or more to the point, have a low level detection to set off a warning lamp.

There's a few ways i have thought to do this.

1. You could try and mount a low level thermistor in to the tank (the same type of unit that is used to detect low fuel in the main tank in most vehicles today). This however will be difficult to seal correctly into a surge tank, they do not come in a packaging that you could mount into an already made surge tank. They also tend to fail.

2. Ultrasonic detection from outside of the tank, There are tiny ultrasonic sensors that you can mount to items that detect if fluid is present or not behind the medium they are mounted too. Unfortunately i do not believe there is one on the market that will detect fuel through aluminium.

3. Mount a float switch to the tank, when the level falls below the tank, the switch closes. Simple. However, if the fuel is sloshing around which it will, the level may get low and activate the switch but the sloshing fuel may cause it to be on and off rapidly and you may never see the warning lamp light up. There is also the issue of trying to mount and correctly seal this type of switch on an already made surge tank. I dare say fuel, especially E85 will eat most of these switches.

4. Put a DC current transformer into the mix. These will monitor the amount of current that a device is drawing. For example, the lift pump may draw 5 amps of current under normal operating conditions. If the circuit for the pump fails or a fuse blows the current will drop to 0 amps. The DC current transformer will pick this up and can be set to switch a relay which in turn lights up a lamp. The other way around is under fault/short/overload/pump seizure the pump will draw well over the normal 5 amps. The current transformer will also pick this up and can throw the relay in the same way. The issue with this setup, it doesn't account for a mechanical failure of a split line from the lift pump to the surge tank. They are expensive and are proving very difficult to find in an exact form that will suit this purpose. These are usually used in automation where a PLC will monitor a 4-20mA range.

5. This is the way that i am thinking will be my best bet. Fit a hobb switch that is set to (1-2PSI) inline with the feed line from the lift pump to the surge tank. When the lift pump is working fine the hobb switch is in a healthy state, if the pump fails, the pressure will drop to 0 causing the hobb switch to activate therefore turning on a warning lamp. The only issue i'm having with this type of setup is what pressure does the line between the lift pump and the surge tank operate at? Obviously it is bugger all, but surely there must be some sort of pressure there? Whether it be 1, 2, 3 or 4 PSI or whatever it is. They must have some pressure. I'm using an aeromotive pump as the lift pump. Its a high volume high pressure pump, so one would thinking, pumping stacks of fuel through an 8mm fuel line will create some sort of pressure which is what i want in this case.

This idea is cheap, simple and will work quite well if all goes to plan. Unfortunately this method does not detect the level in the surge tank, however it will let me know very early that the lift pump has failed and that my engine is going to run lean quite soon.

6. Using a flow meter to monitor the amount of fuel being pumped into the surge tank. These devices are extremely expensive and i highly doubt there is an exact item on the market for my specific purpose.

These are my current ideas, im sure there are a few other simple ideas out there, which is why i'm asking you guys. Please let me know your thoughts. Sorry for the long post, but this is a real life scenario and i'm looking for a solution.

Daniel.

  • Replies 48
  • Created
  • Last Reply

Top Posters In This Topic

Set it so that when the surge tank is low and dangerous, to cut the engine. Nto sure how but it seems to be a good solution. Hope you find a way that works!

What ECU do you have?

Edited by The Mafia

Set it so that when the surge tank is low and dangerous, to cut the engine. Nto sure how but it seems to be a good solution. Hope you find a way that works!

What ECU do you have?

I dont particularly want it to cut the engine. Just warn me early that it is running low.

Just a NIStune for the time being. So no 'extra' engine controls or monitoring.

Why not set a low pressure switch to kill the engine if pressure drops to nothing in the feed line to the surge tank?

You might have to use a low pressure reg between the main tank and the surge tank and another return

Get a propery designed surge tank and lift pump, then you shouldn't have any worries about running the surge low.

All you going to do is create another "worry system" in the car.

Where does it end. Warnings for surge tank, warning for fuel rail, warning for radiator header tank, warning for oil level, warning for everything that has a level.

Get a propery designed surge tank and lift pump, then you shouldn't have any worries about running the surge low.

All you going to do is create another "worry system" in the car.

Where does it end. Warnings for surge tank, warning for fuel rail, warning for radiator header tank, warning for oil level, warning for everything that has a level.

Little cameras inside the bores to check for det?

Or in the sump to see if there is oil surge?

Lol

Why not set a low pressure switch to kill the engine if pressure drops to nothing in the feed line to the surge tank?

You might have to use a low pressure reg between the main tank and the surge tank and another return

I can't mount an FPR to the surge tank as my surge tank is not running "pressurized". The return from the engine is at the highest point of the tank, with the return to the main tank being the second largest tank. This is to create a small air pocket so the large main pump is not working overtime to try and pump fuel into the surge tank and then over flow into the main tank. This is a large reason why main pumps run so hot, the main pump is working so hard to do its job. This is off topic though haha.

Magnets on a float inside, with a reed switch outside?

This will be almost impossible to do as the surge tank has already been made. Most setups like this require you to have access on both sides of the plate that you are mounting the reed switch setup too.

Dual lift pumps lol

How big is the surge tank?

Dual lift pumps are an idea, but not practical in an R31 cradle/fuel tank.

Surge tank is approx 4 Litres, but 500-700mL is useless as the inlet for the main pump is 40mm off the bottom of the surge tank.

Run a frp on the overflow from the surge tank and maintain say 10psi in the surge tank. Then just run a pressure switch on the pressure side. Just make shure you have a decent lift pump...

Cheers

Justin

Get a propery designed surge tank and lift pump, then you shouldn't have any worries about running the surge low.

This doesn't help in anyway if the lift pump fails...

All you going to do is create another "worry system" in the car.

Where does it end. Warnings for surge tank, warning for fuel rail, warning for radiator header tank, warning for oil level, warning for everything that has a level.

This may be so. But its not as bad knowing how many sub woofers to stop at..

Run a frp on the overflow from the surge tank and maintain say 10psi in the surge tank. Then just run a pressure switch on the pressure side. Just make shure you have a decent lift pump...

Cheers

Justin

This is a good idea, i do like it. Its very similar to one of my ideas above, but it involves fully pressurizing the surge tank which puts the main pump under even more strain and heats the fuel further. Which is not what i want to do.

Ah... beat me to it.

As above.

Constant recirc of fuel, pump speed, cavitation and fuel volume has had more of an impact on fuel temp than pressure in my experience (not just talking about hi-po applications).

Have a look at closed fuel coolers (non tube/fin types), pump speed/pwm controllers, surge tank heatsink and return line size too.

Cheers

Justin

I can't mount an FPR to the surge tank as my surge tank is not running "pressurized". The return from the engine is at the highest point of the tank, with the return to the main tank being the second largest tank. This is to create a small air pocket so the large main pump is not working overtime to try and pump fuel into the surge tank and then over flow into the main tank. This is a large reason why main pumps run so hot, the main pump is working so hard to do its job. This is off topic though haha.

I would have thought it would run harder and get hotter under pressure?

Ah... beat me to it.

As above.

Constant recirc of fuel, pump speed, cavitation and fuel volume has had more of an impact on fuel temp than pressure in my experience (not just talking about hi-po applications).

Have a look at closed fuel coolers (non tube/fin types), pump speed/pwm controllers, surge tank heatsink and return line size too.

Cheers

Justin

Fuel pump speed will be varied dependent on TPS posistion. I am installed a big f**k off fuel cooler :)

Feed to and from fuel rail are 1/2", feed and return from the main tank to the surge tank are 5/16"

Just stick a good quality lift pump in there and get rid of the stock one will prob cost the same as going through this trouble of adding a sensor and it should be reliable.

I've already got a good quality new lift pump. I'm wanting to do this after speaking to a few people that have had relatively new lift pumps fail on them.

I would have thought it would run harder and get hotter under pressure?

Yes, thats what i'm trying to say. The pump works harder and gets hotter the higher the pressure.

Screw monitoring the surge tank side why don't you monitor the injector rail - that's what's important. Use a pressure differential sensor that is set to the pressure you want in your rail and set the bounds to say +/- 2psi. Hook 1 side up to fuel rail pressure the other side to manifold pressure.

Use this http://www.mamcoswitches.com/differential-pressure-switches/mam0071.html

Screw monitoring the surge tank side why don't you monitor the injector rail - that's what's important. Use a pressure differential sensor that is set to the pressure you want in your rail and set the bounds to say +/- 2psi. Hook 1 side up to fuel rail pressure the other side to manifold pressure.

Use this http://www.mamcoswit...es/mam0071.html

Yes i'm going to do this as well using the Engine Monitor 2 kit that Paul is offering on here. It monitors fuel pressure vs boost pressure, coolant temp, oil pressure vs rpm (i love this one). Plus a few other things.

I suppose i'm being a little over the top wanting to put in a surge tank level monitor, butif it can be done a low cost. Why not do it?

I was wondering what you were going to say actually Rob. Would you have any idea what pressure the line would operate at between the lift pump and the surge tank?

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



  • Similar Content

  • Latest Posts

    • Any update on this one? did you manage to get it fixed?    i'm having the same issue with my r34 and i believe its to do with the smart entry (keyless) control module but cant be sure without forking out to get a replacement  
    • So this being my first contribution to the SAU forums, I'd like to present and show how I had to solve probably one of the most annoying fixes on any car I've owned: replacing a speedometer (or "speedo") sensor on my newly acquired Series 1 Stagea 260RS Autech Version. I'm simply documenting how I went about to fix this issue, and as I understand it is relatively rare to happen to this generation of cars, it is a gigantic PITA so I hope this helps serve as reference to anyone else who may encounter this issue. NOTE: Although I say this is meant for the 260RS, because the gearbox/drivetrain is shared with the R33 GTR with the 5-speed manual, the application should be exactly the same. Background So after driving my new-to-me Stagea for about 1500km, one night while driving home the speedometer and odometer suddenly stopped working. No clunking noise, no indication something was broken, the speedometer would just stop reading anything and the odometer stopped going up. This is a huge worry for me, because my car is relatively low mileage (only 45k km when purchased) so although I plan to own the car for a long time, a mismatched odometer reading would be hugely detrimental to resale should the day come to sell the car. Thankfully this only occurred a mile or two from home so it wasn't extremely significant. Also, the OCD part of me would be extremely irked if the numbers that showed on my dash doesn't match the actual ageing of the car. Diagnosing I had been in communication with the well renown GTR shop in the USA, U.P.garage up near University Point in Washington state. After some back and forth they said it could be one of two things: 1) The speedometer sensor that goes into the transfer case is broken 2) The actual cluster has a component that went kaput. They said this is common in older Nissan gauge clusters and that would indicate a rebuild is necessary. As I tried to figure out if it was problem #1, I resolved problem #2 by sending my cluster over to Relentless Motorsports in Dallas, TX, whom is local to me and does cluster and ECU rebuilds. He is a one man operation who meticulously replaces every chip, resistor, capacitor, and electronic component on the PCB's on a wide variety of classic and modern cars. His specialty is Lexus and Toyota, but he came highly recommended by Erik of U.P.garage since he does the rebuilds for them on GTR clusters.  For those that don't know, on R32 and R33 GTR gearboxes, the speedometer sensor is mounted in the transfer case and is purely an analog mini "generator" (opposite of an alternator essentially). Based on the speed the sensor spins it generates an AC sine wave voltage up to 5V, and sends that via two wires up to the cluster which then interprets it via the speedometer dial. The signal does NOT go to the ECU first, the wiring goes to the cluster first then the ECU after (or so I'm told).  Problems/Roadblocks I first removed the part from the car on the underside of the transfer case (drain your transfer case fluid/ATF first, guess who found out that the hard way?), and noted the transfer case fluid was EXTREMELY black, most likely never changed on my car. When attempting to turn the gears it felt extremely gritty, as if something was binding the shaft from rotating properly. I got absolutely no voltage reading out of the sensor no matter how fast I turned the shaft. After having to reflow the solder on my AFM sensors based on another SAU guide here, I attempted to disassemble the silicone seal on the back of the sensor to see what happened inside the sensor; turns out, it basically disintegrated itself. Wonderful. Not only had the electrical components destroyed themselves, the magnetic portion on what I thought was on the shaft also chipped and was broken. Solution So solution: find a spare part right? Wrong. Nissan has long discontinued the proper sensor part number 32702-21U19, and it is no longer obtainable either through Nissan NSA or Nissan Japan. I was SOL without proper speed or mileage readings unless I figured out a way to replace this sensor. After tons of Googling and searching on SAU, I found that there IS however a sensor that looks almost exactly like the R33/260RS one: a sensor meant for the R33/R34 GTT and GTS-T with the 5 speed manual. The part number was 25010-21U00, and the body, plug, and shaft all looked exactly the same. The gear was different at the end, but knowing the sensor's gear is held on with a circlip, I figured I could just order the part and swap the gears. Cue me ordering a new part from JustJap down in Kirrawee, NSW, then waiting almost 3 weeks for shipping and customs clearing. The part finally arrives and what did I find? The freaking shaft lengths don't match. $&%* I discussed with Erik how to proceed, and figuring that I basically destroyed the sensor trying to get the shaft out of the damaged sensor from my car. we deemed it too dangerous to try and attempt to swap shafts to the correct length. I had to find a local CNC machinist to help me cut and notch down the shaft. After tons of frantic calling on a Friday afternoon, I managed to get hold of someone and he said he'd be able to do it over half a week. I sent him photos and had him take measurements to match not only the correct length and notch fitment, but also a groove to machine out to hold the retentive circlip. And the end result? *chef's kiss* Perfect. Since I didn't have pliers with me when I picked up the items, I tested the old gear and circlip on. Perfect fit. After that it was simply swapping out the plug bracket to the new sensor, mount it on the transfer case, refill with ATF/Nissan Matic Fluid D, then test out function. Thankfully with the rebuilt cluster and the new sensor, both the speedometer and odometer and now working properly!   And there you have it. About 5-6 weeks of headaches wrapped up in a 15 minute photo essay. As I was told it is rare for sensors of this generation to die so dramatically, but you never know what could go wrong with a 25+ year old car. I HOPE that no one else has to go through this problem like I did, so with my take on a solution I hope it helps others who may encounter this issue in the future. For the TL;DR: 1) Sensor breaks. 2) Find a replacement GTT/GTS-T sensor. 3) Find a CNC machinist to have you cut it down to proper specs. 4) Reinstall then pray to the JDM gods.   Hope this guide/story helps anyone else encountering this problem!
    • So this being my first contribution to the SAU forums, I'd like to present and show how I had to solve probably one of the most annoying fixes on any car I've owned: replacing a speedometer (or "speedo") sensor on my newly acquired Series 1 Stagea 260RS Autech Version. I'm simply documenting how I went about to fix this issue, and as I understand it is relatively rare to happen to this generation of cars, it is a gigantic PITA so I hope this helps serve as reference to anyone else who may encounter this issue. NOTE: Although I say this is meant for the 260RS, because the gearbox/drivetrain is shared with the R33 GTR with the 5-speed manual, the application should be exactly the same. Background So after driving my new-to-me Stagea for about 1500km, one night while driving home the speedometer and odometer suddenly stopped working. No clunking noise, no indication something was broken, the speedometer would just stop reading anything and the odometer stopped going up. This is a huge worry for me, because my car is relatively low mileage (only 45k km when purchased) so although I plan to own the car for a long time, a mismatched odometer reading would be hugely detrimental to resale should the day come to sell the car. Thankfully this only occurred a mile or two from home so it wasn't extremely significant. Also, the OCD part of me would be extremely irked if the numbers that showed on my dash doesn't match the actual ageing of the car. Diagnosing I had been in communication with the well renown GTR shop in the USA, U.P.garage up near University Point in Washington state. After some back and forth they said it could be one of two things: 1) The speedometer sensor that goes into the transfer case is broken 2) The actual cluster has a component that went kaput. They said this is common in older Nissan gauge clusters and that would indicate a rebuild is necessary. As I tried to figure out if it was problem #1, I resolved problem #2 by sending my cluster over to Relentless Motorsports in Dallas, TX, whom is local to me and does cluster and ECU rebuilds. He is a one man operation who meticulously replaces every chip, resistor, capacitor, and electronic component on the PCB's on a wide variety of classic and modern cars. His specialty is Lexus and Toyota, but he came highly recommended by Erik of U.P.garage since he does the rebuilds for them on GTR clusters.  For those that don't know, on R32 and R33 GTR gearboxes, the speedometer sensor is mounted in the transfer case and is purely an analog mini "generator" (opposite of an alternator essentially). Based on the speed the sensor spins it generates an AC sine wave voltage up to 5V, and sends that via two wires up to the cluster which then interprets it via the speedometer dial. The signal does NOT go to the ECU first, the wiring goes to the cluster first then the ECU after (or so I'm told).  Problems/Roadblocks I first removed the part from the car on the underside of the transfer case (drain your transfer case fluid/ATF first, guess who found out that the hard way?), and noted the transfer case fluid was EXTREMELY black, most likely never changed on my car. When attempting to turn the gears it felt extremely gritty, as if shttps://imgur.com/6TQCG3xomething was binding the shaft from rotating properly. After having to reflow the solder on my AFM sensors based on another SAU guide here, I attempted to disassemble the silicone seal on the back of the sensor to see what happened inside the sensor; turns out, it basically disintegrated itself. Wonderful. Not only had the electrical components destroyed themselves, the magnetic portion on what I thought was on the shaft also chipped and was broken. Solution So solution: find a spare part right? Wrong. Nissan has long discontinued the proper sensor part number 32702-21U19, and it is no longer obtainable either through Nissan NSA or Nissan Japan. I was SOL without proper speed or mileage readings unless I figured out a way to replace this sensor. After tons of Googling and searching on SAU, I found that there IS however a sensor that looks almost exactly like the R33/260RS one: a sensor meant for the R33/R34 GTT and GTS-T with the 5 speed manual. The part number was 25010-21U00, and the body, plug, and shaft all looked exactly the same. The gear was different at the end, but knowing the sensor's gear is held on with a circlip, I figured I could just order the part and swap the gears. Cue me ordering a new part from JustJap down in Kirrawee, NSW, then waiting almost 3 weeks for shipping and customs clearing. The part finally arrives and what did I find? The freaking shaft lengths don't match. $&%* I discussed with Erik how to proceed, and figuring that I basically destroyed the sensor trying to get the shaft out of the damaged sensor from my car. we deemed it too dangerous to try and attempt to swap shafts to the correct length. I had to find a local CNC machinist to help me cut and notch down the shaft. After tons of frantic calling on a Friday afternoon, I managed to get hold of someone and he said he'd be able to do it over half a week. I sent him photos and had him take measurements to match not only the correct length and notch fitment, but also a groove to machine out to hold the retentive circlip. And the end result? *chef's kiss* Perfect. Since I didn't have pliers with me when I picked up the items, I tested the old gear and circlip on. Perfect fit. After that it was simply swapping out the plug bracket to the new sensor, mount it on the transfer case, refill with ATF/Nissan Matic Fluid D, then test out function. Thankfully with the rebuilt cluster and the new sensor, both the speedometer and odometer and now working properly!   And there you have it. About 5-6 weeks of headaches wrapped up in a 15 minute photo essay. As I was told it is rare for sensors of this generation to die so dramatically, but you never know what could go wrong with a 25+ year old car. I HOPE that no one else has to go through this problem like I did, so with my take on a solution I hope it helps others who may encounter this issue in the future. For the TL;DR: 1) Sensor breaks. 2) Find a replacement GTT/GTS-T sensor. 3) Find a CNC machinist to have you cut it down to proper specs. 4) Reinstall then pray to the JDM gods.   Hope this guide/story helps anyone else encountering this problem!
    • perhaps i should have mentioned, I plugged the unit in before i handed over to the electronics repair shop to see what damaged had been caused and the unit worked (ac controls, rear demister etc) bar the lights behind the lcd. i would assume that the diode was only to control lighting and didnt harm anything else i got the unit back from the electronics repair shop and all is well (to a point). The lights are back on and ac controls are working. im still paranoid as i beleive the repairer just put in any zener diode he could find and admitted asking chatgpt if its compatible   i do however have another issue... sometimes when i turn the ignition on, the climate control unit now goes through a diagnostics procedure which normally occurs when you disconnect and reconnect but this may be due to the below   to top everything off, and feel free to shoot me as im just about to do it myself anyway, while i was checking the newly repaired board by plugging in the climate control unit bare without the housing, i believe i may have shorted it on the headunit surround. Climate control unit still works but now the keyless entry doesnt work along with the dome light not turning on when you open the door. to add to this tricky situation, when you start the car and remove the key ( i have a turbo timer so car remains on) the keyless entry works. the dome light also works when you switch to the on position. fuses were checked and all ok ive deduced that the short somehow has messed with the smart entry control module as that is what controls the keyless entry and dome light on door opening   you guys wouldnt happen to have any experience with that topic lmao... im only laughing as its all i can do right now my self diagnosed adhd always gets me in a situation as i have no patience and want to get everything done in shortest amount of time as possible often ignoring crucial steps such as disconnecting battery when stuffing around with electronics or even placing a simple rag over the metallic headunit surround when placing a live pcb board on top of it   FML
    • Bit of a pity we don't have good images of the back/front of the PCB ~ that said, I found a YT vid of a teardown to replace dicky clock switches, and got enough of a glimpse to realize this PCB is the front-end to a connected to what I'll call PCBA, and as such this is all digital on this PCB..ergo, battery voltage probably doesn't make an appearance here ; that is, I'd expect them to do something on PCBA wrt power conditioning for the adjustment/display/switch PCB.... ....given what's transpired..ie; some permutation of 12vdc on a 5vdc with or without correct polarity...would explain why the zener said "no" and exploded. The transistor Q5 (M33) is likely to be a digital switching transistor...that is, package has builtin bias resistors to ensure it saturates as soon as base threshold voltage is reached (minimal rise/fall time)....and wrt the question 'what else could've fried?' ....well, I know there's an MCU on this board (display, I/O at a guess), and you hope they isolated it from this scenario...I got my crayons out, it looks a bit like this...   ...not a lot to see, or rather, everything you'd like to see disappears down a via to the other side...base drive for the transistor comes from somewhere else, what this transistor is switching is somewhere else...but the zener circuit is exclusive to all this ~ it's providing a set voltage (current limited by the 1K3 resistor R19)...and disappears somewhere else down the via I marked V out ; if the errant voltage 'jumped' the diode in the millisecond before it exploded, whatever that V out via feeds may have seen a spike... ....I'll just imagine that Q5 was switched off at the time, thus no damage should've been done....but whatever that zener feeds has to be checked... HTH
×
×
  • Create New...