I've done a fair bit of testing recently with cheap crap China sensors whether on oil or water injection.

Compared to better quality sensors on my car, the readings from them do seem a lot more erratic and require more filtering. But for non critical items ( and most only cost me £10ea ) that's of no real concern. Using them mostly for information so no harm can come if they do fail.

 

I do a running average when datalogging, just like most datalogging software does. Nice smooth graph with the noise filtered out.
Someone I know has managed to fail a couple of them by supplying them with 12v instead of 5v because he didn't want to tap into a 5v rev and ground in his factory harness.
There are a few unused 5v and ground pins on the factory PCM connector that he could have easily tapped into. They did manage to last a few months on 12v.

 

I used a Fuel On Demand from Professional Products. Works awesome. Programmable for desired pressure and also for return style or returnless. One problem... They're not available anymore lol

Quick video of mine running on the laptop:

 

During a pull on a Mustang Dyno:

 

Do Fore or Vaporworx or offer something similar ? Although perhaps without the on screen info ?

 

I talked to professional products last November and they said they were developing a new PWM without the glitz and glamour of the computer software. Will be a low cost option for the LS crowd. I'm not sure when it's expected to be done though.

 

what ever came of this?

 

Is there any more news about this, or other DIY PWM fuel pump control options?

I'd love to be able to shove a modern GM fuel pump module into a tank it wasn't designed for, and get access to a schematic, parts list, and the basic coding to assemble all of that into my own control box. The problem is that any company who has developed a control unit to the point of being dependable will also want to finalize and profit from the design. But I'm looking for something that I can build into a can-bus wiring harness of my own.

 

I'm not aware of a commercial product, but if you want to get into small-scale manufacturing this seems like a reasonably straightforward project.

There's a guy on Facebook who advertises automotive electronic design services, I'm sure he could design the electronics, and probably do the software too:
EDIT: Are Facebook links not allowed? Search for AutomotivePCBdesign if the link doesn't work.

Then get a contract manufacturer in China to build 10 or 20 of them. Could probably get the built for $10-$20 each and sell them for $50. Might take a couple iterations to work the bugs out.

For the sake of flexibility I'd suggest supporting a few kinds of input:
* CANbus for newer cars
* 5v variable-frequency like the MAF sensors on P01/P59 PCMs
* and 0-5v analog for MAF sensors on Subaru and other ECUs
* PWM input so that you can tap a fuel injector signal

The software would just need a way to choose which type of input to use, a few parameters to choose which CAN message and bits to use, and a table to map each type of signal to the PMW duty cycle.

I've never done business with the Facebook guy linked above but he seems legit. I've been thinking about working with him to do a 52mm touchscreen gauge for random CANbus parameters, but I have too much going on already.

 

Looks like Big Bad Brother is stalking you....

 

how would you support can bus?
look at tps? rpm?

 

ive started to design a bunch of various controllers like this and the problem is it needs way more time/effort/development in config than actually doing what its supposed to do. controlling a pump is easy. making it configurable for consumers isnt worth the effort imo