For any discussion not specifically related to your project
By theonewithin
#44701
noisymime wrote: Tue Aug 11, 2020 2:36 am Do you mean PWM pump control?
He doesn't know what PWM means.

He posted this question on Facebook and he got similar response.

Myself and others said it was a bad idea and his response was just that were know it all's.(in regards to he said to just turn the pump off while driving).

Either he deleted the post or admin removed it..
User avatar
By PSIG
#44708
JayCarShopper wrote: Mon Aug 10, 2020 9:49 am Maybe I can build a FP ECU from Speedy products to mainly control/ shut the fuel pump on & off during normal driving.
An FP ECU (fuel pump controller) is not a feature of Speeduino at this time. A FPC may be created stand-alone, or coded into Speeduino, or purchased aftermarket. To verify we are on the same topic, an FPC may:
  • control fuel pump current (amperage), or
  • fuel pump speed, or
  • fuel pump flow, or
  • fuel rail pressure,
or a combination of those. An FPC may employ a fuel pressure sensor, flow sensor, MAP, TPS, current, voltage, RPM input, etc, and a combination or none of the above. They may or may not use a fuel pressure regulator, e.g., they may be a simple power resistor with a throttle position switch to reduce pump power consumption at low-throttle. Is any of this generally what you mean by "FP ECU"?

David
User avatar
By JochemsTR
#44816
David, if there is a way to build or get such controller, I would be very interested. In order to have a return-less system I am struggling to solve this with the UA4C...

Jochem
By moonie223
#44817
Pretty much all fords from like 96+ have a FPDM. It's a PWM multiplying fuel pump driver that can switch like 20A.

You feed it a PWM signal from 50-100Hz and it varies the duty cycle to the pump accordingly.

MS3 allows full closed loop fuel control, no FPR, no return, just a fuel pressure sensor. I doubt it works worth a damn. If the fuel pressure is not spot on at all times your car will run like crap.

You can have a returnless fuel system, you just need to mount the pressure regulator in the tank and have it atmospheric referenced. That's how all NB miata work from the factory in the USA, and they can all be tuned just fine without advanced fuel pump control.

Only reason anyone working with these levels of standalones should be messing with PWMing a pump is to slow down a massive oversized pump as you need on a high boost, high horse e85 application. That would be open loop control, and would still need a fuel pressure sensor to safely set up.
By dazq
#44818
JochemsTR wrote: Wed Aug 19, 2020 7:47 am David, if there is a way to build or get such controller, I would be very interested. In order to have a return-less system I am struggling to solve this with the UA4C...

Jochem
Have you looked at this?
https://www.madhu.com/content/Main/FuelPumpController
User avatar
By PSIG
#44830
JochemsTR wrote: Wed Aug 19, 2020 7:47 am David, if there is a way to build or get such controller, I would be very interested. In order to have a return-less system I am struggling to solve this with the UA4C...

Jochem
@Jochem - there are many uses for FPCs, and your example is common, as are several other needs. In your case, @dazq offered a reasonably-priced minimal unit for fixed pressure control that may do the job for you. More specifics on your use and setup would allow more specific suggestions for design.

For example, if you are using very large injectors (power adder or something), the useful range of the injectors is limited by fixed pressure. Tuning at low power levels becomes difficult as pulse widths are very small for situations such as idle and cruise, causing injectors to fall into non-linear flow range. So, in that example, I would suggest adding a MAP sensor to the FPC (or borrowing the signal from Speeduino) in order to allow maintaining fuel pressure differential across the injectors. In this way it would act more like a MAP-referenced regulator, regulating rail pressure relative to MAP pressure, extending the useful linear-flow range of the injectors.

While that is a common example, other FPC uses are to act as a rising-rate regulator, or to stabilize control at extreme injector capability, and more. While power adders are a common reason, in-fact almost all EFI setups can work with fixed-pressure control, but can benefit from MAP-referenced pressure control. Just food for thought. Now, for a different direction on the same topic …
moonie223 wrote: Wed Aug 19, 2020 8:19 am Pretty much all fords from like 96+ have a FPDM. It's a PWM multiplying fuel pump driver that can switch like 20A.
...
You can have a returnless fuel system, you just need to mount the pressure regulator in the tank and have it atmospheric referenced. That's how all NB miata work from the factory in the USA, and they can all be tuned just fine without advanced fuel pump control.

Only reason anyone working with these levels of standalones should be messing with PWMing a pump is to ...
True that, but there are many reasons for using PWM pump control, as we see here from time-to-time. Likely the one of the most common with users here is vehicles with small or non-existent charging systems. Drag racers, motorcycles, scooters, ATVs, etc. In these cases the fuel system is typically regulated by an inline fuel pressure regulator, bypassing excess pump flow when not needed.

At low speeds (or always with no charging system), the pump is running full-power, consuming far more power than necessary, and sapping battery capacity. Electrical system voltages drop, lights dim, injectors and coils need more dwell, etc. We can avoid this to some extent with a simple FPC routine to lower pump power while still maintaining target fuel pressure and with limited regulator bypass.

While an Arduino could reduce pump power versus reduced MAP for example, Triumph used a simple example of FPC consisting of a throttle switch, relay and a resistor. No "brains". Fuel pressure was handled by a conventional bypassing regulator. It would pass full pump power through the relay unless the throttle was low, where there was little need for fuel, and charging was poor. The throttle switch would energize the relay to divert pump power through the resistor, reducing power to the pump. No, not efficient, but it did the job required. Anyway, that's why I was inquiring to the specific purpose or goals of the OP, so the best control and design strategy could be formulated, whether constructing, coding or buying a FPC. I hope that helps in seeing where I was going with this.

For a pump power control unit your suggestion of the Ford FPDM is excellent IMO, shortcuts the hardware design, and again easily controlled with an Arduino or similar. Code to allow employing one in various modes (standalone or using Speeduino) would give users options to solve issues with a common, inexpensive (used ≈$15+, new $30+), and higher-amp unit. 8-) Others may choose to design hardware from scratch (like me), but the code would be similar and useful.

David

Guts of the Ford FPDM @moonie223 mentioned. It is only the pump driver, needing control signals:
Ford_FPDM_sm.png
Ford_FPDM_sm.png (78.36 KiB) Viewed 6540 times
User avatar
By JochemsTR
#44855
Have you looked at this?
https://www.madhu.com/content/Main/FuelPumpController
Yes, I stummbled upon this site before. This seems most interesting at this point since I hardly need to modifiy anthing, just putting the FPC between Pump and ECU....

I struggled with the fact whether MAP Input is really required or not. My fuelpressure varies from
41 PSI in Push-Downhill -
43,5 PSI in Idle and
45 PSI in WOT

So the debate is really, keep FP in all modes consistent or not....I have 6 - 200cc injectors, but for my car I also have a setup with 2 - 350cc injectors, depending on the inlet manifold I can use.

Jochem
User avatar
By PSIG
#44871
JochemsTR wrote: Thu Aug 20, 2020 7:11 am…So the debate is really, keep FP in all modes consistent or not....
The general question is if there is advantage to varying fuel pressure to MAP. In your case, it is running well and apparently has reasonable control of the injectors across the full range, so adding MAP reference may or may not help. MAP-referenced tends to provide a "flatter" fueling table as pressure differential is maintained, but do you need that?. It would not appear to be required in this project, so at this point, my perspective is to not complicate more than necessary to reach the goals. Your perspective may be different.

As with any system, they carry pros and cons — compromises. Just a few things to keep in-mind when changing to a deadhead system are primarily related to heat. Many or most mechanical FPRs are placed close or on the fuel rail, permitting continuous fuel flow. This helps to maintain a better average fuel temperature at the injectors for consistent tuning results, purge trapped or entrained air, and provides cooling flow through the pump.

A closed deadhead fuel system has none of these advantages (although there are some work-arounds), so consideration for their effects should be on your list for your specific project and effects it may have. Many deadhead systems use a fuel temperature sensor to allow compensation for those effects. Likewise, and depending on the fuel used, may require a higher pressure in order to avoid issues such as vapor lock due to heat soak. Most deadhead systems run 4-bar or higher for reasons such as this.

I would analyze the advantages you seek with a deadhead system, versus the potential disadvantages, and make a choice based on net gains. Is it worth it? That's entirely up to you. If you're not sure, then try it and see what the real-world effects are and if it is still worth it. Get your answers for you, and do your thing.
8-)
David

I uploaded logs and last tune and dyno plots look[…]

you could set 'skip revolutions' to 3, then it wou[…]

Hi, I am trying to assign Signed values to the x-a[…]

blitzbox

I've finally figured out why MAP and Lambda weren'[…]

Still can't find what you're looking for?