HFS-3 Eaton Charged 1600

[Caspergti]

Hi guys,

Ive been on this forum a while, and was part of the group buy of the HFS-3 system which i have basically just finished installing.

My engine is a Eaton supercharged 1600cc 4cyl engine, forged internals, Gapless rings, SS Valves etc etc, It runs with an FMIC, Running about 12-14psi of boost and currently has 206bhp at the wheels. It runs emerald management with EGT sensor and techedge wideband control, no knock sensing apart from egt. I bought the HFS-3 system to assist in lower intake temps and ignition timing advances.... But with a small amount of knowledge about the subject, this forum has helped me a great deal with understanding it. Just hoping a few of you could share your views and give me any help, as i will need it!

I have the system set up, but haven't decided on where to put the jet in yet, from what i see people try all sorts of things. I wanted to run the car with a 50/50 mix with running the jet in to the flow after the I/C but further enough away from the air temp sensor so the mix has enough space to fully mix and give a good reading for the sensor... Am i right in thinking this?

It the dawned on me that i do see fairly high intake temps sometimes, after 20mins or so on the rollers, i can see intake temps of about 40-45 degC on hard boost, this is mainly down to the proximity of the charger to the exhaust manifold, its well protected and has a water jacket on the charger, but it still gets warm, Id sat it operates at about 70-80 degC but im unsure on exact air exhaust temps.

Now would it be a good idea to inject a small amount of fluid in before the charger intake?? will the methanol affect the aluminium case of the charger and the aluminium I/C? Bearing in mind the injector sizes are only 380cc, so im looking at a fairly small jet anyway, 0.4-0.5mm to be precise. Would i actually see any benefit from the pre charger injection? and what size jet would be right for it as ive read that smaller jets clog up easier? and if i use this, will i be injecting too much? I will be using R/O water (Reverse Osmosis) and race methanol.

I have set it up so the failsafe and the switch on the dash swaps my maps over on the ECU, going between a W/I map and a normal map at the flick of the button, I will be using EGT to monitor the engine, and have wideband control to monitor the rest. It will be all based on my IDC, i dont plan on using any map sensor input, although i could do, is this advisable??

If anyone could give me an idea in how there eyes they would set it up, or have any information or want to give me any help then it would be greatly appreciated. I will be keeping this thread as updated as i can, its never been done before on this engine, in fact everything ive done to this car has never ever been attempted! So im well any truly in the dark on this one!!

If anyone is interested in the build thread of the car to see the work gone in to it already then please visit here: http://forums.clublupo.co.uk/index.php?showtopic=31082

I understand there's about 1320 posts, it gets interesting about 5 pages in to the thread, so if you get bored, i understand!

Thanks,

John

[Howerton Engineering]

Casper,

I have a 1600cc M45 engine running 18psi and very hot temps out of the SC. I spray right after the IC (100% meth) to give the longest mixing and cooling time. I have tried spraying before the SC, I really didn't see much change, as the mass of the case is so much that in normal driving it didn't seem to help. Maybe in a endurance race with a sustained spray you could lower the temps out of the SC.

Do try to put the jet as far away from the manifold as possible, right after the IC if possible. Water doesn't turn as easy as methanol, so the more time to evaporate and mix the better.

I've used this example before, on a 100F day my SC outlet temps at redline is 350F, and the temp in the manifold in the midrange is 60F climbing to 110F at redline.

[Caspergti]

Wow, 18psi, how did you achieve that??!!

Would you sugest I run more meth in the mix then? Say 60/40? I can get the jet to the exit of the I/c no problems so I will do that if you say it makes no difference to the charger if you inject there.

Am I right in my jet size calculation? 380cc injectors which will probably run to about 80% duty, so a 0.4 or 0.5mm jet after the I/c is the right size? 60/40 mix using about 20% mix to fuel.

Will I have enough resolution on the gauge or adjustment on it to get full gauge deflection due to my lower flow rates??

John

[Howerton Engineering]

Put a small enough pulley on to spin it to 19,000rpm and you can hit 18psi. But it takes a lot of meth to combat the heat.

I'd start with a .5mm jet. You will see some IAT cooling but some will be combustion cooling due to the water. If you want max IAT cooling the more meth you use the better.

I think the gauge should work fine, if you have a big issue once running it maybe possible to re-flash your board for better resolution for a single small jet.

[Caspergti]

I see, i wont be running the W/I all the time, for motorway cruising and pottering about i will be on a different map without W/I.

I spin my charger to 16500 at the moment, dont really want to go much further than that.

I will keep this updated as i move along with the tuning of the system, It shoudl be ready to test by the weekend i hope.

John

[Caspergti]

Well im pretty angry.
Go to start the system for the first time and power up...... Nothing.

Check all my connections, all fine, and correct as the manual states.

Wobble the power receptacle to the controller, and i get buzzing and an array of lights. Had a look at the back of the controller and the solder joints connecting the receptacle to the controller board are all a different colour to the rest, con checked them out and 3/4 of them are cold joints.

Surely if you pay nearly $800 for a full system they would test this first? Im now stuck with half an injection system, and a controller that is about as much use as a chocolate teapot!

Who do i speak with to get a replacement?

John

[Richard L]

I am not sure your comment about the receptacle/board are of different colours, please clarify. You can return the controller back to us so that we can examine the "cold joints" you memtioned. This is the first time I hear about this.

Who charged to $800 for the system? The system was $650 at group buy without the tank.

Send it back to:

ERL Ltd
Repairs department
Iroko House
Bolney Avenue
Peacehaven
East Sussex
BN10 8HF

[Richard L]

If you want use the system to run at a low flowrate, the controller has to be reflashed otherwise the resolution will suffer by upping the gain from "SC" trimmer on the gauge.

[Caspergti]

Inducing tanks and taxes... sorry if i sounded a bit abrupt earlier.

The connection i am talking about is the main power connection to the board, 4 Pin plug, black, white, purple and red wires. If i hold the connector at angles whilst the connection is made, i get a variety of different lights on the gauge and on the board, plus a buzzing noise from one of the relays, i take it one connection is not being made correctly on to the board. I will email you a picture of the board.

I searched the forum before posting to see if anyone else had had the issue, so i was a bit miffed to find out that mine seems to be the only one?

Richard, ill send you an email, but i havent had a reply to one i sent about a week ago so im wondering if i have the right address?

John

[Richard L]

The problem with soldering is the absence of lead (by law). lead-less solder is not very good but improving all the time. The main problem is the ability to flow into the hole of the circuit board properly.

Send the board back and I will reflow the joints and reflash it for a lower flow. Address is no my last post.

[Caspergti]

Will do that now. i will send it same day delivery if i can, id appreciate it if you could do the same when it is repaired.

I have a rolling road session booked and paid for on the 23/8/10. I hope you can repair it and get it back to me by then.

John

[Richard L]

Will do for sure! Do forget yo include the mailing address.

[Caspergti]

Thanks, it will be with you tomorrow before 12.

I have included the return address on it, also can you reflash the controller so it accepts the lower flow rates? i am only using a 0.5mm jet.

John

[Richard L]

I would use a bigger jet and reduce the flow electronically.

Richard

[Caspergti]

what would you suggest?

Im running 4 380cc injectors which will probably be used to about 80% duty using a 60/40 meth to water mix.

John

[Richard L]

A supcharged engine normally produces more heated air than a turbo due to bulk mass of the housing. At low throttle opening typically cruising at high speed, the temperature of SC housing is going rise rapidly due to lack of cold air passing through it, this is on top of the SC efficiency.

The moment you snap open the throttle, the first gust of charge air coming out of the SC is going to be very hot. This may set off knock. Once knock has started, it often continues.

I belive you need a 0.6mm nozzle minimum. As the HFS-3 is progressive it is unlikely to over inject. I will aslo send you a smaller flow FAV so that you will get a more linearly and progressive flow. The valve we supplied with the kit is capable of flowing 1600cc/min. Although you can use a restrictor but nothing like a custom valve that is tailored to your range.

We just return the entire FAV assembly back to me.

[Caspergti]

Thanks Richard,
My plan is now to run w/i all the time, and just have a second map in place for the failsafe.

So I'll use a 0.6 jet, and control the flow by adjusting the gain, I take it that it's also less likely to block the jet when using a larger orifice?

I've put my number in the box being returned to to you, if you could give me a call when you receive the box that would be greatly appreciated.

John

[Richard L]

0.5mm upwards s large enough for most filtered debris to go through.

I have received your HFS-3 today. I have returned the controller (reflashed) via 1st class post. I have also put in a lower flow FAV and a new molex power harness.

I examined the power connector and cannot found any bad soldering joint. I tested the engaged molex conenctor by pushing and pulling it in all directions, not once has it made a buzz or flicker of any leds. Not sure what happen to you. I took a picture of the soldered joint and you can tell me where the bad workmanship is... The joints are untouched nor tampered with.

When you recieve the unit back, please let me know if the probelm, repeats itself again.

http://www.aquamist.co.uk/forum/molex2.jpg

http://www.aquamist.co.uk/forum/molex1.jpg

[Caspergti]

Thanks for turning it around so fast.

I really cant understand what was going on, those ate the 4 joints that I thought looked suspect, one of them had a large dull area in the joint.

When I tested it I tested the power coming out of the connector and the power on the board, at the solder joint and that's where I saw my differences, unless it was the molex plug making connection when I spun the board over to test both sides of the joint? I'll replace the connector and loom aswell, see if it sorts out the issue.

Oh and I have a little question regarding the Ecu loom and the power +12v connection (red wire) earlier in the manual it says to use a switched 12v and later on it says to use the positive on the injector? I take it you need the positive on the injector as that's the only way it's going to see the duty. Just want to make sure so I don't damage anything.

[Richard L]

I have also sent you an extra molex harness (power in) just incase one of the pin backed out of the plug. Unfortunately I have not checked other pins and joints as I was very busy for the whole day yesterday, only have time to build a new valve and reflash the controller and did a general test and catch teh 4pm post round.

Anyway, I like to get to the bottom of this. If you have a solder iron, please reflow the suspect joints. Alternatively you can mark it can return the board to me and we do it on the same day return as before. If it is the molex harness, please return it back to me for closer examination.

The HFS-3 receives the main 12V power from the molex connector but it will not be fully connected to the controller until the internal relay is powered up by the grey connector's red wire. This HFS-3 creates it own switched (+) internally rather relies on the users pick the correct switched (+).

Molex harness: any 12V source but preferred a switched +12V so that the controller is totally powered down when the ignition key "off" the pre crank position.

Grey harness: the red wire on this harness powers up the internal relay so the the controller can receive full power from the molex. To ensure that the controller is "only" powered up at the pre-crank key position, the fuel injector (+) is the only place where it can be certain. This red wire only draws 20 mA or so. The other job of this wire is to reference the (+) supply voltage to the fuel injectors, make correction/adjustments to our IDC decoding circuitry's interpretation.

I hope you it make sense to you.

[Caspergti]

Ok, so the grey harness red wire, does it power a latching relay in the controller? As I know I don't have power there all the time. I assume it sees the priming pulse to latch the relay in to provide power to the controller through the molex connector?

Ive tested the positive connection at the injector and I don't have a constant 12v at pre crank position, as I would expect, that's why I ask if it's a latching relay.

[Richard L]

The red wire of the grey harness powers one side of the relay coil, the other side of the coil is grounded. The relay is not a latching type.

It is very unusual not to have 12V present at the injector(+) at the pre-crank position.
Some cars switchje doff the (+) rail of the fuel injector after 10 seconds or so if no cranking has taken place during this period.

You might have one of those cars.

I am not comfortable for you to power-up the relay other than the fuel injector positive. It can cause false reading on the IDC signal.

[Caspergti]

The car runs off an emerald K3 standalone Ecu not the vw item. It uses a common ground on the -ve and a pulsed 12v signal to energise the injector coil. I'm not sure if it has 12v at pre crank for a small ammount of time though, i don't think it does though.

When it turns up tomorrow I'll check it all out again, this is very strange. So that red wire has two functions? To see 12v at the injector and the to sense the pulse to work out the duty??

Thinking about it, I won't have 12v at the injector at pre crank as the fuel pump primes when the door opens, if 12v was applied then it would inject fuel.

How can we get round this??

[Caspergti]

Ok, so I just checked it out this morning, plugged at all back in for system initialisation, grey harness, red to injector +ve and green to injector -ve which seems to be always earthed. Turn the key to position 1 of 2 (pre crank) and on getting nothing at all, I'm assuming that's because I don't have any power at injector +ve at pre crank as expected.

Bit stuck as to what to do now.... I assume the way the the hfs-3 is set up is that it uses the -ve to sense the duty, as the +ve is constantly powered?

Richard, I'll give you a call later on today, it's better to talk about this over tue phone I think.

[Caspergti]

Looked in to it more this evening.

The controller is ok now, i have replaced the molex connector you supplied me, and will change the FAV tomorrow, and post it back, must have been something connector/loom related, all i know now that its now working. Thankyou for your efforts Richard.

Now the wiring, I have looked thoroughly at the wiring and tried out various different things. Firstly the emerald ecu is an earth switched ecu. But on the injector loom there is no constant 12v at pre crank, only a 2 second pulse. So the aquamist system comes on and then goes off.

If i wire the red wire from the grey harness to the ignition coil pack positive, it comes on as it should at pre crank. the only thing is how i can connect the -ve injector wire up, as the injectors seem to be grounded, the aquamist will just think its at 100% duty all the time.

The problem there seems to be is, the ecu is supposedly an earth switched circuit, but one side of the injector loom is earthed, and the other is connected to the ecu, so i cant see how the hell that works i dont know, but the cars been running like that for thousands of miles......

Any help would be appreciated!

[Richard L]

Ok, so I just checked it out this morning, plugged at all back in for system initialisation, grey harness, red to injector +ve and green to injector -ve which seems to be always earthed. Turn the key to position 1 of 2 (pre crank) and on getting nothing at all, I'm assuming that's because I don't have any power at injector +ve at pre crank as expected.

Bit stuck as to what to do now.... I assume the way the the hfs-3 is set up is that it uses the -ve to sense the duty, as the +ve is constantly powered?

Richard, I'll give you a call later on today, it's better to talk about this over tue phone I think.

I would like to deal with your earlier problem of a bad soldering joint of the unit, you have to push the molex connector at 45 degrees to make the system work. I rather not moving on until this problem is resolved.

I have checked your controller, examine your claim of bad soldering work on the molex connector. I have found none of your claims were creditable and was disappointed at your assessment of our product being high priced and without the quality of workmanship to match it.

Please confirm that the time we spent on "same day return" and an hour of my time is totally without justification? Some acknowledgment of our effort would be nice.

Yes I do have a solution for you if what you have described is true. I would be grateful if you can confirm with Emerald that the fuel injectors are all grounded and pulsed towards the +12V rail.

I await your reply on the Emerald...

[Caspergti]

I would like to deal with your earlier problem of a bad soldering joint of the unit, you have to push the molex connector at 45 degrees to make the system work. I rather not moving on until this problem is resolved.

I have checked your controller, examine your claim of bad soldering work on the molex connector. I have found none of your claims were creditable and was disappointed at your assessment of our product being high priced and without the quality of workmanship to match it.

Please confirm that the time we spent on "same day return" and an hour of my time is totally without justification? Some acknowledgment of our effort would be nice.

Yes I do have a solution for you if what you have described is true. I would be grateful if you can confirm with Emerald that the fuel injectors are all grounded and pulsed towards the +12V rail.

I await your reply on the Emerald...

I think we must have been posting at the same time, I tried to call you today to talk to you in person, but you weren't in at the time. Again thankyou for checking the controller and supplying the loom, it has fixed the issue.

Please read the last post i made, I have spoken with emerald and they have confirmed that the ecu is an earth switching item. but id like to know the idea of how to get around the 12v switching, so i can try it to see if it works.

[Richard L]

Looked in to it more this evening.
.
.
.

The problem there seems to be is, the ecu is supposedly an earth switched circuit, but one side of the injector loom is earthed, and the other is connected to the ecu, so i cant see how the hell that works i dont know, but the cars been running like that for thousands of miles......

Any help would be appreciated!


Lets move on...

According to your account of the ECU, it doesn't quite make sense.

Logically speaking, if the injectors are earth switched, the other side of the injector must be connected to a +12V source otherwise it will never energise when it is switched to earth.

Are you sure the highlighted text above is true? I cannot believe you have run thousands fo miles on that setup.

I will try to download the emerald user manual and take a look. Tell me against what the ECU model number is.

[Richard L]

Just downloaded the Emerald user manual, page 55 - wiring diagram. Is this the same ECU in your car?

[center]http://www.aquamist.co.uk/forum/M3DK-wiring.gif[center]

[Caspergti]

Lets move on...

According to your account of the ECU, it doesn't quite make sense.

Logically speaking, if the injectors are earth switched, the other side of the injector must be connected to a +12V source otherwise it will never energise when it is switched to earth.

Are you sure the highlighted text above is true? I cannot believe you have run thousands fo miles on that setup.

I will try to download the emerald user manual and take a look. Tell me against what the ECU model number is.

This is what is confusing me, i also cant understand how it logically works.

I am finding out exactly how the car is wired up at the moment, i did it with a friend of mine who is very knowldgable with the emerald unit, we did it about 2 years ago now, from what i can remember he wired up the injection circut. The ECU should be wired up as the picture you have posted, It is a K3 ecu unit.

This is what is confusing me, If i do a continuity check between one pin on the injector plug to the block of the car (obviously earthed) then i get continuity, which indicates that that wire is earthed yes? the other side if the injector is connected to the ecu, continuity to one pin on the ecu connector.

The only way i can understand whats going on is that at the off position and pre crank, the injector circuit is earthed mabe for protection, and then upon turning the key to cranking the engine it switches to 12v, This is something im going to confirm tonight.

I will let you know what i find out

[Richard L]

After our phone conversation yesterday, I think the only solution is to rewire the supply to your custom ECU to power up at the pre-crank position.

As the wiring situation at present, you give little time for the third party ECU to power up during the cranking stage. It is a win/win situation to let teh system prime and ready at the pre-crank positon.

In turn the HFS-3 will work as designed.

[Caspergti]

All done, and it is now working how it should be apart from one thing.

I have run out of trim on the SC pot on the gauge, at full power (currently trying to set it up using the "sys" and "inj" jumpers in for full flow) i can only get 4-5 bars with the gauge pot turned fully clockwise.

When the pump is activated (spool up) i see 7-8 bars which then drops to a constant 4-5 bars, this is after about 4-5 seconds at full flow. I am currently using a 0.6mm jet.

Is there any way for me to get the constant flow reading up a little bit? Or will i have to adjust the WH and WL trim pots to suit that?

John

[Richard L]

Can you set the gain on the controller just before the red led (>95%) comes on. Theris no advantage for the red led to come on.

I need to make a smaller flow sensor map for you. I have never seen the HFS-3 use on a car with such a small delivery. I will create a new map for you and please return the controller for further reflash. By the time the controller arrives, the new map will be ready.

[Caspergti]

Richard, the car is on the rollers today for set up.

Can we set the car up now as it is and then return the controller for the map then? I'll adjust the gain so that at WOT with max duty I see just below the 95% light and then I will report back later today hopefully.

John

[Richard L]

Anytime. It will not alter your actural flow. Let me know the results. I will also suggest trying 100% methanol.

[Caspergti]

Ok well we are getting there with it now.

Got more time on the rollers luckily tomorrow, had today to mess about a bit more with it. We are having a few problems with the car at the moment, seems to be missing some power, nothing W/I related though.

With 100% water, at peak power the air temp drop was from 47c at full boost down to 29c which is good, ambient temp was 22c today, tomorrow I will be trying the 60:40 methanol water mix. Hoping to see a bit more drop in temperature. At full power, egt at the headers were 690c so it's still very cool, alot more advance can be used I think.

The system so far is working very well, for me I think it will be best to inject water from a low injector duration possibly 25% as this keeps temperatures down going in to the engine. After playing with the gain, we have the 5-6 bars at full flow peaking at 7 on spool up and adjusted the wh pot to suit the failsafe configuration.

There is alot to learn with this kit, and the set up on my car, i am contemplating running a 0.5 jet in at charger intake to see what happens.

I'll keep this updated, its been a frustrating day!

John

[Richard L]

When things are not going according to plan, you8 learn a lot more from the experience. EGT seemed to be a bit too low for power.

Please continnue to update.