Engineer/Geek Types, questions about PWM controlled Brushless and Voltage Boosters

[Angrey]

The second spring and ball/seat(component 169) is a bypass for throttle lift to avoid destroying their sensor, per their patent. Is this the swing arm you referenced? The only moving part is the main diaphragm otherwise (once setup).

You are correct. The spring doesn't actuate the sensorship. I stand corrected.

However, as I laid out, when you ACTUALLY READ. The value that the transducer is reporting to the controller has NOTHING TO DO WITH THE PRIMARY DIAPHRAM SPRING. IT IS MEASURING " Pressure transducer 155 measures the differential pressure (pressure drop) between return reservoir 167 and chamber 168 and outputs a signal to ECM 121 based upon that measurement. "

Both of those are clearly AFTER the primary regulator and certainly on the RETURN SIDE of the entire assembly (not the rail side).

Furthermore, it comes with a "special" A/N adapter (to connect to the return line fitting) that has been specially sized to give the desired medium/mid backpressure (they sell the regulator in an 8 AN version and a 6 AN version). That insert into 168 is crucial for developing a mid level steady state for typical ranges where the regulator will operate with enough flow in the return line to provide some buffering, but not enough to suffer the pumps being overly tasked.

The controller measures pressure differential and it has NOTHING to do with the regulator diaphram and spring. It's a transducer that generates a value that varies according to the pressure delta which is the basis for control. It has NOTHING to do directly with the regulator diaphram. The diaphram is free to actuate as necessary to maintain the rail side pressure, the only thing changing the transducer is how much return flow/back pressure is present.

Sponsored

 

[Angrey]

Look, at the end of the day. Whatever. I'm tired of defending it. The pressure on the rail side stays constant. It simply ramps the pumps up and down based upon how much flow is coming out of the return, which is an end result of how much the rails are consuming. When working as designed, (simplified for a naturally aspirated setup) as the motor drinks, the return drops, the controller steps up pump duty, more flow, the return line returns back to steady state, the entire time the set pressure on the rail side remains constant. End of story.

 

[Grimreaper]

You are correct. The spring doesn't actuate the sensorship. I stand corrected.

However, as I laid out, when you ACTUALLY READ. The value that the transducer is reporting to the controller has NOTHING TO DO WITH THE PRIMARY DIAPHRAM SPRING. IT IS MEASURING " Pressure transducer 155 measures the differential pressure (pressure drop) between return reservoir 167 and chamber 168 and outputs a signal to ECM 121 based upon that measurement. "

Both of those are clearly AFTER the primary regulator and certainly on the RETURN SIDE of the entire assembly (not the rail side).

Furthermore, it comes with a "special" A/N adapter (to connect to the return line fitting) that has been specially sized to give the desired medium/mid backpressure (they sell the regulator in an 8 AN version and a 6 AN version). That insert into 168 is crucial for developing a mid level steady state for typical ranges where the regulator will operate with enough flow in the return line to provide some buffering, but not enough to suffer the pumps being overly tasked.

The controller measures pressure differential and it has NOTHING to do with the regulator diaphram and spring. It's a transducer that generates a value that varies according to the pressure delta which is the basis for control. It has NOTHING to do directly with the regulator diaphram. The diaphram is free to actuate as necessary to maintain the rail side pressure, the only thing changing the transducer is how much return flow/back pressure is present.

I added to my previous post.. but putting it here.

The reservoir chamber allows the differential pressure to exist and fluctuate based on how the main diaphragm is behaving. The main diaphragm behaves as it does based entirely off of pressures. Feedback is now available to show how much more or how much less the pumps are over/under producing and going into bypass essentially. The reservoir chamber is dependent on the diaphragm/intake chamber. Not independent. If it was independent there would be no usefulness of the change in pressures wanted to control the pumps.

The extra reserve added in by changing the restrictor is clever. Basically your pumps are providing just enough overage to allow for big throttle hits and then this buys time for the pumps to respond and ramp up.

 

[Grimreaper]

And glad it works for you and it honestly should. It's a tried and true design with a new spin from technology to reduce some of the fuel heating and pump draws the standard return regs cause.

 

[Cory S]

Man.
So much time spent on something that’s already been designed/invented that works. There’s engineering and over engineering/thinking. Holy Shit.
Do the pumps work? Does fuel pressure stay above minimum safe threshold? Do the logs look good. Fucking send it. JFC.

 

[Angrey]

I added to my previous post.. but putting it here.

The reservoir chamber allows the differential pressure to exist and fluctuate based on how the main diaphragm is behaving. The main diaphragm behaves as it does based entirely off of pressures. Feedback is now available to show how much more or how much less the pumps are over/under producing and going into bypass essentially. The reservoir chamber is dependent on the diaphragm/intake chamber. Not independent. If it was independent there would be no usefulness of the change in pressures wanted to control the pumps.

The extra reserve added in by changing the restrictor is clever. Basically your pumps are providing just enough overage to allow for big throttle hits and then this buys time for the pumps to respond and ramp up.

The whole debate is about whether or not the rail side sees a change in pressure to illicit a change in pump duty. It does not.

The regulator spring/diaphragm travels it's range of motion in order to maintain constant pressure with varied rail consumption. When the return flow varies, so too does the pumps. So indirectly, yes, it's based upon the regulator diaphragm. But it's more directly analogous with the return flow/backpressure. Which is why that's the operable place to adjust the base duty.

The central assertion that kicked off 2 pages of brain damage was that the system has to wait fora drop in pressure (at the rails) before it reacts, which is simply untrue. It reacts to the drop in pressure on the return line. IT works.

I've proven that a pair of 440 pumps (without booster) can support up to 1100 rwhp without a stitch of control from the PCM (outside of the trigger. I'm using the FPDM logic to trigger pump power, so they follow the momentary prime cycle and wait for ignition before continued use.)

Interestingly, although the regulator controller doesn't need a full battery source to operate, when I first wired in, I tried to use the FPDM wire to power the regulator and it did NOT like that at all. The voltage source needs to be constant (for the regulator). Outside of that, it was the only hiccup I encountered.

 

[Cory S]

The whole debate is about whether or not the rail side sees a change in pressure to illicit a change in pump duty. It does not.

The regulator spring/diaphragm travels it's range of motion in order to maintain constant pressure with varied rail consumption. When the return flow varies, so too does the pumps. So indirectly, yes, it's based upon the regulator diaphragm. But it's more directly analogous with the return flow/backpressure. Which is why that's the operable place to adjust the base duty.

The central assertion that kicked off 2 pages of brain damage was that the system has to wait fora drop in pressure (at the rails) before it reacts, which is simply untrue. It reacts to the drop in pressure on the return line. IT works.

I've proven that a pair of 440 pumps (without booster) can support up to 1100 rwhp without a stitch of control from the PCM (outside of the trigger. I'm using the FPDM logic to trigger pump power, so they follow the momentary prime cycle and wait for ignition before continued use.)

Interestingly, although the regulator controller doesn't need a full battery source to operate, when I first wired in, I tried to use the FPDM wire to power the regulator and it did NOT like that at all. The voltage source needs to be constant (for the regulator). Outside of that, it was the only hiccup I encountered.

Without typing 16 books of text, I’m just loving the fact my little twin Ti274’s are supporting me on Ethanol like an absolute Boss with trims/lambda in perfection with a -10 feed and no return line to run. .

 

[engineermike]

The whole debate is about whether or not the rail side sees a change in pressure to illicit a change in pump duty. It does not.

You are correct that a pressure change in the rail side is not what directly changes the pump duty cycle. As I’ve said before, forget about the electronics for a second and look at what has to happen first (regulator cross-section helps visualize this). The mechanical regulator has to change position in order to send more or less flow to the return, which then changes pressure in the return side, which then changes the duty cycle. But the first component is a mechanical regulator, which is a feedback control device that can only react to pressure changes after they happen. The rail pressure drops then the regulator closes, reduces return flow, pump speeds up. It’s a cause-effect relationship.

It really is a pretty cool device, and adds a layer of functionality and control to the traditional bypass-regualated fuel system. But it is in no way a predictive device unless it’s tied into the pcm logic or accelerator pedal.

That said, if you’re on e85 and don’t have to worry about knock, then there is little chance this lack of predictive control will really hurt anything. And if on pump gas and tuned properly with borderline tip-in modifiers and lambda offsets, you can minimize the effect as well.

@Cory S don’t those pumps lack check valves and Venturi taps? How are they installed and controlled with the return-less system?

 

[Cory S]

@Cory S don’t those pumps lack check valves and Venturi taps? How are they installed and controlled with the return-less system?

The Ti274’s do have built in check valves. They are setup exactly like OEM operation with Venturis. They are installed in a Radium fuel pump basket/hat combo with the regulator mounted in the hat itself. They are not controlled by PWM. The OEM FPDM is used as a relay trigger and for prime only. It’s a brilliant design as not having to run a return line that just heats up fuel over time. The fuel return is done in the hat itself.

 

[Angrey]

You are correct that a pressure change in the rail side is not what directly changes the pump duty cycle. As I’ve said before, forget about the electronics for a second and look at what has to happen first (regulator cross-section helps visualize this). The mechanical regulator has to change position in order to send more or less flow to the return, which then changes pressure in the return side, which then changes the duty cycle. But the first component is a mechanical regulator, which is a feedback control device that can only react to pressure changes after they happen. The rail pressure drops then the regulator closes, reduces return flow, pump speeds up. It’s a cause-effect relationship.

It really is a pretty cool device, and adds a layer of functionality and control to the traditional bypass-regualated fuel system. But it is in no way a predictive device unless it’s tied into the pcm logic or accelerator pedal.

That said, if you’re on e85 and don’t have to worry about knock, then there is little chance this lack of predictive control will really hurt anything. And if on pump gas and tuned properly with borderline tip-in modifiers and lambda offsets, you can minimize the effect as well.

@Cory S don’t those pumps lack check valves and Venturi taps? How are they installed and controlled with the return-less system?

I'm done arguing with you. The rail side pressure does not change.

Just like without a controller, if the pumps are operating at full duty, the regulator is maintaining the set pressure by bleeding off excess flow. If the motor then consumes more fuel, the diaphragm on the regulator adjusts to reduce the bleed off and maintain the desired set pressure. Not sure why you're either being stubborn or having a difficulty understanding this.

Or are you of the incorrect opinion that a mechanical regulator can not maintain a set pressure value?

I've also further informed you that I've both observed the pressure values myself (on the dial gauge) and we've observed pressure data in the logs.

It's not that difficult of a concept. The controller simply regulates the pump duty based upon how much resultant return line flow/pressure exists, all the while the rail side pressure remains constant, even though the consumption is varying. NOT a very difficult concept. It seems you're so set on your esoteric tuning solution that you either refuse to acknowledge a different approach or you're truly not understanding how it physically works (which is unlikely because you're obviously very intelligent).

The system actually achieves what is most desired. Controlling the pumps, not based off schedules or approximations or predictions, but based off increases and decrease in engine consumption (which propagate themselves as increases and decreases to return flow under constant pressure conditions).

I've tried exhaustively. At this point, I'll just give up. Either you understand it or you don't.

 

[bankyf]

The Ti274’s do have built in check valves. They are setup exactly like OEM operation with Venturis. They are installed in a Radium fuel pump basket/hat combo with the regulator mounted in the hat itself. They are not controlled by PWM. The OEM FPDM is used as a relay trigger and for prime only. It’s a brilliant design as not having to run a return line that just heats up fuel over time. The fuel return is done in the hat itself.

I assume you are tuning this yourself? What HP do you estimate that you are making?

 

[Cory S]

I assume you are tuning this yourself? What HP do you estimate that you are making?

This setup isn’t self tuned. It’s a file that RS setup last February. Uses the same calibration as any other standard return system. Nothing special.
Depending on DA, it puts down between 874-908whp.

 

[markmurfie]

feedback: correcting error after it happens
feedforward: predicting error and accounting for it before it happens


This pump controller is neither. Its operating on a buffer that can absorb the error. As long as the engines fuel demand doesn't change so fast that it over runs that buffer and the control systems reaction time it should work fine. The size of this buffer, the "return pressure" of the return line, would need to be tuned according to the application its being used in. Too little buffer you risk over run, too much buffer you lose the benefits of reduced pump output. Lower fuel temperatures, avoid cavitation, and less strain on the vehicles charging system. Set the pump control in the ECU to full at all times and tune like a normal return style system, use a cautiously high return pressure for this external pump control to target. Make sure the rail pressure is held constant on both rapid accel and decel transitions.

It seems like you have already figured all this out and just want to know how a voltage booster would affect the DW pump controller. According to you seeing 12.5v-13.6V now and their 18V rating I would say your logic on how it will effect PWM is on the right track and it will just give the pumps more head room under high demand. The PWM is reducing the voltage you currently see to what the pump requires to meet the target pressure in the return line. Then allows it to increase/decrease with fuel demand changes from the engine. It will have to reduce more voltage if it has 18V available instead, but should be able too with automatically using a smaller PWM.

I/ We don't know where in the 20Hz(0%) to 200kHz(100%) range you are currently with your base pressure at 12.5- 13.6V, logging and figuring that out would be useful. Multimeter for resistance. If going to 18V you hit the 20Hz limit(high ohms /open), get a smaller diameter or longer return line. With the PWM the oversize return line wouldn't be as important.

Everything compatible with this regulator is 100z-5000hz range.
The DW controller is 0% when open and 100% when its grounded, with a greater frequency range.
This should work fine as 100hz should be "open"(high ohms), and 5000hz "ground"(low/sub ohm) from the regulator.
Logging this would be resistance(ohms) based, like a temperature sensor kind of.

An additional output from the Electronic Regulator can be used to track the fuel system operating condition. This 0- 5 Volt analog output allows the user to use a data acquisition system, Engine ECM, general data logger, or simply a volt meter, to track how hard the pump is signaled to operate. The voltage is typically at about 1.5 to 2.0 Volts (depending on Pump Model used and other factors such as baseline fuel pressure) while engine is operating at idle condition. Typical full capacity has an output of approximately 4.5 Volts.

This is probably more what you want to hook into a data logger like HPtuners or SCT. green+ and black- wire. Typical 0.5v(0%)-4.5V(100%) sensor, probably linear 1V=25%dc. (voltage*25-12.5= DC%)Tune the return restriction based on this to achieve 25-35% low load range.
Red is ignition source, white is to the pump controller. Logging it is the only way to see if the controller/ buffer is keeping up with the dynamics of the engine load.

 

[markmurfie]

Sorry I had to edit my original post after doing more research. What you are doing should work, and you should log your regulator to know what its doing.
This is not as simple as your typical return style fuel system or even just hi/ low pump triggering.
The end results should be fairly interesting.

I don't see remote tuners reccomending this any time soon.

You probably could connect the ECU FDM control wire into this and control it via the ECU.

 

[engineermike]

... If the motor then consumes more fuel, the diaphragm on the regulator adjusts to reduce the bleed off and maintain the desired set pressure....Or are you of the incorrect opinion that a mechanical regulator can not maintain a set pressure value?

What I am saying is that the diaphragm on the regulator adjusts as a result of something, and that something is a change in pressure. Without a change in pressure, there would be no movement of the diaphragm.

....you either refuse to acknowledge a different approach...

Not true, I've said several times that it's better than a traditional return-style system and definitely has its advantages.

Controlling the pumps, not based off schedules or approximations or predictions, but based off increases and decrease in engine consumption

What's best is doing both, using predictions to get it close to the future demands, then using a PID feedback loop to fine-tune from there.

Sponsored