FUEL PUMP RFI ( ham radio related)

i remember it like it was yesterday, by cracky!

 

Think about what that would do to any can bus or other method of digital signal transport….. tiling or pixilation on your tv is what you can see. Imagine your TPS going apeshit

 

yeah. the guy i talked to last night got wedged because in the 70's he did his thesis on semiconductor control of commutation of brushless motors (pretty apropos )
and being that early vested in the technology he expected things to progress and brushed motors go away. he could not believe someone was trying to use PWM with a brushed motor.
so many analogs in my life, too. somehow things do progress though. (for u whippersnappers: just old folks ruminating here )

 

@WileECoyote

 

[3rd gen] Toyota Tacoma fuel pump is a three-phase brushless motor.

The older 12V fuel pumps are bushed motors, btw. The fuel pump sits in fuel, which provides cooling. If the pump stops working a tap with a rubber mallet would help 'bump' the position of the pump's rotor, so it would start spinning again.

The modern fuel system (no return line) went with a fuel pump ECU. Speed is determined by the pressure sensor. The Engine ICU sends out a PWM signal to the Fuel Pump ECU. The Fuel Pump ECU is powered by that ignition relay, FPC - Fuel Pump Contor is a PWM signal from the Engine ECU, and creates the three-phase motor speed control. The FSM will point out the three wires to the fuel pump should all be around 1 Ohm resistance. Spec calls for 0.05 to 3 Ohm per phase.

You could sneak a wire probe into the wiring between the Fuel Pump ECU and Fuel Pump to see what that frequency is.

 

Mike - noticed "Beaverton, OR". So how many Tek instruments do you have?

I loved the old surplus sales. 185th? Sigh, finally passed along my 7603 scope, my logic analyzers, and still have my "free to a good home" 576 Curve Tracer. Moved to Texas in 1993 when the market tanked. Told my bother I'd move back in two years, but I'm still here!

The 576 curve tracer was a godsend in troubleshooting odd issues (alternator diode that would short above 80C!) Test fine on the bench, long hot drive meant a dead battery. wtf? Jeep Wagoneer.

These days with 10Gig signals, I let my work buy those $25,000 scopes! I have my trusty Saleae Logic Analyzer in my travel bag.

My classic time: Prototype boards worked fine. Working Software. The customer did a production board, and it didn't work. Next thing I'm on a flight to Asia. Arrive late, told the Engineers time to get to work let's take a look. Signals didn't look right, severely attenuated. What, SMD components were new, but is that a 102 or 105 resistor. It's a 105, let's change it! "Out technicians have left for the day". I smiled and asked, "where is the soldering station, I'll do it" "Oh, we thought they were sending an engineer, not a technician." I swapped out the wrong part, new one in. Board up and running, self-test passed. The three engineers all were silent, staring at their feet. "OK, we found one issue but let's spend the next week reviewing the rest of the design before we say anything". Three beaming smiles.

Had a great time, spent the mornings "in class" going over subsystems, rest of the day spent "team-building" (i.e. sightseeing).

I've forgotten more than I ever remember, hate finding myself making mistakes but am grateful to have gone from the days of analog fuel injection to modern digital systems.

 

I’m assuming the “G, L and B” represent Green, Light Blue and Black for color code of the wiring ?

for shits and giggles , I’m gonna measure the resistance at the plug exiting the rail module and see how far off the resistance is at that spot. In theory, it should be close if the harness is ok.
I may also attempt to check resistance between the shield and the frame. That I should have no resistance on if the shield of that wire is doing it’s job.

 

Getting a bit extreme here, but one possibility to better shield the fuel pump would be to swap out the factory plastic fuel tank for a long range all metal tank. These folks make a 33 gallon tank that is a direct swap for the factory tank. Very expensive, but you get some additional range out of the deal.

https://longrangeamerica.com/shop-tanks/

 

yikes, yeah that is salty. My Mobile HF operations aren’t anywhere close to being worth that expense.

 

No dice on the shield continuity. Without seeing internal schematic, hard to say if that shield is even stretched to ground ( though is should be for proper shielding)

 

Could you tap the shielding on the wire and ground it? Assuming, of course, the shielding is even connected to the pump and not just a useless sleeve around the cable run.

 

Not trying not to stap the insulation at all. Last thing I want is a puncture in the insulation. My luck, No matter how well i think it’s sealed, it would more than likely suck up moisture

 

You might be able to do it at the plug end, if it has its own pin it wouldn't be to hard.

 

You probably already came to this conclusion, but putting ferrites on each individual motor lead is perfectly fine with PWM control. I design industrial equipment for a living, and we put ferrites on our PWM controlled BLDC motors all the time for to meet EMI requirements (most common for CE) with no demonstrable downsides. One ferrite per power leg, keeping each leg separate. The ferrites go either at the controller, or motor depending on where the noise is coming from. Now if the noise is radiating out from the motor itself, this won't help of course.



I am also a bit interested in that shield. It's hard to tell how it's connected from that picture. If there is no shield continuity (as you indicated) then I doubt that shield is doing anything other than looking nice on a schematic... Of course there may be a cap somewhere in the shield circuit, which could make a continuity test fail. Like you indicated, internal schematics would be best to determine how the shield is wired. I would like to hope that Toyota knows what they are doing, but weirder things have happened.

 

I looked at my paperwork, it’s still under the extended warranty. I’m not sure if that warranty covers the fuel pump or not, I have to read the fine print.

I’m going to try that route first , showing how it’s spewing that RFI And see if that is enough to get it replaced under the warranty. Next step would be the FCC unintentional radiator clause, then see.
I’m not holding my breath on any of that but if I can get the service managers ear to fully hear me out, I might have some hope.

I’ve been looking at different suppliers and it seems as if everyone is out of stock on pumps to attempt an aftermarket solution…..

I went and added yet another bond strap to the antenna today , cleaned up all my current bond straps and even took a set of jumper cables from the antenna shield/ground directly to the frame to see if for some stupid reason it fixes it and it changed nothing.
If it were a band bond/ ground , my antenna wouldn’t tune for crap and it’s never tuned above 1.5-1 between 7mhz and 50 MHz and never gets any RF back into the stereo system, factory amp, dims lights or causes idle surges as can happen with RF ingress.


I knew you could ferrite larger VFD systems , as I had to do it to my neighbors fish pond pump, but was not sure if the added inductance would change anything in the FP signal And cause any RPM reductions and cause a flow issue.

 

I hope you are able to get this worked out with the service manager. Finding a failed component is certainly the best outcome!!!

Actually, while I design industrial equipment, it is very rare that I use VFD's. My typical application is using PWM control of a BLDC motor for precise servo control, and positioning for multi-axis coordinated motion in precision CNC equipment. While the current in these systems is still higher than I would expect in a fuel pump, we are extremely sensitive to motor RPM, and overall control to achieve extremely precise accuracy. While we are holding a part in a fixed position, our typical current draw per motor in a multi axis system, is about 0.3A.

Adding ferrites to the smaller system should be just fine. In a BLDC motor, the PWM is used as a more efficient method of regulating the current in each of the three phases on the motor. Ideally the current would be smoothed out before it gets to the motor, but since a motor is basically an inductor, it smooths out the pulses internal to the motor (yes, I am oversimplifying here). In fact, for highest performance, linear amplifiers are still used to control BLDC motors, to eliminate the loss of motor control when the current in a phase crosses the 0 current point.

 

lol. Funny you should ask and OT but I'm game. I have 4 tek scopes. 465M and 465B that i picked up and fixed. Gremlins came back on 2 of them
and they are completely torn down waiting for me to recap the LV power supplies and deal with, i think, HV restorer circuit problem in one
and something wrong in timebase of another.
I took them apart in spring but I just can't bring myself to finish them up while weather is good. I hunch over soldering iron in the foul rainy weather.
I am from silicon valley, moved to OR, but am late to the party playing with analog tek scopes.
Scope repair really a hobby (though I would prefer they would stop breaking). Not what I did in real life.
They are only 100MHz but very useful around the house. Diagnosed my furnace controller last winter, though probably could have done it
with a DMM. No fun in that though.

ps. 185th: I think you are referring to Wacky Willy's?. Long gone but Surplus Gizmos is cool on Cornelius Pass north of 26.
Back to watching great thread...

 

I’m hoping the RFI alone is enough to show an issue, but we shall see.

a lot of the newer technology in motor controls is well above my
Pay grade, especially when you deal with it infrequently like I do. It’s easy to be left in the dust if you do not keep up with it.

 

Get a Kenwood KLF2 in-line noise filter.

We use them in OffRoad racing for the same type of noise issues

 

It’s not in the power line.