Battling failing fuel pumps, I take a chance on using pulse width modulation to reduce fuel pump speed to help reduce heat and extend pump life expectancy.
I would also suggest running one pump full time and the other to kick in via pwm or or switch based on MAP or MAF load signal from the ecu. This is a more reliable setup and less prone to failure. More importantly, This will help keep your pressure at a more consistent range.
Interestingly this seems to be standard on OEM fpcms. Mopar for example. I need to bench test the 14+ Ram setup. I think it defaults to 80% or something, and when getting a proper signal can output like 15a or something like that
Most of those SSR's don't allow True PWM, it's more of a opto-coupled messy square wave duty cycle. It would be nice if they made a SSR with a I2C or SPI serial interface for that. That said, it should still work with a intermittent duty cycle though it might be a lot of stress on the pump's motor.
No problem, I liked your video. You might even be able to use a MOSFET transistor directly with only a flyback diode in between drain and source to quench the inductive spikes. If the SSR continues to run no problem stick with it, but if it ever burns out try a fairly simple MOSFET switch with the diode. That will allow True PWM right to the pump at high frequency. Look forward to your videos.@@uniqueautomationelectronic5696
Are you sure that you aren’t losing pressure inside the tank? I’ve seen those rubber lines(even the submersible rated) rubber hoses get soft after a short while or shrink enough to lose pressure from the hose clamps in the tank… alot of the time, it only will show up under load. I always use the plastic corrugated stuff with the squeeze type clamps for inside the tanks…. Just something to consider for you possible pressure loss issues.
feed size will also extend pump life i went from -6 too -8 and my fuel temp decreased step it down at the rail there are too many bends too leave the line -6
I can't hear it when I'm driving the car. It's been quite a while since I did this project. I don't remember it being any louder, but the pitch of the pump motor was different when running slower.
I recently got a pd16, I’m thinking of trying this but I hear people say brushed walbros aren’t meant to be pwm. How is this holding up for you 2 years later?
@@Jamxknife I drive the car very seldom. However it is still working fine. I contacted TI, and they did not voice concerns when I was inquiring about this approach.
I have searching different solid state relays and all off them accept duty only up to 90% not 100%. this means with solid state relay i lose 10% off pump capacity. or am i wrong?
U do know that in the z32 300zx tt as well as the aristo with the gte vvti among other cars I guess, there is a fuel computer which runs the pump at 9volts under normal and then 12 volts when triggered by say boost pressure and possibly other inputs. I am not too confident o on off on off. Arcing unecessarily
@russpeed301 Yes I am aware. That was the inspiration for this setup after burning out a handful of pumps in less than a year each. Unfortunately these higher flow pumps draw quite a few more amps than the stock pump and powering them with factory wiring would be risky while also reducing effective voltage at the pump itself. So I had to bypass it.
Hey thanks for your video, I am trying to do something similar with an electric water pump for my Fueltech. For the Haltech is the output you are using for the fuel pump send 0v or 12v. In my situation to get PWM the ECU needs to send 12v. Seems like the Haltech is sending ground in your situation since you didn't mention any grounds in your SSR setup
That is correct. Haltech is wired to ground the circuit. +12 is sourced elsewhere for the relay. For my application it is sourced from the stock factory fuel pump harness. By way of the Halltech, the circuit is completed as it is grounded based on the programmed frequency (HZ) on the table.
@@533L5 thanks for the reply, I may have found a diagram from Fueltech, they have a similar setup for a NOS solenoid. Using the 12v output I am looking to use. Hopefully it works 🤞🏿
A fair question, but I can't recall. I called TI automotive to get their input when I set it up. I want to say in the 120hz range but I cannot swear to that as I don't have the car here at home. Regardless, I would recommend you contact the pump manufacturer.
@@533L5 Thanks for the fast reply. I'm using 3 450's, with a SSR on each pump. Seems to work well , but I'm trying to dial in the frequency for them atm.
I would also suggest running one pump full time and the other to kick in via pwm or or switch based on MAP or MAF load signal from the ecu. This is a more reliable setup and less prone to failure. More importantly, This will help keep your pressure at a more consistent range.
Awesome Video Brother - BTW according to Walbro ....The controller requires a squarewave frequency (battery voltage to ground) at 100Hz.
Interestingly this seems to be standard on OEM fpcms. Mopar for example. I need to bench test the 14+ Ram setup. I think it defaults to 80% or something, and when getting a proper signal can output like 15a or something like that
Most of those SSR's don't allow True PWM, it's more of a opto-coupled messy square wave duty cycle. It would be nice if they made a SSR with a I2C or SPI serial interface for that. That said, it should still work with a intermittent duty cycle though it might be a lot of stress on the pump's motor.
Thanks
No problem, I liked your video. You might even be able to use a MOSFET transistor directly with only a flyback diode in between drain and source to quench the inductive spikes. If the SSR continues to run no problem stick with it, but if it ever burns out try a fairly simple MOSFET switch with the diode. That will allow True PWM right to the pump at high frequency. Look forward to your videos.@@uniqueautomationelectronic5696
very interesting video
Are you sure that you aren’t losing pressure inside the tank? I’ve seen those rubber lines(even the submersible rated) rubber hoses get soft after a short while or shrink enough to lose pressure from the hose clamps in the tank… alot of the time, it only will show up under load. I always use the plastic corrugated stuff with the squeeze type clamps for inside the tanks…. Just something to consider for you possible pressure loss issues.
feed size will also extend pump life
i went from -6 too -8 and my fuel temp decreased
step it down at the rail
there are too many bends too leave the line -6
Does the fuel pump sound louder at less than 100%? Does it also vibrate more as it is pulsing?
I can't hear it when I'm driving the car. It's been quite a while since I did this project. I don't remember it being any louder, but the pitch of the pump motor was different when running slower.
I recently got a pd16, I’m thinking of trying this but I hear people say brushed walbros aren’t meant to be pwm. How is this holding up for you 2 years later?
@@Jamxknife I drive the car very seldom. However it is still working fine. I contacted TI, and they did not voice concerns when I was inquiring about this approach.
@@533L5 Ah I see good to know, I’ll definitely try it then, I just gotta run 1 wire anyway. Also fyi the 535 only draws 20 amps, the 525 draws 25
I have searching different solid state relays and all off them accept duty only up to 90% not 100%. this means with solid state relay i lose 10% off pump capacity. or am i wrong?
U do know that in the z32 300zx tt as well as the aristo with the gte vvti among other cars I guess, there is a fuel computer which runs the pump at 9volts under normal and then 12 volts when triggered by say boost pressure and possibly other inputs. I am not too confident o on off on off. Arcing unecessarily
How bad is the arcing?
@russpeed301 Yes I am aware. That was the inspiration for this setup after burning out a handful of pumps in less than a year each. Unfortunately these higher flow pumps draw quite a few more amps than the stock pump and powering them with factory wiring would be risky while also reducing effective voltage at the pump itself. So I had to bypass it.
@@yujyuj2957 I'm not sure I follow. The solution has been in place now for a year or so. Probably 2,000 miles on it with no known illl effects.
Hey thanks for your video, I am trying to do something similar with an electric water pump for my Fueltech.
For the Haltech is the output you are using for the fuel pump send 0v or 12v.
In my situation to get PWM the ECU needs to send 12v. Seems like the Haltech is sending ground in your situation since you didn't mention any grounds in your SSR setup
That is correct. Haltech is wired to ground the circuit. +12 is sourced elsewhere for the relay. For my application it is sourced from the stock factory fuel pump harness. By way of the Halltech, the circuit is completed as it is grounded based on the programmed frequency (HZ) on the table.
@@533L5 thanks for the reply, I may have found a diagram from Fueltech, they have a similar setup for a NOS solenoid. Using the 12v output I am looking to use.
Hopefully it works 🤞🏿
I know the EKP modules in BMW cars operate the fuel pumps up to 2khz.
Is the 2khz max on the EKP3 or EKP2?
Interesting
What frequency are you using ?
A fair question, but I can't recall. I called TI automotive to get their input when I set it up. I want to say in the 120hz range but I cannot swear to that as I don't have the car here at home. Regardless, I would recommend you contact the pump manufacturer.
@@533L5
Thanks for the fast reply. I'm using 3 450's, with a SSR on each pump. Seems to work well , but I'm trying to dial in the frequency for them atm.
450 is the pump to use