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

[engineermike]

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

This is definitely how I would do it.

Sponsored

 

[Angrey]

I don't know how close to the "floor" of the controller I am in terms of low level operation. I'm assuming at some point, there's a limit (low end) to how slow the controller will try to run the pumps.

Basically does running 18V in shorter pulses equivocate to the same pump output as running 13V in slightly longer pulses?

The DW controller has a very wide frequency range, it was one of the things I was worried about prior to setting this up, whether the Fuelab regulator and the DW controllers would "talk" to each other properly.

Yes, most "I'm too cool for school" tuners won't be interested for a number of reasons. They're not familiar and they're lazy and just want you to configure the car the way they're comfortable and know (and ultimately makes less work for them) and we can't ignore the backscratching that goes on with companies that have sponsorships and kickbacks for parts (i.e. if you run what WE want you to run, my shop gets free shit from said vendor/supplier/manufacturer).

When I started this venture, I spoke with all the major muscle movers in the industry. Fuelab, Fore, Radium, Deatschwerks, Aeromotive, etc. I found it crazy that there was apprehension to attempting an in tank brushless solution. Fore told me it couldn't be done. Aeromotive told me they looked at it and weren't interested because it didn't resolve siphon issues/complaints (from the corvette crowd). Fuelab basically said "we did that once and no one bought it."

Now here we are 3 years later. Aeromotive has come up with their in tank package. Fuelab has updated and jumped on board. Injector Dynamics is now offering their controller to mere mortals for less than a lottery jackpot.

I gotta say, Deatschwerks and Radium were great to work with and on the front end of all this.

I still have work to do. The autometer gauge doesn't work as advertised and after numerous calls with the deepest bowels of their tech department, we still can't get it to work. If I could get that to work, I could have even a fully automated transfer of fuel to the active/driver's side. As it is now, it's on a manual switch. Which isn't terrible, but not my preference.

The automotive industry is so late to the party. Brushless motors have permeated virtually everything else. Only cheap, lazy shit is brush style motors anymore. They burn up/out. They're not as powerful size for size. (part of the benefit of having a separate controller). They are more expensive.

I ran into a lot of stuffy, crusty old set in their way people who were "just put a fore system in it and be done with it." I wanted better. Luckily others don't have to breach through obstacles now and can have a variable controlled, in tank brushless arrangement that doesn't run full tilt all the time wearing and tearing on the electrical system and heating the fuel.

 

[engineermike]

Basically does running 18V in shorter pulses equivocate to the same pump output as running 13V in slightly longer pulses?

I'll admit I'm not 100% sure of the following, so take it FWIW.

In a brushed DC motor, PWM simulates lower voltages by switching full voltage on and off at some set frequency and modulating the pulsewidth when the voltage is "on" to change the power delivered. The frequency does not determine the speed and just needs to be high enough to simulate a constant reduced voltage.

A brushless DC motor runs near "sync speed", such as an AC motor attempts to run constant speed synchronous to the grid at 60 hz. So rather than modulating the pulsewidth at constant frequency, it modulates the frequency and the motor attempts to match it in speed. (That said, it typically does use PWM to create a sine wave since it only has DC available, but that's another rabbit hole). That said, the motor is always attempting to run sync speed, whatever that may be, with some small amount of slip. More voltage wouldn't increase the controller output frequency, unless of course the controller has the smarts to increase the speed as a function of voltage available. If the controller output speed doesn't increase with voltage, then the only effect more voltage would have is to reduce slip. This slip usually isn't more than about 5%.

 

[Angrey]

I'll admit I'm not 100% sure of the following, so take it FWIW.

In a brushed DC motor, PWM simulates lower voltages by switching full voltage on and off at some set frequency and modulating the pulsewidth when the voltage is "on" to change the power delivered. The frequency does not determine the speed and just needs to be high enough to simulate a constant reduced voltage.

A brushless DC motor runs near "sync speed", such as an AC motor attempts to run constant speed synchronous to the grid at 60 hz. So rather than modulating the pulsewidth at constant frequency, it modulates the frequency and the motor attempts to match it in speed. (That said, it typically does use PWM to create a sine wave since it only has DC available, but that's another rabbit hole). That said, the motor is always attempting to run sync speed, whatever that may be, with some small amount of slip. More voltage wouldn't increase the controller output frequency, unless of course the controller has the smarts to increase the speed as a function of voltage available. If the controller output speed doesn't increase with voltage, then the only effect more voltage would have is to reduce slip. This slip usually isn't more than about 5%.

I fully admit that I struggle with wrapping my brain around how all of it works (comprehensively). My initial post was more about the end result and the principle involved. If spinning the pumps at a higher voltage would result in the same power consumption (just less amps at higher voltage) to maintain the pump output the controller is commanding.

I think that's going to be the case, unless as you pointed out, the controller is already near it's low/floor limit.

I think of it similarly to injector pulses. If you want a certain volume of fuel, you can ask for longer pulses at lower pressure or you can ask for shorter pulses at higher pressures. The physical way the brushless pump operates probably gets a vote (eventually) in that the pump either can't operate or can't operate smoothly if you're trying to feed it 2000V in infinitesimal pulses vs feeding it 5V in a nearly steady/ground condition.

Ultimately, I'm hoping (correctly) that the pump outputs the same, it just has more legs up top with 18V instead of approx 13V. We'll see.

If not, I'll have to employ some sort of ramp or stage control (like everyone else does) with the booster.

 

[Angrey]

If I'm not understanding how it works and it's modulated by frequency (i.e. the same pulse width, just faster cycles) then it would still arrive at the same result. The controller would just send a lower frequency at 18V to maintain the same pump output than it does at higher frequency (and 13V).

 

[markmurfie]

The control signal, from the ECU or regulator as you have setup, is an open at low frequency and a ground at high frequency.

The voltage going to the controller and being modulated to run the pump speed is a seperate circuit. The pump is not getting an alternating current.

Think of the pump controller as a relay or transistor.

 

[Angrey]

The control signal, from the ECU or regulator as you have setup, is an open at low frequency and a ground at high frequency.

The voltage going to the controller and being modulated to run the pump speed is a seperate circuit. The pump is not getting an alternating current.

Think of the pump controller as a relay or transistor.

I was off on my terminology, not the "controller" but the electronic regulator signal/command. I realize the controller just passes through, I'm talking about the brain.

If my thinking is correct, the "brain" will simply send either slower frequency (still unclear on that) or smaller pulse widths at 18V to maintain the same pump output as it was requesting at 13V (which has either higher frequency or longer pulses). in the end, the energy consumption from the DC batter should be the same (18V at lower amps or 13V at higher amps). IF that's the case, THAT is what I'm trying to achieve. Then I don't have to employ a stage or ramp to bring in the booster. I can simply run the booster full time and the system will draw the same energy at low demand and just have a higher ceiling/output at full demand.

 

[markmurfie]

What data logging systems do you use?

 

[Angrey]

What data logging systems do you use?

I have an HPT Ngauge but again, there's nothing to log. The fuel system is completely independently controlled. The only thing I can observe is fuel pressure and the PCM injector information.

As far as pump voltage or consumption, none of that is tied into the car anymore. It's all controlled through the Fuelab regulator and the DW pumps/controllers. I use the factory FPDM as a trigger (to turn the pumps on and cycle/prime them per OE strategy) but otherwise once they're triggered everything runs independent of the PCM tune.

The Fuelab regulator has a data wire/output but I haven't tried to approach the brain damage of how I can tap into that.

 

[engineermike]

Angrey, that's not how a brushless DC motor works. The controller output *frequency* is what controls the speed. The motor attempts to run the speed "synchronous" to the voltage input frequency. It's not the same as PWM. So a freqency of 100 hz might coincide with, say, 5,000 rpm and 50 hz would force it to 2500 rpm. If you command a certain frequency and add load to the motor, the amp draw will increase until it over-loads. Running higher voltage will reduce the amp loading for a given power output.

DC PWM use constant frequency but vary the pulsewidth duty cycle to modulate power.

 

[markmurfie]

With one of the gauges inputs log the voltage on the green and black wire of the regulator. I gave you the transfer function, but you can just look directly at the voltage.

 

[markmurfie]

If you trust the grounding in the car, is just attach the green wire to the input pin on the back of the ngauge.

 

[Grimreaper]

Rc hobby world does this everyday. Different cell count on same kv motors/controllers. Increased final rpm from same throttle input with higher voltage. Massive current draw, 60 to 150amps on tipin from logs I've found on those boards. (Yes, people log their rc car/plane/boat apparently!)

And to further the point of how outdated this device is.. the rc world has had tunable brushless motors for 15+ years.

 

[Angrey]

I'm still unclear and you guys are speaking indirectly or in generalities. I want you to respond like I'm a 4th grade science student.

If I gather correctly, the Fuelab regulator sends out a variable voltage to the DW brushless controller dependent upon the desired load (we could represent as a %, 0-100, probably with some sort of minimum floor value, like 15%).

If I'm understanding it correctly, the DW brushless controller then uses VFD (variable frequency) to correspondingly increase or decrease pump output. So it's not pulse width managed but frequency managed.

Again, if that's the case:

Let's say (for argument's sake) that the Fuelab regulator wants to establish a baseline pressure (on the return) that results in 30% of it's voltage range (at 13v). It continues to modulate that voltage until the DW controller and pumps come to steady state. With some corresponding frequency out of the DW contoller.

Keeping everything the same, we increase the voltage from 13V to 18V, now at that previous frequency, the pumps are supplying more fuel. More fuel/flow results in an increase in the return line flow, which triggers the Fuelab regulator to try to reduce the flow by lowering it's command voltage. The DW controller then reduces the frequency to find the desired steady state output/flow.

The end result is that the pump outputs the same baseline flow, it just gets there with either higher voltage and less frequency (18V) or lower voltage and more frequency (13V).

If that's the case, then the resulting draw from the battery is the same ENERGY. The pumps will draw roughly the same current going into the booster, which will convert it into an 18V supply.

In theory, as long as either the fuelab regulator isn't near the bottom of it's available adjustment and if the DW controller isn't near the bottom of it's variable frequency, the system should draw the same amount of power from the battery whether it's on 18V or 13V when trying to establish the low level steady state flow (at say idle).

However, as the engine consumption spikes and the corresponding reduction in return side flow, the regulator calls for more fuel and the DW controller increases frequency to match.

In theory, it should end up being the best of both worlds. The same draw at low level with potentially a higher maximum draw at max effort.

This would preclude the need to employ the booster with either a ramp in or a stage activation in order to maintain that desired low level pump draw at idle and non-spirited driving conditions, and still allow more output capability at full tilt. There'd be some slight inefficiency/losses from the booster (via heat and resistance).

The whole thread started by me wanting to know if running the booster full time on a flow variable system would result in additional low level draw on the battery/electrical system.

 

[markmurfie]

The pumps circuit.

DC voltage goes into the DW brushless controller. Pulses of DC voltage comes out of it and controlls the speed of the pump. The slower the pulses the slower the pump speed, the faster the pulses the faster the pump speed.

" This controller accepts an 8-22v DC input and creates a digital signal to allow effecient and precise speed and torque control of the pump. "

A digital signal is not a sign wave, its a square wave if anything. This wave doesn't make the voltage going to the pump AC. The reason why I would still call it DC voltage out is because the polarity is not flipping. Positive and negative on the pump is always positive and negative.
Anyone saying the pump remains at a constant speed like an AC motor..... Is wrong. It has to change speeds in order to change flow rate.
Im not asking you to log or get into any of this. How it works exactly doesn't matter and discussion on it is pointless. If the regulator is varying and sending a control signal you know its varying the flow of the pump controlling it to a desired amount of return flow. you said it has already supported high power outputs.

The control circuit.

The regulator does not send out voltage to the pump controller. The pump control wire is grounding through the regulator. Therefore, if you want to think about it from a voltage point of view, the pump controller is sending a voltage to the regulator. The regulator is opening and closing very fast, 100hz to 5000hz. The more time is is open(low frequency) the less the pump controller gets to send to ground and it controlls the pump circuit properly. The more time it spends closed (high frequency), the more the controller gets to send to ground and it controls the pump circuit properly. The pump controler supports control devices that can range from 20hz-200khz. The regulator falls in that range, so it should properly ground and control fine, but with out a log of the regulators output, we dont know what its doing or what 18V going into the pump circuit will do.

Sponsored