Series II ESC (Supercharger)
 

I am taking a leap here and designed a open source hardware guide for how to supercharge your Series II with a REAL electric supercharger. No Leaf Blowers, or R/C Turbofans here.
I am going to start with some MATH

In order to make 5PSI of Boost, per liter, at 6,000 RPM, 3.25HP is required on the spinning motor driving a root type supercharger assembly.

1HP = 746 Watts of Power at 100% EFF.

A rotary engine for this formula will be 2.3x the size of the reported displacement or 3.0 Liters.

The kit I am having designed is a total of 20HP on the motor or roughly 15KW.

20HP would equal 10-15PSI on a Rotary at 6,000RPM

So the cost of this part I am having sourced is 2,600 dollars and that isn't cheap. That does not include the batteries to drive the motor, the motor controller, and the custom logic boards I am going to have to make to boost it.

Because all this is expensive and costly, I am building it without a engine for now, and will get a wrecked and salvaged RX8 to test it out on a Dyno and track.

So here are the major parts required

Air Filter
Piping
ESC Device
Insulated Piping with MAF sensor housing
Auxiliary Fuel Device (5th port, 6th port, 7th port or Methanol Injection)

Boost Computer Calculator
Fueling Computer
Health Sensors

Batteries for driving ESC
Motor Drive Controller (PWM)
Charging Mechanism (Inverter switched over to DC for charging)

External MAP Sensor
AFR Wideband
Stock MAF Voltage Tap
Stock TB Voltage Tap
ECT Sensor
Oil Pressure Sensor
Oil Temp Sensor
IAT (MAF)

Tuning Software or External ECU

This is not going to be an easy thing to do, and it would prob just be cheaper to buy a Series 1 and turbocharge/Supercharge it, or just a FD, but it is a puzzle I want to test out.

So the cost is looking like about 3,000-4,000 dollars for the whole thing including the batteries, parts, & tuning costs.

I will be getting the part together for some documentation shortly with some pictures.

The Major benefit is absolutely no parasitic draw from the drivetrain while firing and no backpressure issues from a Turbocharger. The only draw will be a Aftermarket Alternator with an increase of output via FREQ controller.

Going to be some fun so stay tuned.



Got the alternative fueling for the Methanol Injection Sorted out.

Devil's Own 300PSI pump with 3/8NPT fittings instead of 1/4 PTC
Devil's Own 6 Gallon Race tank or 2.25 Gallon Race Tank
Devil's Own DO03 Nozzle (3 GPH)
Devil's Own Progressive Voltage Controller and/or DVC-30 Controller


The tubing and fittings will be all Stainless Steel, via CO2 Tubing from a Fountain Soda Machine, and Swagelock Stainless Steel fittings.


Methanol Tank(7/8 Inch hole) to 1/4 OD Compression
?????

To covert the Devil's Own pump from 3/8NPT to 1/4 Compression (QTY 2)
SS-400-1-6

For the Inline Methanol Injection Filter 3/8NPT to 1/4 OD Compression (QTY 2)
SS-400-1-6

Inline Checkvalve 1/4OD Compression to 1/4OD Compression
SS-4C-TR-10

Nozzle Adapter 1/4OD Compression to 1/8 NPT Female
SS-400-7-2

With all of that sorted the system will be 100% sealed with a very little chance of leaking since the tubing can withstand methanol and fuel at 1800PSI because of the thick sidewalls. That will provide the alternative cooling & Fueling system. Additional Fuel injectors will more than likely be needed in the long run, but this should get things cooling so the compressor doesn't increase the air temps. The methanol mixture will be injected pre-compressor without an intercooler to supercool the intake air.

It is important that EVERYTHING be stainless steel because of the hardness of the tubing. The Stainless Tubing will not stay in a PTC fitting even with Stainless steel grabbers. Compression is the only safe way I can find. I also ran a similar system on my Mazdaspeed 3 and it worked excellent.

There are 2 things I want to add if I can figure them out. A Flow Sensor & Pressure sensor to feed back to a Sub ECU I am designing.

The way the formula was explained to me is such.

6000RPM is the peak flow rate of a ICE where it is moving the max amount of air. The becomes the max reference point of what you need at the top end. Anything lower than that 6,000 pressure will increase because optimal flow is not being achieved. That is exactly why I picked what I did because I only want to use the pressure to force more air in at lower RPMs, not higher ones. The device is only supposed to be a short buster for rapid acceleration. The size of the unit is that of an Eaton M90 supercharger. The flow rates aren't quite known because they vary based on the type of motor used since the pulley input is no longer a constant size. (Diameter) I am trying to archive a boost in toque, not really a boost in HP. One gives you the other in ways, so that is one way to look at it.

All the pressure numbers come from open flow, not below 6,000. Below 6,000 it will flow less so pressure will increase because of a bottleneck. I understand the Renesis isn't the same thing as the ICE the formula was designed for, but I am not trying to use it to widen my power curve, simple increase the torque at lower end by adding a pushing force on the flow rate.

The motor specs allow it to run from 24v to 96v. I never plan on hitting 100% Duty Cycle on the PWM driver so I was going to start with a Gold Cart controller and if I needed to reduce the amperange or gain something I would increase the voltage of the controller. 96v controllers are harder to find than an Ezgo golf cart controller so it was just part of the stepping stones of getting it to work.

Because of the cost of these device, I want to try it on more than just the RX8. If I pick something like a supercharger I have to keep rebuying expensive parts like the charger or the housing. If it do it with electricity it is much more movable. I have 2 types of cars lined up to try it on to see how it effects each car. It isn't just focused on the RX8.

I understand I am wasting time, but like I said, I want to get a chance to understand these kind of problems and how to best solve them. One of the reasons I left pages of data out of the post is because I didn't expect anyone to understand, and asking for a reasonable debate based on facts I figured was impossible. Kudos for at least trying understand instead of saying "it won't work"



So here is the issues with the voltage:

15,000 Watts = 20HP Electric Motor @100% DC
15,000 Watts / 48 volts = 312.5 amps @100% DC
15,000 Watts / 96 Volts = 156.25 amps @100% DC

The 48v golf cart controllers have a max amperage where the controller will either stop sending amperage if it is a good one, or go pop because the load is to high. (If it is a cheap one) So when I am testing with a 48v controller I doubt I will be able to high 200/312.5 amps so 63% DC max of 100% available, if even. Also this is all at 100% EFF which won't happen either. There is just to many open variables to solves for the missing ones without trying to find them. Examples are how efficient the motor is, how efficient the compressor is, how much flow the modified housing can do, etc. It is easier just to get the assembly made and go from there.

@100%DC is irrelevant for what I am trying to do with the project. It isn't going to be switch operated to where it is on or off 0% or 100%. The engine won't need 100% of the power at all times, so I am created a control curve device to see how much to "need" in between. My guess is the power band I want, 2,000-5,000 it won't exceed 66%, so it will be more efficient than at 100%. I am seeking a level of control not available with a bleed based system because I don't want to bleed energy, I want to maximize the use. This is a core difference, but it eliminates parts and makes the system easier to control because a computer is determining the control of the boost.

Organizing this all in plain text is a disaster. When I wrote it all out, it took up 19 pages of graph paper and that was just the trying to "guess" at the flow rate of the parent supercharger the housing was made from. It was just a cluster fuck. I started to put it into Onenote and then figured out I cant really export that to something that works with Forum.

Still moving forward on this one.

Ok so I identified that the compressor I have is from a Thunderbird SuperCoupe. It is a great little compressor. The side has in inlet so I finalized how and where I want to mount it.

I hit a very serious problem I wasn't expecting.....My Compression went UP. Yes that is right UP. When I got the car it was at 7.0:1 and now it is pushing 8.0:1. I am obviously taking to good care of it, so I may need to run out the clock on the motor before putting this on a Series II RX8. I am going to try it out in a RX7 Series 1 or a small B16 motor then try it in the 8. Eventually the compression will dip to the point where I need a new engine, and at that point I will go ahead and throw it on. For now I am still mocking it up on my car.

Basically here is the plan.

Buy an AEM.Mazdaspeed Intake and study the exhaust Flow
When ready remove the filter element from the filter and use the MAF pipe for the Supercharger Build.
Adapt the Exhaust flange from the supercoupe to bolt to the AEM MAF Pipe.
Create a custom flange for the charge pipe that draws ram air from the front of the car like the AEM and RB kits.
Create a passive flow system for the supercharger drive.
Buy & program the motor controller.
Switch over setup to super capacitors instead of batteries.

At 12 volts this motor draws a little shy of 300 amps. I tested it with a 12 volt battery and it immedately started to whine and die out. After I added a 1Farad audio cap it did a bit better. I have a 350Farad supercap set I built to replace a battery in a SCCA car, and tested it with that. With all 6 caps in series it is about 58Farads. I expected it to act like the 2 Farad cap and run for about a half second without a battery, but I was VERY WRONG. It ran at full Speed all the way to 5 volts over 1 minute! That is 400 amps from 58 Farads and it out preformed a 12 volt battery and audio cap! I was very very shocked! So this renewed my faith in using these supercaps, and I learned I dont need the 80 dollar ones to do the job. The only challenge is creating a mounting design to hold them. I will prob need a 3D printer to do so.

Saved for Stage 1 Pictures.

Saved for Stage II Pictures

Added Methanol Injection Information.

I dont think 5PSI at the diameter of the intake at WOT is gonna be only 3.25hp. superchargers that puts out 7psi under WOT is gonna be around 70-100hp range to drive them. you are wishful thinking.
the thing I dont understand the whole electric supercharger thing is why driving it with electricity. why not just use say the engine like all the car manufactures been doing? converting energy is a huge waste. why not just drive the supercharger with a pulley, instead of using alternator to convert mechanical energy into electricity, then convert that into chemical energy, then convert it back into electricity, then convert it into mechanical energy?

1 hp = 0.746 W

Posted From RX8Club.com Android App

746w:yumyum:

Total waste of time an effort IMO.

If electric superchargers worked in any practical form, they would be using them on race and production cars. You would need a pretty large motor to have enough power to compress the air. You are not just blowing air into the engine. A fan wont cut it. So, big electric motor attached to a supercharger is what you are basically doing. Adding lots of weight over a light little pulley.

A lot of people fail to grasp the concept of a fan versus a compressor.

Why not make it a real challenge and just tie one arm behind your back while your at it ....

Another thing I just thought about. Where the hell are you going to find an electric motor that can turn fast enough to run a blower? I know centrifugal superchargers are basically turbo compressors with a pulley in place of the exhaust turbine. Those turn at speeds over 50,000 rpm. Some times a lot more. Not sure on roots style, but I imagine it's up there a ways. The only electric motors I have seen that are that fast are tiny things. Not ones big enough to run a compressor.

That is per liter, which technically the way the Renesis acts it would be more like 7-11HP would be required of the motor. So based on what I have heard, it would be 7 or 10hp for the Renesis at 6k rpm for 5 psi. Displacement for the Renesis isn't straight forward, so it is something to experiment on.

The reason electricity, is load bearing and spooling speed. It is faster than N2O, and can be controlled with a PWM signal instead of unmetered energy controls of a supercharger or turbocharger, aside from load.

Car manufacturers are starting to develop Hybrid Electric superchargers and have been for about 5 years. The issue they have is the dual rail charging systems & how to approve a second voltage standard for the car. Some manufactures are already working on that like GM. People want V8s, so they make V8s. Consumers drive the need for cars to be made a certain way, and if it is cheaper and nothing is stopping them, why would they stop?

Yeah, I wrote one thing on paper then blanked when I transcribed it. I used 750w for all my calculations, not 725.

Exactly what I am proving. The concept only works on little engines, anything bigger than 2 liters isn't worth the "pool" of electricity you need.

Every form of Forced Induction moves energy around. The assumption is that Supercharging and Turbocharging and the best way because no one else does anything else. There is very little to prove this concept could not be done besides Chinese people advertising leaf blowers are turbofans.

You don't do something like this to make power, you do something like this to expand and collect knowledge about how something works. Not everyone is looking to "go fast." If I wanted to do that I would just buy a bottle and be done with it. Can you say you bought an RX8 to go fast? There is always a cheaper way of making power, but I bought my 8 because I wanted a different engine platform.

A Transmission is technically built into a supercharger based on the side of the pully driving it. Because of the EV conversions out there, DC electric motors are everywhere, and cheaper because of it. I am not designing it, just buying it and tinkering with it.

I expect a lot of hate, and so does the guy helping me with the project. He deals with it all day, but the truth is, this is an engineering project for fun, and some for a potential benefit. I am documenting my results, because I want to share my results, negative or positive. If it isn't cost effective or possible, it cost you 15 minutes to read the thread and figure that out. If it is, then you just figured out something new to do that may make power for cheap. If it was already clean and documented, I wouldn't waste my money or time developing and documenting it.

If you want the truth, I get itchy just sitting around unless I am in the garage making something. That is just the way I am.


....FUCK, I guess RX8 doesn't add quick posts together like MSF. Sorry about that.... :(

It just seems to me you are doing a lot of work for nothing. why use electric motor when you can simply just drive off a pulley. why use electricity when you have mechanical force available. the only reason electric motor is good is in situations when no coupling is allowed meaning it will have to spin up from complete stop such as an electric car cuz you get instant torque from stop. you dont need to stop your supercharger. also there is no such thing as spooling up a supercharger since its driven off a pulley.
its not we are saying you are doing the impossible, and there is no hate going on. its just that you are doing something not necessary when there is good solution already available.
and how did you get the 6000rpm 5psi at 7-10hp calculation? why is there are rpm related to psi? dont you want 7 psi across the whole rpm? are you treating an engine as if its a sealed box? remember, engine sucks air. pressure is BS when there is no flow. my electric power washer makes 1700psi but at only 1/16 square inch area maybe less, when you fan out the water, the pressure drops significantly.
also the equivalent 4 stoke engine to a 13b renesis is a 4 cylinder with 2.6L displacement. if you cant figure that out you probably need to go back to the drawing board.
and have you seen the size and weight of a 70hp or even a 10hp electric motor? its massive. also what kind of voltage are you gonna run the electric motor at?? even at 48V new EV standard, you are pulling 153A with the 10hp electric motor. its over 1000A when using 70hp motor.

Most of these aren't my calculations, the formula was given to me as such. If it is wrong I can't find the right formula to fix it. I was told it was either 2.6 or 3.9, I knew it was not 1.3. The engineers I am talking to stated the 3 rotor faces make it a 3.9 and the other set said one 2 faces contain the compression and ignition stroke so they dont matter. Either way they stay within the 20hp tolerance for the motor so I said screw trying to figure out what side is right.

Even if it did work, you still added a couple hundred pounds of battery and electric motor to the car. Maybe more. That's never a good thing. That spoils everything about the car. Just makes the whole thing more pointless. I don't see any reason for doing this at all. Even if it works, the cons far out weight the pros.

its twice the displacement. the rotor turns once per eccentric shaft rotation makes one explosion vs 4 stoke with one explosion per two rotation.

here is your problem. open flow is different from induction. the pressure measured inside the intake manifold is the combination of the supercharger pressure minus the engine suction. a normally aspirated engine at wide open has a suction pressure equal to atmosphere pressure meaning it cant suck more than the atmosphere pressure.. so if you have a supercharger that puts out same pressure as the atmosphere pressure, it will be only enough to satisfy the normal operation of the engine. to have any power increase, you will need the supercharger to put out more pressure than the atmosphere pressure which is 14.696 psi. so if you want 5psi in the manifold, you will need a supercharger that will at least put out close to 20psi at open flow.

Considering that many hybrids use a gas motor to run the electrical generator should tell you something.

I'm still trying to imagine where you can fit a supercharger belted to a big honking motor in the engine bay. :scratchhe

Love your enthusiasm, I will follow with interest....

People are fucking idiots for buying Hybrids. An gasoline car is leagues more eff than any damn hybrid. People are sheep for going along with the idea that they work for the price you pay. If you ran the ICE on something besides gas for your Hybrid it would make sense.

You are describing a Electric car.

Most Hybrids use the Gas generator to turn the transmission same as the engine.

https://www.fueleconomy.gov/feg/hybrid_diag.gif


Electric cars use a Gasoline generator to provide power in addition to the batteries. The Gasoline generator never mechanically touches the drive shaft. That isn't a Hybrid, it is an electric car.

http://www.boosthead.com/images/prod...ge105-full.jpg

The displacement takes guesswork about of it thanks for that.

The answer to his lies in the validity of the formula. I didn't write the formula so I don't know if it included Atmo or not. The was rated on the Dyno as positive pressure, so I can only assume it did. I am going to check on that one.

it reads 5psi on the gauge but it actually is with respective to atmosphere pressure, which is about 15psi. you need to add your pressure on top of that to have more power.

I figured all would compensate for this as such. The good news is most of the testing for this was done at Sealevel so it is pretty accurate for my build since I am also at Sea Level.



Throwing an idea out there to use Supercaps instead of batteries. The Power is passed through a PWM Motor Controller anyway, so the caps might be all I need. They weigh very very little. You can replace your car battery with them and it weighs less than a pound. I would want two 72-96v banks, which would amount to about 10-15 pounds, but that is nothing. The battery budget is around 1k if even. The primary disadvantage is discharging over time, but that isn't really an issue for the application. Another big issue is our Alternators aren't normal alternators. They use Frequency to dictate charging rate. They aren't the kind to just up "upgraded" like 90% of the alternators on the market. I am not sure the effect the caps would have if you tried to replace the battery with them on our cars, so I don't recommend trying that. Just wanted to throw that in there.

The primary issues with standard lead acid batteries are the purposeful high resistance. Very little resistance would be dangerous for car electronics, so a small amount of resistance is added in the form of less conductive pathways in the battery. This is what determines "cranking amps." More conductive rods = more cranking amps but less battery life. My idea was to use some 16ah VERY low resistance batteries for this project to keep the weight minimal and power maximum. 100 pounds max was the estimate for all the batteries.

Some people think this resistance adding is a bad thing. Well if you do, but one of the lower resistance batteries and short it.(Don't do it, youtube it) Trust me, you will want that high resistance after you see a hydrogen explosion out of a battery.

I am not sure how many people have heard of the Phantom Electric supercharger, but it is been tested on the FRS. It is basically a Turbocharger with the exducer cut off and a R/C type ESC driven motor replaced in it's place. I think they are being a little generous with the efficiency.

FTS-TQ25024V - New site

This setup is basically 4KW to produce 2.5psi of boost. That is a real world adjacency I can use to see how mine will go. I applaud them for making it a kit, but for <70% Efficiency, that is kind of a waste. They would get a much better number if they changed the motor to a higher voltage, and switched compressor styles.

Ok I have been playing with the ESC device since it showed up, and I have learned one thing: This is to much for an RX8 to handle. That is exactly what I was aiming for, too much.

Total power is 18HP for the Electric Motor at peak voltage, but even at 12volts it had enough torque to try and escape the bench. I am having to postpone testing till I can build some kind of holding device.

I have been developing the Motor controller myself from almost scratch. My father chimed in and gave me a big push when it came to what transistors to use, but I have mostly been planning the power mapping out myself. I am still going to use SuperCapacitors, but I am going to use a unique way of gating the Boost converter (not boost controller), so it won't have more amps and volts than it needs at the given time it is operating.

The ideal Supercap is the Maxwell 3000 Farad in 2 banks of 30. (200 Farad Total.), but since they have become known to the world the price has gone up severely to almost 64 dollars a cap. This isn't ideal for a prototype, so I am going to use 12v 20ah batteries and some 350F caps to test with.

This is going to take a whole new kind of Dynotuning, so it will have to wait till I get the PC interface built for my custom power adder computer done. Right now it can only be programmed by a Basic Stamp and a USB cord with a computer attached.

Pictures are coming, but I want to control where they hit the internet, so they will be hosted inhouse.

I'll follow this thread, in houston too so i'd love to see it when i can. My only issue is mainly with my distrust with anything electrical and the known abundance of large amount of wasted energy in the rotary exhaust stream.

what kind of compressor you going to use?

It is a Root Type. Twin screw would be to unpredictable, and a centrifugal takes way to much rotation to do anything with it. Also the centrifugal takes way to much balancing of the inducer if you don't have the bearing assembly, and if you do, you have to worry about oil or the bearings for cooling and lubrication. There isn't any room for a transmission or drive ratio adapter, so your motor would have to spin several thousand RPM to get any non parasitic boost.


I tested the supercharger this weekend on 12 volts, (I will be running it at 84), and it moved pretty good at 12, enough to scare you. 24 is supposed to be it's minimum voltage so I didn't want to leave it at 12 for very long, plus it would be drawing a ton of amperage at 12 and cook the battery it was connected to.

For now I am going to get a 72 volt golf cart controller and 7 20ah 12 volt Sealed Lead Acid Batteries till I can find the motor's speed curve at 72 and what it will take to build positive boost and what it will take to move air at idle.

If you use a motor, you are independent of engine speed, unlike pulley. Massive advantage.

In my work I've seen watercooled permanent magnet motors running 7.5t@ 2.5m/s, that means 187kW continuous if im not wrong. It was actually not that big, but fitting it into an engine bay would be a challenge.

Problem I see is power delivery to motor from dynamo/accumulators, and weight/space.

Good news for all of that. At the end of the day it will almost break even.

I am removing the stock airbox, stock battery box, stock battery, and anything else in that area to bolt the supercharger up.

It weights about 50 pounds total, so that nearly breaks even.

The battery will be replaced with super-capacitors near the windshield wiper tank. eliminating some necessary wiring length, engine load, & space.

The supercaps weigh less than 2 pounds total for the board and the modules assembled. They are about the size of a small brick.


For the 72v/96v size I am using a create way to pulse DC-to-DC converters so they draw less total power per second than if you straight powered the motors.
so it does 12v Alternator (200amps) to Supercap bank #2 to DC-to-DC (round robin 1:6 Ratio) to motor controller(PWM), to motor.

So here is where I am at right now. I am testing it off the car, and working out the motor's calculations. The car is stock and still under warranty for now, so I want to wait and see the Engine's health before I start dynoing/testing. Right now the engine is getting worse and not starting well. I want to see if it will get to low enough compression to get replaced and then start this process. That will also give me a chance to strengthen some things such as the exhaust mani & coils.

Right now I am going to do 7PSI at 2500 RPM tapering down to 2PSI at 6000, and +/-0 pressure at 8000RPM.

Was that a 3 or 4 wire perm motor? If you make the contact points supercooled by water, the motor's fan and balance can be smaller because of the less space needed. Also that prevents the motor's dust from Atomizing.

One other question. Was the 7.5t, was the T a typo or a unit? I wasn't sure if you meant electric KW or power KW. Example 187,000watts (electric) or HP(as unit of power, not electricity)

There is a place in England that is making some amazing electric motors that are watercooled and are 4 wire controlled for extreme precision and regeneration control. You must buy the whole kit because of how customized it is, but that is ok. I have to find out what they were called again, but last I checked they were trying to make a "drop in kit" for an EV conversion.

good luck finding 187kw of electricity too. stock alternator is 1440w, so you need 129 alternators to continuously feed the supercharger, if you use less, think about batteries.
there is a reason why superchargers are driven off the engine because electricity is not the most efficient way of delivering power.
any why would you want independent of engine speed?

I am not saying its not doable. anything is doable given time and money. it all comes to weighting pros and cons.

I dont think he was insinuating that at all.

187KW is 200HPish or 90PSI of boost if you can get a compressor to compress that much. My guess is that is for an EV and he was just saying the technology has come a long way.

Independent of the engine speed means direct boost anywhere without any restrictions or need to take power from anything including the alternator.

-_-

-_-

Sub'd. I've seen something similar on the FRS/BRZ forums used with great success.

I actually called the owner for some information and at first he dismissed my idea, but after I explained it to him he was interested but a bit skeptical. I would say you could you that same model on the RX8, but it doesn't flow enough for anything past 6K RPM, where as the M90 will flow all the way to 9,500 RPM on a ported Renesis. It will draw a ton of power, but will keep going. I plan to just keep it under 6k right now to keep the power usage minimal and the torque temporary to where it is missing int he curve.

The same place I got my M90 version from was the original design for the BRZ model. He stopped using the Centrifugal type (Turbos, Superchargers, and ESCs) for the rotor type because the Centrifugal have to spin at XX,XXX RPM where as the M90 does about 2,000 RPM max. More torque is required, but the parts last longer and require no transmission to make the wheel spin as fast as required.

Super caps aint gonna cut it, at least not with the size/weight constraints you're talking about. I expect you will want the supercharger to run for >30 sec in a single pull, not <15. I don't have the math handy, but I was considering super-caps as a battery replacement. When I figured out there would only be enough juice to crank the car a couple of times I threw out the idea. Don't look at the total power available in the cap, look at the amount of power available from peak voltage to minimum voltage. As soon as you start discharging a cap voltage drops off, as opposed to a battery where voltage stays around whatever the cell chemistry produces until it's fully discharged.

Other than that I like the idea. Keep us posted.

Ok, just looked at several supercaps. Six of the Maxwell 3000F would make a decent battery replacement. It would be enough to run a supercharger for a little bit, but it would be an expensive prospect. The 350s would only give you one good crank, so forget about them.

Oh and any PWM pump controller for W/M injection is a joke. The sureflo pump will not provide variable flow, at least not in a linear fashion. Pressure will just bounce off the pressure switch setpoint even at reduced PWM and then will start dropping off FAST when the pump can't keep up with demand. Also flow through an nozzle is not linear with pressure so you will end up with 70% of the flow at 1/2 the pressure, or half the flow at 1/4 the pressure. Not good, not easy to tune around.

As the pump bounces off the high pressure switch, even without the PWM, it will cause flow out the nozzle to surge as pressure surges.

If you want to go that route use a water hammer arrestor as a mini accumulator and use solenoid valves and nozzles to stage your injection. That will give you predetermined amounts at predictable times.

Oh and if you're set on getting a PWM controller, just let me know I have one in my toolbox that you can have for next to nothing. (What it's worth.)

Based on your comment I think I left something out in between. I would be using 36 supercaps in series, (90 Volts 10 Farads) and possibly two sets of 36 (90 volts 20 Farads) in parallel. That would be the primary bank being charged by a round robin DC to DC boost converter set of 6 (200watts each running at a 1/6 duty cycle) that would work on a 1/6 distribution. At some point I am going to need a battery type and an isolation circuit to prevent from frying the alternator. I am trying to break it down in a format that I could post here, but I have noticed some of the math on the supercaps just isn't matching up. I don't want to look stupid by posting bad math. The Farad to Joules conversion isn't matching up in some of my tests because I think it is tailored to what you were saying about the battery peak to dead voltage. That motor will keep running all the way till about 5 volts before it burns up the traces on the board, so my voltage range would be from 90 to 5 volts which is huge. I took the calculations and formulas I found, and tested them with the exact load I put into the math and the load lasted 4 seconds longer than the math said it would, even with the amps on the meter being correct.

Something left unsaid on those caps is that at 2.5 they have more power than at 2.7 or 2.3. 2.7 is their peak, but they just don't work as well after 5 to 20 cycles at 2.7volt vs 2.5. All my math is using them to a max of 2.5 so it appears that, that is working in my favor.

The biggest problem with the 350s is their weird ass contacts shape. The negative is in the center and the positive is a half inch away. I think I would need a 3D printer to try and make something to hold them. A huge benefit is they are not marked up 200% like the 3500s because of Laserhacker's video. Even if you can fit them together you need a balancing circuit across every cell, so that is a difficult one to do with their weird ass shape. I thought about trying to make my own board with this as a template. http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=301224051229I bought one and tested that 350 kit. I started and ran that supercharger off it. I stopped because I didn't have a shunt to figure out the amperage. It would have melted my meter's amp capacity in a second.

Here are some links to make your life easier:
http://www.electronics2000.co.uk/cal...calculator.php
(I'd run it twice once for peak voltage and once for minimum voltage then take the difference. You do not want to run all the way to 5volts, unless you want huge motors/wiring. The current draw at 5volts is going to be HIGH. Hence the traces burning up, the motor would be next.)

Joules to watts (W) conversion calculator
Then convert.

It's going to take serveral KW to run the supercharger (5-10ish). Also the boost converters will be inefficient and running at 200w total, or maybe you meant 1200w total (the 1/6 duty cycle comment is a bit confusing), so it will still take a while to recharge your cap bank.

I'm not trying to discourage you, but just pointing out that the math has to work out on paper before you can hope anything else will.

Also don't miss what I wrote on PWM water injection pumps. I'm serious it doesn't work worth a darn.

I am using the 250 PSI Aquatec Pump, not the Surflow one. I would not trust my engine to a Surflow at any point. The Aquatec is what everyone is using because everyone has moved up to 200 PSI regulated micronozzles. I can a Setup on my Mazdaspeed and it let me make more power on the stock cat than anyone expected even without tuning. I used all Stainless Solid Lines and Swagelock fittings with a AEM Failsafe Gauge. The controller I used the DO progressive controlled worked great. The pump doesn't have enough load on it to warrant using more than a 10amp controller. Everything is handmade at DO, and I am skeptical to trust anything that could mess up an engine. If I do end up using it as secondary fueling, I would do a precompressor mix to make ti more vapor than atomized mist so it would be optimal to tune for. I prime my pump so it won't surge using solid tubing so it isn't a problem. If you mount everything with the right check valve, it is a non issue.

Those links have the same math bug I have been fighting for awhile now, the corrected Farad factor of the dielectric in the Supercaps. They don't act like cap, but act like batteries so the formula never matches up to what it is supposed to do. For example, if you have 3 caps in series at 3,000 farads each, it is actually 1000 farads total and does not work like amperage staying constant, yet the formula treats them like they are 1/3 the power captivity each. SO if you have a 2.5 volt battery at 1.2 amps and you put them in series you keep your 1.2 amps, but on that math, it treats it like you don't. The other problem is half those calculators don't say when you use Corrected Farads or when to use captivity of the capacitor's cell since they are assuming you aren't using them in series like a supercap bank. It is been really annoying trying to find out where the flaw in the formula was. After a notebook full of fact checking I did a test where I added a 2 amp load to the caps charged at exactly 12 volts, and the math said it would last 4 seconds less than it actually did taking into account the voltage drop. I tried to find the difference, but it was just not worth the day of fried brain cells to do so.


With the DC to DC converters, I would be using 6 to 10 of them and using them in a gated formula of a 1/6 duty cycle varying to a 1/2 duty cycle meaning the gates open for 10 out of 60 seconds in a round robin 1-2-3-4-5-6-7-8-9-10 section so that each gate opens for 1 second, charges a cap, then the DC to DC converts it, then moves to the next set. At full load it would increase the number running to 3 to 5 gated "sets" and have a maximum draw from the alternator of 1kw. Like I said I can't explain it without a diagram.

Also keep in mind the motor will be on a motor controller when running on the car, I just haven't forked out the money for one yet. The 72 volt ones are expensive.

Sureflo pumps are damn fine at what they do, and with the right diaphragms they handle W/M. I have the DO branded one and can't complain about the pump one bit.

Using all SS lines will prevent residual pressure in the lines which is good, but using PWM to control a pump will not give you linear progressive flow. Check it yourself a simple bench test with a measuring cup will tell you all you need to know about PWM on a pump. Very few things are handmade at DO, most are off the shelf parts re-branded, some are contracted out from china.

Surge is not about priming the pump. Surge is from hitting the pressure limit of the pump over and over again. For example it will be at 240psi and get that last squirt of water in then be at 270psi, then it will flow out the nozzle until 240psi again and then annother squirt of water to 270. The aqua-tech pump is going to have the same problem because of the internal pressure switch. Again, bench test. If you can see flow surging then your engine will see it too.


The AEM failsafe is a good idea, especially with the micro nozzle because they will clog.

I have no doubt you have used the same stuff to make lots of power before. You can make good power with a windshield washer pump and some mcmaster-carr nozzles. If you get the results you want then it's a win regardless, I'm just trying to share my experience.


The math on capacitors in series is 1/Total = 1/C1 +1/C2 +1/C3. It is no different for supercaps. If you have three 3000F capacitors in series then it is 1000F total. Just the nature of the beast. The actual energy stored adds straight across.

Just a guess, but to load test a circuit you need a resistor of some kind. A 2 amp motor will not draw 2 amps unloaded and also the amperage will not be constant as voltage changes.

1kw max alternator draw should be fine. Just expect that to be the bottleneck in your system. Run boosted for 20 seconds then recharge for a couple minutes.

I might be able to add something in regard to the W/I .
My setup gets to 16psi by 4500rpm and holds that to redline . I found 450 ccs of 75/25 W/M was at the limit of what the engine could digest without severe power loss . I now run 300ccs which starts around 12psi. and find no detectable power loss. I originally tried a pwm controller for a 2 stage setup but decided to bin that idea once i researched a bit more . Seems its a good idea to run a higher % W/M early on in the rev range which is what happens naturally if you keep flow constant.

As an aside : Cooling mist are wankers .
When my nozzle output didn't match the advertised output from their webpage I complained . I got an answer back that blew my mind ....." The rated output advertised was from when our pumps were running 150psi . Our latest pumps run at 200psi . "
They changed the pump psi YEARS ago yet left the nozzle outputs as is !
they then proceeded to ignore my request for a nozzle swap.
End rant.

Appreciate the feedback. Cooling mist looks to still be using the Shurflo pump Harlan is describing. The nozzle is the scary part. I went with DO's setup because of the microns on the nozzle. One thing I noticed is that those Aquatec pumps are legit. I was outside a very large yellow tractor manufacture because I worked for a parent company, and they were doing flow tests on them for a new kind of emissions standard. They copped the leads off and threw them away. For fun I took them home and they held exactly 200 PSI even after they had been beat on. One thing I did notice is that the fluid they ran was corrosive, so I never did use it on my methanol setup, but I used it to transfer fuel from a few carb bowls.

I was considering doing 50/50 or 25/75 Water/meth because it will get compressed inside the M90 and turn into a cooling vapor. If I insulate the intake tube it will operate like a air conditioner for the intake.

Yeah the Aquatec is much better about "bouncing off the pressure switch" than the Surflo version. The reason it does that is because it is designed for a higher flow, but reduced down a few sizes to increase the pressure without increasing the flow. I don't think the Aquatecs even have the pressure relief switch you are referring to. I have a tested setup from the Mazdaspeed 3 that I use for a few hobby like things. It still works great. I have never had it surge on my because of the check valve I used.

DO's stuff used to be made in China a few years back. That changed big time. The hose, and motor for the pump are still made in china like everything else, but the Nozzle isn't. That is the part that matters. I am not sure about the check valve, but I don't plan on using their's anyway. I plan on doubling the cracking pressure and going with one from Swagelock. Higher cracking pressure means less of a surge. I also had 20 feet of tubing, so the surge wasn't as much of an issue because enough pressure had to build to crack the check valve. I tested all this stuff extensively before I even put it on the car. I was being a pussy about putting it on my car before I did. I had about 5 months of happy trails on my car with that meth running in 100 degree Houston weather with a 70 degree intake. I logged every drive and made the failsafe log too.



That is the formula I did by hand, but it still doesn't add up. Either something is wrong with the specs of the caps, or the formula does not apply to supercaps. In one of the tests I did I used a perm motor from a old fuel pump. it was 2amps exactly. It had a blocking diode to prevent feedback. I get what your saying about math, but it just didn't add up, so I bought one to confirm that it didn't match. For instance the 350 Farad caps should not have enough energy to start the car 3 times, yet people test it, and even on Diesel starters it appears to work many times more than expected. Either they are more efficient than people are claiming the math is, or something is out of balance.

I think you missed one of my points. You can't test power draw with a motor. The free energy guys pull this scam all the time, they take a 2kw motor and then run it off a 12volt battery with a mystery circuit in between and say that they are producing 2kw for much longer than the battery could normally output. Just because it's a 2kw motor doesn't mean that whenever it's spinning it uses 2kw because an unloaded motor draws next to no current.

The same is true of starter motors, just because it's a 2kw starter that doesn't mean it's drawing exactly 2kw all the time when it's spinning.

The thing that is more efficient here is the motor not the capacitors and that's what's out of balance. If you want to get a clearer picture of current you have to use a shunt resistor and measure for yourself.