When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Stevieturbo I have the alkycontrol and have it set to come on at 7psi and all in by 23psi. One thing that was odd to to me was the voltage to the pump at turn on and at max. It was like 12.2v coming on and max was like 12.8v, resolution seemed small, was expecting more like 10v-12.8 Just food for thought.
Because the voltage is not varied, and you cannot measure what is going on with a multimeter. You need a scope.
In effect to reduce speed, they're simply turning the pump off/on very fast. So it never sees anything other than either 12v, or nothing.
For 50% duty, say in a time period of 10s, just have it turned on for 5, and off for 5. But it's just a rapid off/on/off/on etc
Strangely with the recent tests.....
At 50Hz PWM, the pumps range of fl....pressure seen ( was going to say flow and it will be related, but as not measuring flow would be incorrect to say it ) vs other frequencies is a little strange.
At 50Hz, the pump certainly kicks into life better and offers more pressure at low duties. So overall it would seem less linear from low boost, to high boost.
At around 2-300Hz, the motor did not sound happy, and at the same base 55% duty tested output pressure was much reduced. However when increasing boost in steps, output pressure mirrored boost pressure, so it would offer a very good linear output of water vs boost. Which would be a good thing, because at 50hz it just seems like at low boost it might be dumping in far more water than is really necessary for the respective duty cycle
For control purposes for what we want at the engine, 2-300Hz would be best with the test I did. I just dont feel the electric motor is happy at that though. Will probably test it first actually running around 100hz
That's really interesting that the PWM frequency makes such a difference. I would have guessed that the pump would be happiest at higher frequencies than lower, but only because I read somewhere that Subaru uses 250hz for their fuel pump (and I'm not even sure that's accurate).
Did you try 400 - 500 Hz? I wonder if the pump really gets progressively less happy at higher frequencies or if it's just unhappy in that 200-300 band.
I suspect the ideal thing would be a short pulse of DC to get the motor spun up, and then drop down to whatever duty cycle is appropriate, but that might be expecting too much from an "ancient" controller.
I've never tested the Subaru controllers...generally anything that's going to make some power it would get dumped and a proper fuel system put in.
I was very surprised at the results myself, especially as a loot of people would say motor control PWM would generally be very fast with 10, 20Khz range, nevermind down as low as a SSR. Part of that is to keep the switching out of our audible range.
And certainly at 50Hz...I just hear the pump chugging. 100hz...sounds good...but 2, 3, 400hz I can really hear a whine and the pump doesnt sound happy either.
I didnt measure current to the pump in these tests...and again these were with a small single 640cc, D07 nozzle only.
Pressure is just applied to the MAP sensor via Mity vac, but are in roughly equal steps. Pump duty rising in similar steps from around 55% to 100%
But it's surprising that pump pressure output at 50Hz is very non linear..so flow maybe not changing very much throughout that range So maybe not the best for what we actually want.
With both nozzles my initial testing saw quite low pressures at say 50% duty...so a high PWM value might make that worse. Would maybe need to recheck with both nozzles installed to see how that goes.
Scaling not hugely important, red line is pump duty cycle ( directly related to boost 11psi up to 32psi at 100% duty ), white line is pressure in the water line in psi, with a 1/2sec filter applied as actual reading is quite erratic.
Maximum pressure doesnt really change much throughout.
Last edited by stevieturbo; Jul 12, 2018 at 06:06 AM.
I did also test my dual 450 Walbros that I PWM via Crydom SSR. I had run these at 650Hz for the past while and no issues.
Just a static 10s prime to see if PWM did have any effect, although of course there is an FPR there to even things out too. But around 200Hz did seem to see the highest pressure readings from the test. Only very small differences though, only around 0.1 bar, maybe 2-3psi change. So I'll leave that at 200Hz for now. Again, I've never measured current on those.
Thanks for the info Stevie. FYI I did my MS3 meth injection testing by measuring volume over time and noting spray pattern quality. What I found was the lowest freq of 11.1Hz gave me the greatest resolution from 10%-100%. The higher I went with the Hz, the less resolution I got..25% was not far off from 100% at the 78Hz or whatever was highest.
I am not sure what the pressures were but as mentioned, at the lowest Hz, spray pattern was still acceptable even with a DO14 nozzle.
It would be interesting to compare current draw vs. flow rate at different PWM frequencies and duty cycles. I have no clue what that relationship would look like, but I'm guessing that the pump would run coolest and live longest at whatever frequency gives the lowest current draw for the required flow rate.
TBH as it's only used in short duration I really dont think longevity in that respect is an issue.
I would guess pump current will be pretty closely related to output pressure, but if the pump is not performing at the higher frequencies, it just seems wrong to run them in those ranges.
I'll maybe get a BNC plug and I could hook up the current clamp to a spare input to log it on the ecu. Normally I'd use it with the scope but no reason why it couldnt just be logged on the ecu. Although I'm just going to run it at 50Hz, just dont see a lot of point at the higher frequencies.
I did more tests yesterday, and despite D07 and D10 nozzles...ie almost 1600cc worth of nozzle max flow seen was only around 1200cc/min. Pump voltage was only around 11v during those tests, so I guess things will improve a bit once the engine is running and things are up around 14v.
But it also begs the question that there must be a happy medium in terms of nozzles and maybe a 1200cc nozzle might work better ( or pair of 600's ), flow the same anyway and run at higher pressures which would be better overall.
6 Common C5 Corvette Failures and What's Involved In Repairing Them
Slideshow: From wobbling harmonic balancers to failed EBCMs, these are the issues that define long-term C5 ownership and what repairs typically involve.
Retro Modern Bandit Pontiac Trans AM Comes With Burt Reynolds' Autograph
Slideshow: A modern Camaro transformed into a retro icon, this limited-run "Bandit" build blends nostalgia with brute force in a way few revivals manage.
Top 10 Greatest Cadillac V Series Performance Models Ever, Ranked
Slideshow: Cadillac didn't just crash the high-performance luxury vehicle party, it showed up loud, supercharged, and occasionally a little unhinged...
Coachbuilt N2A Anteros Is an LS2-Powered C6 Corvette In Italian Clothes
Slideshow: A one-off sports car that looks like a vintage Italian exotic-but hides a C6 Corvette underneath-just sold for the price of a new mid-engine Corvette.
