Flow vs Pressure

[parmas]

I am making this thread to ease my mind that I am going the right route.

This Aquamist graph shows that with an Aquamist nozzle of 1.2mm @ 160psi the pump is able to flow 600cc.

If you use 4 x 1.2mm nozzles will the pump be able to flow 600cc x 4 = 2400cc?

[Richard L]

No, it won't flow 2400cc/min. This is because the pump cannot sustain 160 psi at 2400cc/min. You need a more powerful pump.

[parmas]

No, it won't flow 2400cc/min. This is because the pump cannot sustain 160 psi at 2400cc/min. You need a more powerful pump.

So the chart displays the maximum flow in case one nozzle is used. In that case if I want the maximum all I can do is 200cc x 4 nozzles per pump?

[Richard L]

The maximum flow at 160psi is approximately 1600cc/min. This is the aquatec 5800-series pump.

All pumps on the market are quoting "dead-headed" pressure. Totally meaningless. As soon as the spray starts, the pressure will drop. Higher "dead head" pressure setting will drop much faster than a lower "dead-head" pressure setting.

[parmas]

The maximum flow at 160psi is approximately 1600cc/min. This is the aquatec 5800-series pump.

All pumps on the market are quoting "dead-headed" pressure. Totally meaningless. As soon as the spray starts, the pressure will drop. Higher "dead head" pressure setting will drop much faster than a lower "dead-head" pressure setting.

Let's take 1600cc/min as Maximum flow (0.9mm(M5.3) nozzle) for a 4cylinder engine :

A - 1 Pump = 1600cc @ 160psi = (400cc x 4 nozzles)

Good for mixing 75% pump fuel and 25% Water/Methanol


B - 2 Parrallel Pumps = 3200cc @ 160psi = (800cc x 4 nozzles)

Good for mixing 50% pump fuel and 50% Water/Methanol


C - 2 Series Pumps = 1600cc @ 320psi = (400cc x 4 nozzles)

BEST for mixing 75% pump fuel and 25% Water/Methanol. SUPER FINE MIST


D - 3 Parallel Pumps = 4800cc @ 160psi = (1200cc x 4 nozzles)

Good for mixing 25% pump fuel and 75% Water/Methanol


E - 2 Parrallel + 2 Series Pumps = 3200cc @ 320psi = (800cc x 4 nozzles)

BEST for mixing 50% pump fuel and 50% Water/Methanol. SUPER FINE MIST


F - 4 Parallel Pumps = 6400cc @ 160psi = (1600cc x 4 nozzles)

BEST for operating engine @ 100% Water/Methanol injection


Calculating Fuel flow for 400Bhp @ 1.2 BSFC 100% Methanol = 1260cc/min x 4nozzles. Using 25/75 Water/Meth, one should compensate for water flow and Setup F configuration of 4 Parrallel Pumps would be needed.


Calculating Horsepower loss per WATT taking account pump is operating at 14Volts x 10Amps

1 Pump = 140Watt = 0.19Hp loss
2 Pump = 280Watt = 0.38Hp loss
3 Pump = 420Watt = 0.57Hp loss
4 Pump = 560Watt = 0.75Hp loss

Maximum of 1Hp loss @ 100% Meth fuel injection. Is it worth it ?

What is your point of view Richard ?

[Richard L]

Are you trying to flow 1600cc/min of methanol per cylinder at 160psi?

Just use one large pump with a "return to tank" pressure regulator is best. This is because the pressure drop across the entire flow range in not linear.

You will experience starting problem in cold morning, much worse than ethanol due to high latent heat value of methanol.

[parmas]

1600cc what I calculated, It could be more or less but better be prepared for much than for less.

If you know one big pump ready for methanol able to use 150+psi pressure and flow 6400cc/min please share.

I will not have issues with starting cause I will work with pump fuel for starting/idle/cruising/low boost. At certain boost levels the base fuel will eventually be replaced by methanol injection.

Seems like the best possible setup a street car should actually have although I am ready for critics to say their way.

[Richard L]

I don't know of any pump and can run at 160psi continuously.

They might be able to make you one:
http://www.micropump.com/product_overview.aspx

[rotrex]

I would suggest you use a different architecture
Set of petrol fuel injectors at the OEM install location for petrol
2nd set of methanol compatible fuel injectors further up the intake runner or pre turbo fed by a methanol compatible fuel pump with a pressure regulater and return line into a small fuel cell, say 10l. Essentially a second fuel system. This way you get all the flow you want with the control you want. This has been done before and it works great.

Use your current water injection system to spray any water you need or want on top of that. As flow requeirements are now much reduced, use your current system as is. You may add 15% methanol to the water tank to keep it sterile.

There is no need to inject methanol and water at once as a mix. It has its merits, but I would not try to build a classic water injection system with membrane pumps to provide the entire fueling of a car. There are suitable and reliable pumps,fuel lines, connectors, fuel pressure regulators and injectors on the market for just that. Fuel injectors will also give you the precise metering capability out need for the same money as 4 water/meth valves and jets. They are plug and play to any ECU with 2nd injector bank capability such as the Emerald K3 I am using. Hondata will likely support this as well.

Besides methanol those pumps, injectors etc. would also be E85 compatible if you shall have the desire to try it. It is more stable than pure methanol and available at petrol stations. You could also run a mix of e85 and methanol. You gain flexibility.

Run petrol off boost and it automatically switches to e85/methanol, 100% ethanol etc. as you wish. Water comes on top of this from the system you already have. If you go that far, do it right.

[Richard L]

Couple of issues need to be addressed:

1. All wetted part has to be water resistance, if you are not running 100% methanol.
2. There are no high pressure fuel pump on the market that exceeds 100psi

Why not replace gasoline with methanol in the fuel tank, switch to a methanol map.

[parmas]

ANSWERS IN CAPS

I would suggest you use a different architecture
Set of petrol fuel injectors at the OEM install location for petrol
2nd set of methanol compatible fuel injectors further up the intake runner or pre turbo fed by a methanol compatible fuel pump with a pressure regulater and return line into a small fuel cell, say 10l. Essentially a second fuel system. This way you get all the flow you want with the control you want. This has been done before and it works great.

WHEN SPRAYING LOTS OF METH, SOME DISTANCE IS NEEDED TO LET EVAPORATION AND AIR MIXING TO HAPPEN EFFICIENTLY. STILL THINKING OF THE POSITIONS...

I THOUGHT OF A SYSTEM LIKE YOU DESCRIBED BUT THERE ARE TWO PROBLEMS.

1. COST

4 NOZZLES + 4 PUMPS + TANK IS LESS EXPENSIVE THAN 4 METHANOL INJECTORS + REGULATOR + METHANOL PUMP + TANK

2. CONTROL/EFFICIENCY

THE INTENTION IS TO USE A MAP ACTIVATED BY PWM DUTY RPM VS BOOST TO CONTROL FLOW. HALTECH SPORT 1000 ECU IS NOT CAPABLE TO CONTROL A 2ND INDEPENDENT SET OF INJECTORS.

Use your current water injection system to spray any water you need or want on top of that. As flow requeirements are now much reduced, use your current system as is. You may add 15% methanol to the water tank to keep it sterile.

THE REAL PROBLEM IS SINCE THE CAR IS STREET DRIVEN I NEED TO TAKE PRECAUTION FOR MIXTURES ABOVE 50/50. METHANOL AT SUMMER TEMPS IN THE TRUNK WILL BE DANGEROUS

There is no need to inject methanol and water at once as a mix. It has its merits, but I would not try to build a classic water injection system with membrane pumps to provide the entire fueling of a car. There are suitable and reliable pumps,fuel lines, connectors, fuel pressure regulators and injectors on the market for just that. Fuel injectors will also give you the precise metering capability out need for the same money as 4 water/meth valves and jets. They are plug and play to any ECU with 2nd injector bank capability such as the Emerald K3 I am using. Hondata will likely support this as well.

AS SAID ABOVE ... ADDING THAT METHANOL LIKES IT RICH. NO MATTER HOW MUCH YOU INJECT THE BETTER IS ABOVE STOICH. NO NEED FOR PRECISE METERING.

Besides methanol those pumps, injectors etc. would also be E85 compatible if you shall have the desire to try it. It is more stable than pure methanol and available at petrol stations. You could also run a mix of e85 and methanol. You gain flexibility.

E85 IS NOT PRESENT IN OUR COUNTRY AT PETROL STATIONS. YOU NEED TO BUY DIRECTLY FROM RACING FUEL COMPANY JUST LIKE METHANOL. ALSO E85 OR ETHANOL IS MORE EXPENSIVE AND LESS POWERFUL THAN METHANOL. WHY WOULD I USE IT THEN?

THE ONLY DISADVANTAGE OF METHANOL IS FUEL CONSUMPTION.

Run petrol off boost and it automatically switches to e85/methanol, 100% ethanol etc. as you wish. Water comes on top of this from the system you already have. If you go that far, do it right.

YES THAT IS THE PLAN... I WOULD APPRECIATE ANY PRECAUTIONS

[parmas]

Couple of issues need to be addressed:

1. All wetted part has to be water resistance, if you are not running 100% methanol.

I THOUGHT 100% METH HAS MORE SEVERE CONSEQUENCES THAN MIXING WITH WATER? COULD YOU EXPLAIN THIS ISSUE?

2. There are no high pressure fuel pump on the market that exceeds 100psi

EXACTLY. ONE OF THE REASONS USING COMBINED PUMPS.

Why not replace gasoline with methanol in the fuel tank, switch to a methanol map.

I DO NOT RACE EVERYDAY OR EVERYTIME. IF I WENT FOR A SUNDAY DRIVE AND IT DECIDED TO TAKE IT EASY OFF BOOST, METHANOL WILL NOT BE USED.

ALSO THE STOCK FUEL TANK IS NOT PREPARED FOR METHANOL SO CORROSION MIGHT BE AN ISSUE OF INCREASING COSTS.



,,.......,......

[parmas]

Thinking back at the configuration of the injection setup :

Trial number 1 100% Water Mixture :

Would be using 4 pumps : two in series and two in parallel.

From experience and studies when injecting water the primary aspect is droplet size. Droplet size works in direct relation to pressure. The more pressure used, the less the droplet size is. Working 2pumps in series would achieve a pressure of 320-400Psi .

Nozzle size is still unknown although positioning two nozzles one pre-turbo and one post turbo could be enough.

Direct port would be ideal for non-balanced plenums as not all cylinders could benefit equally from the setup. The problem is that water would not have enough time to evaporate if "large" nozzles are used.

Combination of direct and post/pre turbo total of Six nozzle would be the best viable option for best distribution and cooling.

[rotrex]

Fuel injectors, also ones for methanol, only need 45PSI. So any alcohol compatible fuel pump should do.
If you use a methanol compatible fuel cell, e.g. Small steel tank, there is no issue with having it in the car. The tank is vented as any other fuel tank.

The boiling point of methanol is not that low. Folks use plastic canisters of methanol to run fuel cells in their camper vans. They typically are used in hot areas.

[rotrex]

Check flow rates of air assisted foggers. Some have rather high flow rates.

[parmas]

Fuel injectors, also ones for methanol, only need 45PSI. So any alcohol compatible fuel pump should do.
If you use a methanol compatible fuel cell, e.g. Small steel tank, there is no issue with having it in the car. The tank is vented as any other fuel tank.

The boiling point of methanol is not that low. Folks use plastic canisters of methanol to run fuel cells in their camper vans. They typically are used in hot areas.

This is a billet methanol fuel injector :

https://www.moranmotorsports.com/product/160-lbhr-mre-billet-atomizer-injector

Cost : $188 x 4 = $752 = £518

Aquamist Jets = £35 x 4 = £140

And that is only injectors.... A high flow methanol pump like Aeromotive costs more than 6 pumps altogether apart fuel regulator and engine management....

Fuel tanks: I always so plastic tanks in methanol setups and in a way I like it. Steel tend to heat the liquid more plus could be prone to corrosion plus it is more expensive and "heavier" aswell.

To be exact methanol's boiling point is 64.7DegC. Let the car sit for a few hours in the sun on a summer day. I am sure you will see 60+DegC in it (tried and tested). Liquid inside would be less though.... just taking pre-cautions...

[Richard L]

,,. All wetted part has to be water resistance, if you are not running 100% methanol.

I THOUGHT 100% METH HAS MORE SEVERE CONSEQUENCES THAN MIXING WITH WATER? COULD YOU EXPLAIN THIS ISSUE?

- Water causes sever corrosion on iron alloy than methanol. Fuel injectors will seize.

- There is also issue regarding using an intermittent pump for continuous duty.

- 300-400psi line pressure will impede the operation of and inline valve.

[parmas]

- Water causes sever corrosion on iron alloy than methanol. Fuel injectors will seize.

What mixtures/flow% do you recommend then?

- There is also issue regarding using an intermittent pump for continuous duty.

The pump will not be used continuous but on demand according rpm and boost. Check ecu table attached. Also check attached screenshot directly from Aquatec. The pumps could be used to run continously for hours

- 300-400psi line pressure will impede the operation of and inline valve.

You mean the check valve ? If the cracking pressure is 15psi, the valve still needs 15psi before opening. The only possible issue is if the valve withstand the 300-400psi range



..........

[rotrex]

The injection control valve will only work to some max. Pressure. Check the data sheet of the valve you use.
Doubling pressure only gains you some 40% more flow. Pipe fittings become challenging at such high pressure.
The pump case might not sustain 450PSI vs. the outside and leak or crack.
The serial arrangement is not necessary. Just use them in parallel and get the jets yOu need for the flow you need. you can always just add more.

My system is based on the old AM 2c system. I have now replaced the race pump with a Flotec pump. Otherwise it stil works the same. 150 PSI rail pressure buffered by a accumulator. The mix is metered by a HSV PWM control valve. I think AM now seems a FAV (fast acting valve). My flow indication is the pump running indicated by a LED: if it blinks or illuminates, the mix is flowing. The faster it blinks, the more I flow.

There is no reason to build something like this, just with 4 pumps and the according number of jets. Likely a single FAV will flow enough.

Fuel injectors are ethanol and fuel compatible these days. MEthnaol compatible injectros are likely identical except someone did the effort to test compatibility.
Common to all fuel injectors is that they cannot tolerate water similar to fuel pumps with ther fuel cooled rotors. They corrode from the water.
You need special metering valves that are designed to be compatible with with water. They are not common. Aquamist is one of the few (actually the only one) selling those.
I have a HSV under ECU control and it is one fine piece of valve. Current modulation frequency is set to 80Hz.

Pure methanol corrodes aluminium if exposed over prolonged periods of time in a container, especially if contaminated by acids or bases breaking down the passivity on layer of aluminium. Water reduces this effect. That is why storing 50:50 mixes causes no issues.

Are you sure your race fuel shop cannot provide race spec e85? It would make your life much easier as you could use mostly common fuel pumps and injectors for a secondary injection system.
Then again, add water as necessary with your current WMI system.

[Richard L]

..........

- You should not have any trace of water in the methanol. Even absorb from the water vapour from the air over time. Leave the fuel injector

- High pressure demands high current. Hight current means heat rise. The Aquatec specification is quite vague, " ... some pump can run for hours .... "

- I cannot see a valve that will work at 450psi at 1600cc/min. Certain not from us.

[parmas]

The injection control valve will only work to some max. Pressure. Check the data sheet of the valve you use.Doubling pressure only gains you some 40% more flow. Pipe fittings become challenging at such high pressure.

The check valve used on the Pre-turbo is branded Hamlet 316_Stainless and works in 3000psi systems. Also much expensive though worth it. The inlet/outlet is 6mm I don't know how much I can flow to it until I test it. See attached..


The pump case might not sustain 450PSI vs. the outside and leak or crack.
The serial arrangement is not necessary. Just use them in parallel and get the jets yOu need for the flow you need. you can always just add more.

Well the pump case could be an issue, I have to check with Aquatec about it. Parallel would be best for methanol flow needs while Series would be better for water droplets

My system is based on the old AM 2c system. I have now replaced the race pump with a Flotec pump. Otherwise it stil works the same. 150 PSI rail pressure buffered by a accumulator. The mix is metered by a HSV PWM control valve. I think AM now seems a FAV (fast acting valve). My flow indication is the pump running indicated by a LED: if it blinks or illuminates, the mix is flowing. The faster it blinks, the more I flow.

I have an AEM flow gauge and software datalogging. I can datalog how much injected during a run. Could you give more details on the HSV PWM control valve and Flotec pump you have like links etc..

There is no reason to build something like this, just with 4 pumps and the according number of jets. Likely a single FAV will flow enough.

Did not understand the above...

Fuel injectors are ethanol and fuel compatible these days. MEthnaol compatible injectros are likely identical except someone did the effort to test compatibility.
Common to all fuel injectors is that they cannot tolerate water similar to fuel pumps with ther fuel cooled rotors. They corrode from the water.
You need special metering valves that are designed to be compatible with with water. They are not common. Aquamist is one of the few (actually the only one) selling those.
I have a HSV under ECU control and it is one fine piece of valve. Current modulation frequency is set to 80Hz.


Could you indicate what Aquamist valve are you referring to?

Pure methanol corrodes aluminium if exposed over prolonged periods of time in a container, especially if contaminated by acids or bases breaking down the passivity on layer of aluminium. Water reduces this effect. That is why storing 50:50 mixes causes no issues.

Are you sure your race fuel shop cannot provide race spec e85? It would make your life much easier as you could use mostly common fuel pumps and injectors for a secondary injection system.
Then again, add water as necessary with your current WMI system.

I meant E85 is not available at the pump station, you need to go buy it directly from the Racefuel company like VPracing. As you see I like to go big so methanol is the first priority fuel if it happens the case of any issues along I might reconsider E85 or pure ethanol then



....................

[parmas]

- You should not have any trace of water in the methanol. Even absorb from the water vapour from the air over time. Leave the fuel injector

This issue will rise only if I use fuel injectors instead of nozzles right or will effect the primary pump fuel injectors?

- High pressure demands high current. Hight current means heat rise. The Aquatec specification is quite vague, " ... some pump can run for hours .... "

Agreed... need to clear this out with Aquatec...

- I cannot see a valve that will work at 450psi at 1600cc/min. Certain not from us.

Jus posted one I already have from Hamlet able to work in the 3000psi region, don't know the flow though



...................

[Richard L]

I was referring to a PWM valve, not a checkvalve.

[parmas]

I was referring to a PWM valve,

What if I use 4 pumps each connected directly to each nozzle (4) and each pump has its own PWM control valve. This will eliminate problem of pumps overheating/overpressing + PWM control valve flowrate.

Option 1 : All 4 PWM control valves wired together so they activate all together according duty cycle managed by ecu map.

Option 2 : Each PWM control valve connected to primary injectors (example injector 1 with PWM nozzle 1 etc..) Since the injection needs to happen on boost the ecu switches on pumps when the engine is on boost and manage flow according PWM duty cycle managed by ecu map.

What do you think Richard/Rotrex ?

[rotrex]

As the jets mostly flow all the time under boost,I.e. Valve is open, you might get away with a single or two fast acting valves. Timed sequencial injection has little benefits for what you want to do.
Richard should know the max flow rate of the current fast acting valve FAV.

[Richard L]

There are two FAVs with flow rates 1600cc (806-239) and 1300cc (standard 806-239B).

[parmas]

There are two FAVs with flow rates 1600cc (806-239) and 1300cc (standard 806-239B).

99% Methanol needs to be used....

I am trying to find find one that actually can work with 100% water if possible.

Any ideas Richard?

[parmas]

There are two FAVs with flow rates 1600cc (806-239) and 1300cc (standard 806-239B).

Richard just noticed the attached statements :

Aquamist FAV : Warning: never run hydrocarbon contaminated fluid through this valve. The internal EPDM seal is subject to severe swelling if 99% pure methanol is not used. Always use 99% methanol from a factory sealed canister. Transfer pump or jugs should be free of any previous hydrocarbon usage. Washing the container with detergent will not stop methanol drawing out the hydrocarbon from a plastic container laced with gasoline.

Aquamist Pump : Warning: Prolonged use of 100% methanol may cause
premature pump failure and may not be covered under
warranty - this warning applies to all Aquatec pumps.

Is this contradictory? 99% Methanol may damage pump while less may damage FAV....

[rotrex]

He just means 99% purity methanol, but this mixed with distilled water.

Richard, can the older HSV take a ethanol / water mix?
Ethanol is way cheaper than methanol for me here in Germany.

[Richard L]

The HSV has EPDM seals, OK with methanol or ethanol as long as it is 99% pure.

[parmas]

He just means 99% purity methanol, but this mixed with distilled water.



Maybe Richard could explain it better?

[Richard L]

I meant 99% pure methanol without traces of hydrocarbon.

[Richard L]

1600cc what I calculated, It could be more or less but better be prepared for much than for less.

If you know one big pump ready for methanol able to use 150+psi pressure and flow 6400cc/min please share.

I will not have issues with starting cause I will work with pump fuel for starting/idle/cruising/low boost. At certain boost levels the base fuel will eventually be replaced by methanol injection.

Seems like the best possible setup a street car should actually have although I am ready for critics to say their way.


I have just remembered your set up is based on progressive pump speed (PPS) rather PWM valve. Ignore my PWM valve remarks.

If you are flow this large amount, I am not sure the PPS algorithm will work satisfactorily due to the non-linearity and dynamic range.

[parmas]

Actually I am still determined to find the best setup PPS or PWM . The question is finding the right components that work efficiently without too much complexity and too much expenses

[rotrex]

I personally like the PWM set-up.
Initially it was a Aquamist 2c. Since the race pump dropped in flow rate, I have replaced it with a 160PSI 2.xl/min pump of eBay for €26!. The harness works for the most part except that the pump flow indicator LED is now driven by the 12V line after the pump's micro switch.

The pump has a integrated pressure switch. It stops at about 10bar. Once pressure drops below 8 bar, the pump automatically activates itself. The switch closing fires also the LED. This is my flow indicator. As in the 2c system, I maintained the 30cc accumulator. This dampens the pump switching cycles.
So overall it behalfs like a fuel injection system with pump based pressure control. A bit more crude, but hey.
With this system you end up with a line that is permanently pressurized with your mix.
Next you need to control flow. For this purpose Aquamists sells very suitable valves. I have the older HSV (high speed valve). Now they sell a version called FAV (fast acting valve).
The valve is controlled by my ECU. I use a boost control map with PWM output for this purpose. Most ECUs feature auxiliary PWM maps. The boost control map in open loop mode allows to set a PWM percentage based on throttle position vs. rpm.
This allows me to deterministically set my injection onset and flow rates over the rpm range. The beauty is the proportionality of PWM. If I set it to 70%, I get about 70% of the max. flow rate, if I set it to 30%, I get about 30% flow. It makes the injection a very reproducible and mappable process. I do not need any Hobbs switches to control the system. I still have one to automatically switch to a non WMI map once rail pressure drops below like 5 bar.
Cableing is simple. 12V and ground to the loom, LED and tank level lines to the front., 1 wire from the ECU to control the HSV and 1 line for the map switch as a fail safety.
There is no separate controller and its associated wiring and adjustment knobs. It is all ECU controlled.

Under PWM control, you get spray even at low flow rates as during each opening of the HSV, a full pressure wave passes through the line to the jets.
I can test the pump by pressuring the throttle with the engine stopped.
Over the last 25% of the throttle the valve opens from 0 to 100%.
It works beautifully and you get spray over a big part of the control range.

Pump speed based systems only increase flow by about 40% for a doubling of the pressure.
This is probably why those 250PSI pumps are so popular. You need the dynamic range on the top to get a sufficent controllable flow range as at low pressures, spray is poor.

If you change jet size, all the scaling does not work anymore as pressure vs. flow is now changed. Exception are internal bypass pumps with a constant pressure.

[MaestroB]

Hello!

Sorry for jumping in but actually I'm not replying to this post but as a new comer into the forum, I just want to say hello to you old folks.


Thanks

[parmas]

I personally like the PWM set-up.
Initially it was a Aquamist 2c. Since the race pump dropped in flow rate, I have replaced it with a 160PSI 2.xl/min pump of eBay for €26!. The harness works for the most part except that the pump flow indicator LED is now driven by the 12V line after the pump's micro switch.

The pump has a integrated pressure switch. It stops at about 10bar. Once pressure drops below 8 bar, the pump automatically activates itself. The switch closing fires also the LED. This is my flow indicator. As in the 2c system, I maintained the 30cc accumulator. This dampens the pump switching cycles.
So overall it behalfs like a fuel injection system with pump based pressure control. A bit more crude, but hey.
With this system you end up with a line that is permanently pressurized with your mix.
Next you need to control flow. For this purpose Aquamists sells very suitable valves. I have the older HSV (high speed valve). Now they sell a version called FAV (fast acting valve).
The valve is controlled by my ECU. I use a boost control map with PWM output for this purpose. Most ECUs feature auxiliary PWM maps. The boost control map in open loop mode allows to set a PWM percentage based on throttle position vs. rpm.
This allows me to deterministically set my injection onset and flow rates over the rpm range. The beauty is the proportionality of PWM. If I set it to 70%, I get about 70% of the max. flow rate, if I set it to 30%, I get about 30% flow. It makes the injection a very reproducible and mappable process. I do not need any Hobbs switches to control the system. I still have one to automatically switch to a non WMI map once rail pressure drops below like 5 bar.
Cableing is simple. 12V and ground to the loom, LED and tank level lines to the front., 1 wire from the ECU to control the HSV and 1 line for the map switch as a fail safety.
There is no separate controller and its associated wiring and adjustment knobs. It is all ECU controlled.

Under PWM control, you get spray even at low flow rates as during each opening of the HSV, a full pressure wave passes through the line to the jets.
I can test the pump by pressuring the throttle with the engine stopped.
Over the last 25% of the throttle the valve opens from 0 to 100%.
It works beautifully and you get spray over a big part of the control range.

Pump speed based systems only increase flow by about 40% for a doubling of the pressure.
This is probably why those 250PSI pumps are so popular. You need the dynamic range on the top to get a sufficent controllable flow range as at low pressures, spray is poor.

If you change jet size, all the scaling does not work anymore as pressure vs. flow is now changed. Exception are internal bypass pumps with a constant pressure.

Thank you rotrex for your explaining. I like your system although monitoring flow with an LED seems too simple and I am sure 90% of the time you actually look at the road.

Anyway, just spent some time checking other Aquatec pumps particularly the 550 Series and the 230Volt series. I made a chart for you to review.

Notes :

- The 230Volts consume more power than the 12V series but flow slightly better

- The 550series have double the flow or more the 5800series but they are not capable to withstand pressures for long period of time


Taking into account the rate of evaporation of Methanol vs Gasoline. If gasoline fuel injection systems work with 45psi to 80psi would methanol injection systems need less pressure to vaporize the fuel ?

So why do I need 160psi pressure if I could be good with 80psi @ Pure Meth?

What about Water/Meth mixtures would I still have good evaporation @ 80psi and W25/M75 ?

[rotrex]

If you just want to get the methanol to burn in the cylinder as a fuel, the spray quality (small droplet size and narrow distribution) is not as important as if you like to preferentially cool the air. For water this is even more important as its evaporation speed is so low.
For methanol as a fuel and power as the main objective with emissions lower on the list, you get away with a crude spray. Most of the fuel, being petrol or methanol or ethanol evaporates hitting the hot valves and combustion chamber walls.

Hydraulic nozzles produce finer mist at higher pressures. This is where the need for high pressures come from. Making fine mist of water or water / meth mix.
Air assitted nozzles work with much lower fluid pressures, but for very high flow rates, need high air pressures.

Fuel injectors have overcome this droplet size to a degree by adding holes to the injectors with some having as much as 12 holes, e.g. The Denso injectors for the 2003-2009 Toyota Prius. For this car this was done to achieve a very fast fuel evaporation and homogenous distribution lowering emissions. Lean pockets increase Nox emissions, rich pockets CO and HC.

so yes, for pure methanol or mixes high in methanol content, you get away with bigger droplets. This is why I proposed a methanol or ethanol compatible secondary fuel injector system. But unfortunatly your ECU does not support it.

the LED is crude, but all I need. Closed loop fueling, a very fast knock controller and pressure based fail safety take care of the rest. If it fails, I get misfires and knock. Nothing brakes. I have tried it when the race pump dropped in flow and for a strange reason the priming pump controller was also not activated anymore by the pressure switch of the race pump.
fitting a membrane pump solved this. I have a spare on in the shelf shall I got problems again.

This is new pump I got on eBay, except it says 1.1 Mpa on the label instead of 1 Mpa as on the picture.
I cut the barbs off and cut 1/8" threats into the plastic to fit pushy on connectors.
http://g03.a.alicdn.com/kf/HTB1pExxGpXXXXXaXpXXq6xXFXXXe/220361854/HTB1pExxGpXXXXXaXpXXq6xXFXXXe.jpg

[parmas]

ANSWER IN CAPS

If you just want to get the methanol to burn in the cylinder as a fuel, the spray quality (small droplet size and narrow distribution) is not as important as if you like to preferentially cool the air. For water this is even more important as its evaporation speed is so low.
For methanol as a fuel and power as the main objective with emissions lower on the list, you get away with a crude spray. Most of the fuel, being petrol or methanol or ethanol evaporates hitting the hot valves and combustion chamber walls.

AGREED


Hydraulic nozzles produce finer mist at higher pressures. This is where the need for high pressures come from. Making fine mist of water or water / meth mix.
Air assitted nozzles work with much lower fluid pressures, but for very high flow rates, need high air pressures.

AGREED


Fuel injectors have overcome this droplet size to a degree by adding holes to the injectors with some having as much as 12 holes, e.g. The Denso injectors for the 2003-2009 Toyota Prius. For this car this was done to achieve a very fast fuel evaporation and homogenous distribution lowering emissions. Lean pockets increase Nox emissions, rich pockets CO and HC.

THESE INJECTORS ARE STOCK USED FOR MY ENGINE. TRIED TO USE THEIR ORIFICE MATED WITH A BILLET NOZZLE. SPRAY WAS ALMOST LIQUID MAYBE DUE TO PRESSURE DISTRIBUTION ON 12HOLES. ALSO THE ORIFICE CORRODED AND BLOCKED HOLES WERE AN ISSUE.


so yes, for pure methanol or mixes high in methanol content, you get away with bigger droplets. This is why I proposed a methanol or ethanol compatible secondary fuel injector system. But unfortunatly your ECU does not support it.

NEED TO TRY A W25/M75 @ 80PSI PRESSURE L

the LED is crude, but all I need. Closed loop fueling, a very fast knock controller and pressure based fail safety take care of the rest. If it fails, I get misfires and knock. Nothing brakes. I have tried it when the race pump dropped in flow and for a strange reason the priming pump controller was also not activated anymore by the pressure switch of the race pump.
fitting a membrane pump solved this. I have a spare on in the shelf shall I got problems again.

This is new pump I got on eBay, except it says 1.1 Mpa on the label instead of 1 Mpa as on the picture.
I cut the barbs off and cut 1/8" threats into the plastic to fit pushy on connectors.
http://g03.a.alicdn.com/kf/HTB1pExxGpXXXXXaXpXXq6xXFXXXe/220361854/HTB1pExxGpXXXXXaXpXXq6xXFXXXe.jpg

[parmas]

the LED is crude, but all I need. Closed loop fueling, a very fast knock controller and pressure based fail safety take care of the rest.

Thinking ahead.... could you give more details on the above ?

[rotrex]

Hello Parmas,
I already did so I believe.

Fueling is controlled by the ECU running closed loop at all times employing a AFR target table.. Shall the methanol supply be reduced or seize, the ECU can add up to 30% of fuel to maintain that target AFR. I could adjust the regulation range even further.

The knock controller is a J&S Safeguard vampire unit. It works by being able to retard ignition by up to 20 deg (currently set to 10 max) within one single revolution for individual cylinders depending on knock intensity. It then reapplies fully advance by bringing back 1 (for 10 deg max range) or 2 deg (for 20 deg max retard range) every couple of revolutions.
It uses a standard BOSCH knock sensor. I have mounted their gauge, too so I can see what is happening.
It sometimes misses light knock, but the mid and heavy stuff is reliably regulated.
If you get a heavy misfire, e.g. Due to water meth failing or fuel air lock, it retards by 10 deg within a single revolution. This drops all power, but with it also all force. Light knock at part throttle you can barely feel. It typically pulls 2-4 deg.
I have tried the bigger range of 20 deg retatrd, but this essentially feels like someone pulled the plug out of your ignition coil. I believe this is more useful for NA two valve engines with old and slow combustion chamber designs that need up to 40 deg of advance at WOT to make power.
http://www.jandssafeguard.com

Then last but not least, I have mounted a pressure sensor in my water / methanol fuel rail. If the pressure drops in there, e.g. Due to pump failure or running out of fluid, it triggers a map switch. This is a ECU feature of my Emerald K3 ECU. It switches to a map I have set-up to run richer and with less ignition advance at WOT, so it runs without water / meth, albeit with less power. I do not have any boost control. If you have boost control, most turbo cars have, you can also just reduce boost and run with petrol alone.

[parmas]

Hello Parmas,
I already did so I believe.

Fueling is controlled by the ECU running closed loop at all times employing a AFR target table.. Shall the methanol supply be reduced or seize, the ECU can add up to 30% of fuel to maintain that target AFR. I could adjust the regulation range even further.

The above safety feature shall only work if you inject a small amount of methanol circa 15%. If you normally see 11Afr with meth and suddenly dropped to 13Afr, the ecu could adjust fuelling back to 11Afr but still ignition timing is according water/meth. Knock should still happen although it's more safe than nothing

The knock controller is a J&S Safeguard vampire unit. It works by being able to retard ignition by up to 20 deg (currently set to 10 max) within one single revolution for individual cylinders depending on knock intensity. It then reapplies fully advance by bringing back 1 (for 10 deg max range) or 2 deg (for 20 deg max retard range) every couple of revolutions.
It uses a standard BOSCH knock sensor. I have mounted their gauge, too so I can see what is happening.
It sometimes misses light knock, but the mid and heavy stuff is reliably regulated.
If you get a heavy misfire, e.g. Due to water meth failing or fuel air lock, it retards by 10 deg within a single revolution. This drops all power, but with it also all force. Light knock at part throttle you can barely feel. It typically pulls 2-4 deg.
I have tried the bigger range of 20 deg retatrd, but this essentially feels like someone pulled the plug out of your ignition coil. I believe this is more useful for NA two valve engines with old and slow combustion chamber designs that need up to 40 deg of advance at WOT to make power.
http://www.jandssafeguard.com

Thanks for the link. The system looks promising especially because of using stock knock sensor, features, fine-tuning and realtime-gauge.

The problem is I don't like external controllers apart ecu to manage the ignition system. I like better to let the ecu be the manager like changing map or reduce timing accordingly. Also the system mentions that retard timing by increasing dwell on coils which is not a good idea (depending how much dwell it actually increases). Coils could overheat that way.

Also how do the system understand that knocking happens on cylinder 1 and not cylinder 4 if only one middle knock sensor is used ?



Then last but not least, I have mounted a pressure sensor in my water / methanol fuel rail. If the pressure drops in there, e.g. Due to pump failure or running out of fluid, it triggers a map switch. This is a ECU feature of my Emerald K3 ECU. It switches to a map I have set-up to run richer and with less ignition advance at WOT, so it runs without water / meth, albeit with less power. I do not have any boost control. If you have boost control, most turbo cars have, you can also just reduce boost and run with petrol alone.

Good feature. What pressure sensor did you used and how did you set it from the software ?



....................................

[rotrex]

If your ECU offers knock control, even better.
The coil does not overheat as your engine is not permanently knocking to the maximum degree, say for minutes at the time. It also extends the dwell by max 10 deg in my config. So that is a few ms the most and that every few seconds. Nothing happens. At 6000 rpm for a I4 engine this corresponds to a extra 0.27ms dwell or say you have a total dwell of 3ms about 10% extra dwell every here and there. Nothing to worry about.
Most of the time the J&s does not do anything, even in Spa being at WOT some 50% of every lap. As you now get a good feel where the knock spots are, you eliminate them from your map. This way it only takes care about the exceptional knock instances.

Regarding cylinders. The units are configured for a specific engine configuration, say a inline 4 and distributor ignition with a single coil. If knock occurs, it knows that the very same cylinder is ignited 4 sparks later. It does not know if it is 1,2,3 or 4 as there is no cam phase sensor. But it knows when the same one is about to fire again. :-) And that is all that counts. You get versions for whatever engine config you have. V8 flat crank, Cross crank, I6, distributor, distributorless, 8x coil on plug, batch ignition, sequential ignition. Just tell him what you need. It is a one man company. The owner knows the designer and the programmer pretty well :-) His wife does the admin.

A single knock sensor captures knock all round the block. Aluminium is a good sound conductor. Most engines get around with a single sensor per four cylinder bank. I6 engines often have two, V12 sometimes 4.
I have observed it applying 4 different levels of retard on 4 different cylinders as indicated by 4 LEDs illuminated. If 1 to 4 cylinders have the same retard you only see 1 LED. You can also set the J&S to apply global retard on all cylinders.

The pressure switch is a Hobbs switch that was left over in my Aquamist 2c kit and the SW is a standard feature of my Emerald K3 ECU. It works from a analog input. It tells the ECU to switch maps past adjustable voltage threshold. The rest is a resistor across the switch and a internal pull up resistor :-)

[parmas]

If your ECU offers knock control, even better.
The coil does not overheat as your engine is not permanently knocking to the maximum degree, say for minutes at the time. It also extends the dwell by max 10 deg in my config. So that is a few ms the most and that every few seconds. Nothing happens. At 6000 rpm for a I4 engine this corresponds to a extra 0.27ms dwell or say you have a total dwell of 3ms about 10% extra dwell every here and there. Nothing to worry about.
Most of the time the J&s does not do anything, even in Spa being at WOT some 50% of every lap. As you now get a good feel where the knock spots are, you eliminate them from your map. This way it only takes care about the exceptional knock instances.

Regarding cylinders. The units are configured for a specific engine configuration, say a inline 4 and distributor ignition with a single coil. If knock occurs, it knows that the very same cylinder is ignited 4 sparks later. It does not know if it is 1,2,3 or 4 as there is no cam phase sensor. But it knows when the same one is about to fire again. :-) And that is all that counts. You get versions for whatever engine config you have. V8 flat crank, Cross crank, I6, distributor, distributorless, 8x coil on plug, batch ignition, sequential ignition. Just tell him what you need. It is a one man company. The owner knows the designer and the programmer pretty well :-) His wife does the admin.

A single knock sensor captures knock all round the block. Aluminium is a good sound conductor. Most engines get around with a single sensor per four cylinder bank. I6 engines often have two, V12 sometimes 4.
I have observed it applying 4 different levels of retard on 4 different cylinders as indicated by 4 LEDs illuminated. If 1 to 4 cylinders have the same retard you only see 1 LED. You can also set the J&S to apply global retard on all cylinders.

The pressure switch is a Hobbs switch that was left over in my Aquamist 2c kit and the SW is a standard feature of my Emerald K3 ECU. It works from a analog input. It tells the ECU to switch maps past adjustable voltage threshold. The rest is a resistor across the switch and a internal pull up resistor :-)

I will surely send him a message, if you want I can put you in my reference for your good name.

Cheers