Project-M45 super charger (developments)

[d6dph]

Will the screen wash tank be ok with the pressure of the charge cooler coolant? What happens when it gets hot and expands, will it not blow the cap off?

Sorry not trying to be negative, I'm in awe of the skill that's going into this. just mentioning something that may cause an issue.

[stuartgallafant]

dave, AFAIK, matt is only using a low pressure pump to circulate this coolant, so i cant miagine that there'll be pressure issues...

[appletree]

as regards the inlet link and pressure relise, due to it being a supercharger and being driven all the time it only makes X psi boost at X rpm, the idea is you do lots of maths and work out what ratio pully you need to run to make the boost you want were yo want (top of th revs) .

which i dont realy know as i want the most the engine can take before det/pinking occurs so ive done lots of maths and spec'd it a little high (more boost than i might need) so this means that the boost pressure might only reach enough to lift the valve at say 6000rpm so it will only resurculate for 250rpm before i reach the red line and change up. but yes you are right if i hit max boost at 3000RPM then i would end up heating up the charge to much.

as for the tank the resivour its self wont be under pressure only the pipe work after the pump will be under pressure with a return in to the resivour, i might put alittle breathe nipple or somthing in the cap aswell to compensate for the heat change.

[d6dph]

I'm just thinking if the coolant absorbs sufficient heat, be it from the intake charge of from the engine bay, it may cause issues?

I'll go back to playing with my lego, I understand that stuff

[appletree]

the collent tank will be in the usual place behind the headlight so it will/....should have a cool air supply round it, the pump also pumps the coolent through front mounted radiator that will take out alot of the heat

The coolent will heat up as it starts to cool the charge but as i said in a much earlier post the car will only be on max boost for a short period of time followed by a longer period of time at little/no boost so the heat that built up in the short burst and then will have time to be removed when off boost, so it should equel its self out and stay at a pritty constant temp unless i'm on a track day or its realy hot out.

[Ant]

I'm just thinking if the coolant absorbs sufficient heat, be it from the intake charge of from the engine bay, it may cause issues?

Dave, I'm impressed dude damn good question fella

given enough headroom to expand, and a vent hole the tank will work fine as Matt has planned it, hence the term Headertank

[stuartgallafant]

matt, what kind of sums have you done to work out what pressure you will be getting where and how much, and so on...

what size pulley you running and also, what compression ratio?

[Ant]

@ Stuart

Peak boost after peak N/A torque is best, any earlier and Det could be a pain, but can be controlled with timing tweaks, the end result is less optimal ign timing and a power dip or worse....

there's no reason why 10psi on a 9.5:1 CR mill would be an issue, provided charge temps are kept down.

Heat = Det
Excess boost in the wrong place = Det
Excessive AFR's = Det
Lean AFR= Det
High EGT= Det

See a trend

[appletree]

theres about 4 main somes you hve to do taking in to account desired bhp, current bhp, CR, engine efficency, supercharger efficency, intercooling efficency, exhust gas temp.

theres a realy good book am using amongst others thats called supercharged! very very good worked examples of sums all using a M44 form a z3 as a model!! so almost ale the same figures, the design and installation. the bloke is using a twin screw type charger and looks pritty good.

the book

http://www.amazon.com/gp/reader/0837601 ... eader-link

[d6dph]

Ant, Am i right in thinking a lower boost setting can sometimes be of benefit, if the boost is too high, the charge temps rise due to the pressure stacking up in the inlet pipes? The compression causes extra heat and a less dense charge?

From what I can make out, you have to strike a balance between boost + sufficient charge cooling to make good power and stay reliable.

[appletree]

d6dph you totaly right you have to match evert thing up and have the right size components or the system wont work

[d6dph]

I am picking it up slowly!

[appletree]

matt, what kind of sums have you done to work out what pressure you will be getting where and how much, and so on...

what size pulley you running and also, what compression ratio?

The pulley works out about 2:1 of the the supercharger pully size, which i think worked out should give 9 psi at the charger and about 8.2psi after pressure dropof the ICing

CR is still standard 10:1 which seams to be about the max to start charging things with out IC

[stuartgallafant]

so you'll be running about 8 psi with 10:1 compression ratio?! isnt that a little high? i mean, i dont know, that just sounds high!!!

is the pulley on the charger the standarxd size pulley? also, will the one on the crank be twice as big or half as small?!

[jkarran]

Looking nice Matt,

The coolent arrangement should be spot on if I understand the plumbing: Cold from tank -> pump -> IC -> radiator -> chilled water back to tank. As you say, just vent it to atmosphere and it'll be fine.

How are you proposing to provide a reference pressure for the waste gate actuator? Most pressure regulators I can think of require some minimum pressure drop accross them to regulate. Likewise most designs only flow in one direction (from unregulated to regulated port) so some sort of bleed valve will be needed to avoid over pressurising the top port (no regulator is perfect, even good two stage ones fluctuate and spike).

Does the wastgate have an adjustable preload spring? I might be miss understanding but I'd have thought the best bet would be to set the preload spring so that the valve dumps boost when the bottom port (connected to SC outlet/plenum) reaches XPsi with the top port open to atmosphere.

Now, if you connect the top port to the SC inlet (behind the throttle butterfly) the valve will lift quicker when you snap the throttle shut. My brain's a little fuzzy tonight but I think that should give you a reasonable low load bypass as well.

Of course all that assumes you have adjustable spring loading I'll give this some more thought.

As described it's very similar to mine (dump valve is spring loaded albeit lightly, bottom port/piston is open to boost pressure and the top port is connected to the MAP hoses whioch tap into the TBs behind the butterflies) although I have downstream butterflies.

Give it some thought, I could be totally wrong

There's no progress on mine other than thinking about the management issues, I've been away/too lazy. I'm currently thinking: extra fuel rail, MegaSquirt and spark running distributor/crank trigger wheel spark code and a home made, logic level distributor feeding 4 home made coil drivers, this will take timing signals from the MS coil driver logic level output and the cam pos signal. Should neatly and simply enough get around the wasted spark/exploding manifold problem plus it'll run at light load (or without the plenum - NA) off the bike ecu fueling, all I'll have to get working initially is the sparks then I can start mapping the >atm pressure MAP fuel sites (plenum and extra injectors fitted) which should simplify the setup somewhat I may however have to electronically limit the MAP signal the bike ECU sees to avoid it throwing a hissy fit and assuming a sensor fault Likewise I will need to fit spoof coils to the standard ignition system, might just get awway with a bank of high power resistors, depends how smart the ecu is, my bet is 'very'

jk

[Turbo-Brown]

Am I missing something here, or should you not be seeing pretty much the same boost everywhere that the throttle's open with a positive displacement supercharger?

And then to prevent excessive loading of the belt when the throttle's shut, you have your valve open so that the SC is bypassed.

Would it not be a bit counterproductive to vent boost that's already cost you power to make?

I probably am missing something there, will have to have a proper read later on

[jkarran]

Without a bypass, boost should be constant at all rpm, in reality, there's a fairly constant leakage back through the SC so boost rises with rpm as that leakage becomes smaller in relation to the amount of air being pumped in/unit time.

I don't think you are missing much, you're right about the SC needing a bypass when the throttle closes if you have downstream throttle plates (it'd stall or burst the plumbing/bend the throttles otherwise), with an upstream throttle it'll just evacuate the inlet side behind the throttle. Whilst that will knock efficency it shouldn't be destructive. It'd still be worth bypassing though to reduce losses/stress.

I think the plan to control boost with a wastegate is for easy adjustability during development if I understand it correctly. Longer term it'd make sense to run the correct pully ratio and not vent at peak boost. Personally I'm just planning to cut a new pully once mine's up and running if I want more boost (unlikely... yeah right ).

jk

[appletree]

Thats right the valve is more for development and is only for pressure releise at peak boost, as i dont realy wont to make 4 different sized pullies to take to the R/R to find the correct pressure.

I am also running a seperate bypass valve (standard Dump valve) to recirculate boost when the throttles closed to prevent the charger from being loaded when it doesnt need to be, you can't truly "bypass" the chager when the throttles closed or only cracked open unless you have a clutched pully like the ones on the mercs, using the dump valve should mean that some of the S/C boosted air is recirculated so its not trying to suck against a closed throttle and sort of freewheeling in air thats nearer equel pressure on both the inlet and the outlet side than if there where no valve.

JK i was looking at using a in-line pressure regulater like the ones you get on a compressor but alot smaller to regulate the pressure on the top of the waste gate, it wouldn't rely mater if there was a non-return in the line aslong as the pressure didnt creep above the desired pressure and make the valve vent at a higher boost pressure. i could use a bolt with a cup pushing against the spring thats already inthere just as easily to be honest i just liked the idea of the pressure reg:D

also to mount the valve the correct way so the boost is pushing against the bottom of the valve is a little more difficult .

al have alook tomorrow and see

[jkarran]

Thinking about it, I have the boost on the dump valve side port (like Porsche/Lancia I think) rather than on the bottom port so mne is purely vaccum opperated and does not provide any backfire protection. Wasn't the initial plan but that's the only way it fits

I might make up a replacable bursting disk from an offcut of the stock I got for the pully and a beer can Or maybe a spring loaded lifting disk, might be less annoying (and require less beer consumption ) whilst I get the set up right.

jk

[appletree]

JK what type of dump valve have you got ??? is it like a HKS/MR2 style with the flange?? a normal baileys/forge/VAG bosch recirc type can be put in either way as the connections are the same on either side.

Am not to sure if you realy can do that much to combat the relise of pressure after a backfire anyway, a book i was reading suggested a good size is 2 square inches Per liter which is nearly 4" on mine which is a pritty big valve!! anyway this waste gate should offer abit of protection against it which has go to be better than nothing right

[jkarran]

Yeah, got a Bailey recirculating (Bosch type) valve. I stripped it ( Don't try it, it's a slippery git to put back together ) to see how it works, it'll work either way but if the boost is on the bottom port it'll help lift the piston, on the side port it wont. It wont physically fit the other way though it'd be a simple mod to make it work.

Backfire on mine would be bad, it'd blow the whole bonnet off if the plenum pops off the engine Better do something about that then!

Gotta get the E30 rally turd running tonight, hope I can, I haven't been offroad in ages

jk

[stuartgallafant]

this is all sounding really complicated now!!!!

[appletree]

little more done today, moved the top radiator hose out the way and re-routed the coolent supply for the head to keep it all nice and tidy and make it easyer to fit the S/C inpalce.

few pics

before


After


and a few of the top hose mod





the top hose being move out of the way means that theres plenty of room for the S/C belt/tensioner. I'll take some more pics of the supercharger in place so you can see how it all looks whne its light again.

soz the pics are abit dark only just come in from playing about.

[stuartgallafant]

matt, the 'after' doesnt show... what did the top hose look like before, i cant picture it?!

[appletree]

its like a "L" shape that goes streight out from the fom thr rad and then 90 dergees towards the timing case

i'll take sum pics in the light and all will become clear

[stuartgallafant]

yeah, i remember... so you only modded it for clearance for the charger then.

[E30_BO1]

I have just spend 1 1/2 hrs reading through all of this and all I can say is I want one!

[appletree]

So i've just got to work out how to put all these bloody bits together lol

any way had a trial on a spare bit of alloy bar at making some laminover so al post up the results later

[Andy335Touring]

I've got a load of clay left over from making the mold for my G/F elbow, any one need some ?

[jkarran]

any way had a trial on a spare bit of alloy bar at making some laminover so al post up the results later

Come on man... this I want to see
jk

[appletree]

calm down Jk here you go!!

this is just a little practice me nmy dad had on a bit of bar to see what size of fins we could make and what was the best method of making the groves.


screw cutting is on the left and then just using a power feed on the cross slide is towards the right.



The limiting factor is going to be getting a parting tool that is thin enough, on the laminova web site they say the fins are 0.2mm wide with a gap of 0.3mm between them so i cant se me being abl to reprouce those!!! but i'll have ago, secondly laminova tubes the fins are only very shallow and mine will be 9mm deep on the flat of the square and nearer 15mm deep from the corner of the square.

theres abit about half way through the right hand side which is is pritty good, machined alot nicer than the screw cutting and gave tha same results gaps and fins seem to be around 1mm wide.


think we might be making some sort of guide to put in the rear tool post tsoyou dont have to index the tool along but instead you can drop a guide into the last grove you made.

any one got any thoughts or suggestions??

[jkarran]

If you're cutting them 9mm deep the screw thread really isn't an option anyway, it'd be a nightmare to cut

Have you thought about using something like a ground hacksaw blade offcut to make a very thin parting tool, maybe braze it onto some scrap square bar to hold it nice and square with no surplus unsupported overhang. It might be rubbish but if it works it'd give you a very nice thin cut and it's an essentially free experiment.

You could probably also go significantly thinner on the fin without any problem, down to 0.5-0.6mm they'd still be plenty strong enough.

Are you going to support the middle of the bar while cutting to stop it chattering, it'd be a real shame if you got half way along and the tool dug in?

jk

[appletree]

Mats dad here
With the laminovas being made out of square bar it would be very difficult to support the bar in any way that will stop deflection when "tooling" in. I have considered a thin tool like a hacksaw blade, but fear a disaster should the tool dig in. Currently I am airing towards a slitting saw driven by a small gearbox I have kicking about somewhere, mounted on the rear of the cross slide. If I drive the lathe very very slowly (probably much less than 1 rpm) by an independent drive, or by the 3phase motor/thyristor drive (i've had one ready to fit for some time) I think we should be able to achieve narrow groves relatively painlessly. Might even be able to use the thread method in one pass.
Having looked into Laminovas some more, I think some cleaver baffles will be required to ensure the air flows over all of the fins and does not take any short cuts. also thermally speaking the fins may require segmenting radially (to provide a thermal barrier) so that the heat does not travel from the hot supercharger side via the fins themselves to the cold side and actually heat the air as it travels towards the cylinder head.

[appletree]

Try putting Laminova in Google images ignore the ones which are formica samples lol
Also try this link some interesting work on a Turbo E21 http://www.bmwpower.lv/my.php?op=galler ... e&pid=1927 may take a while to download loads of pictures foreign text

[WillG]

so you'll be running about 8 psi with 10:1 compression ratio?! isnt that a little high? i mean, i dont know, that just sounds high!!!

it depends on a few other things most importantly rpm, so for example 8 psi at 2000 rpm on 10:1 might be abit high but at 4000 rpm it wouldnt be a problem. thats how people can boost 2bar and stock compression cos they control the high boost to come in really late where its safe to use