Alfa Romeo Forums banner

61 - 80 of 100 Posts

·
Registered
Joined
·
4,090 Posts
Nice!

Are those individual water injection nozzles on the runners?

Jes
 

·
Registered
Joined
·
4,075 Posts
They sure are! With 9.5:1 cast pistons, you can't be too careful. On this car I don't think there is any location post intercooler where a single nozzle could provide even distribution to all six cylinders.

This entire setup looks amazing. I really like the water injection parts, the one way check valve, and 6 way splitter/maniold look right at home.

Greg
 

·
Registered
Joined
·
400 Posts
Discussion Starter #64
Rolling road session booked for 5th November (hope there are no fireworks in the engine!).

Looking promising that the ECU man is going to be able to do a software mod so I will have a dedicated map of boost/revs with values for the water/fuel ratio in the cells, with the ECU working out the right PWM drive for the HSV on the fly. Certainly make mapping easier!

Jes, Greg is spot-on. There really wasn't any other way that didn't leave me worrying about uneven distribution. I did consider 2 or 3 in the plenum after the cooler core, but they'd have been facing the wrong way and no way to predict where the mist would go.


....wish me luck....
 

·
Registered
Joined
·
1,733 Posts
Impressive :D:D:D

Looks like you have everything covered to make bulk horsepower and stay in 1 piece :cool: Hopefully the same can be said for the transmission, especially when you have the mid-engine traction advantage (over a frontwheel drive):p
 

·
Registered
Joined
·
400 Posts
Discussion Starter #66
I'm fairly happy about the transmission at these power levels. (He says, being uncharacteristically optimistic!) A mate of mine had a twin-turbo V6 that was making around 400/400 and apart from a triple-plate clutch and an LSD, the rest of the chain was stock. It never broke a 'box, but he did break the odd drivesahft (our cars use Lancia Beta shafts).

Probably the main thing that might let me down is the clutch - it's a stock one, but again, I think (hope) it'll be OK, as I was involved with another twin-turbo 12V that, at the time of mapping, was using a stock clutch and made 350/350 - clutch was fine on the RR but didn't live long on track once it had 'enjoyed' a few race starts and had the Porsche 'box drip oil all over it.

The traction certainly is impressive though!
 

·
Registered
Joined
·
400 Posts
Discussion Starter #67
Well, that was a bust! Safe to say the WI won't be needed. The setup made a grand total of 5psi of boost at 5500rpm. Given that poor a result, we didn't bother optimising the map. At one point, I almost decided to pull it all off and go back to aspirated, but the guys there presuaded me to try one more time with a smaller blower pulley. We did a safe 'get you home' map which yielded a net loss of 1bhp over aspirated and about +10% peak torque. Again, have to emphasise this was was running very retarded timing so it could do better...but not that much by the look of it.

So, some questions mostly aimed at Greg!! Simple pulley ratio calculations with an allowance for cam overlap suggested about 7psi+ Clearly didn't get anywhere near this, so my main question is about the losses I saw. Are they a simple 'offset', a percentage, or more likely, a bit of both? Ignoring the overlap correction, the basic ratio said 8.8. I got 57% of that. I'm now thinking I need to spin the blower a 2x crank rather than 1.6 x crank to stand a chance of getting any worthwhile boost. Charge temps where low (never went over 50C, often saw high 30s+ aspirated), so I think I can rule out restrictions down-stream of the blower. Bypass was tight shut.

Boost did rise slightly with revs as I expected it would, lowest recorded was about 3psi @ 2000rpm, highest was about 5.25 @ 5600rpm.

Dangerous to try and draw too much from this, as the map was far from optimised, but peak torque and peak power had moved a long way down the rev range. Torque was at 4480, now at 3600, power was at 5600, now at 5300. If time and inclination had allowed, it would have been helpful to optimise it, but we'd all had enough by that point.

Spin it faster is the obvious conclusion, but how fast.

Other warts from the day, took me well over an hour to put the blower drive belt on. Last time I did it, it took about 30 minutes to swap the bits over (pulleys, brackets etc) but of course, I've added all that WI stuff since then and it's all in the way. Some good bruises on the arms today. To my huge embarrasment, we found that the throttle wasn't opening fully by quite a bit. After getting over the dismay of such a basic error, I was then optimistic we'd found out why it wasn't making much boost.....needn't have bothered! It made about an extra 1/4psi. Final insult was down to another of my bright ideas. The plug caps under the blower are a bit recessed for clearance, so I was worried about water pooling in there. I put some vaseline in there to help the seal....but of course, it gets hot, runny, then trickles down, hits a very hot spark plug, erupts back up then dribbles onto the exhasut making great clouds of smoke.

There, told you I'd report honestly!!
 

·
Registered
Joined
·
2,269 Posts
Before seeking more boost, I would take a hard look at all the other metrics. AFR's, ignition advance etc. According to the Eaton information (and our previous debate [Greg and I]) your M90 should have been flowing close to 400cfm at 5600 engine rpm (x1.6 pulley). This should be sufficient to support a minimum of 240hp. Something is wrong.
 

·
Registered
Joined
·
400 Posts
Discussion Starter #69
It's not wrong as such I don't think, as I said we ran a very quick and dirty, but safe, mapping session so I could drive it home as it was. No effort was made to extract best power. The plan was to creep up on a safe setting so we could do a run and see what boost it made. The map won't change the boost of course, so that was a sensible approach. If it had made enough to be worth the trouble, we'd have done a proper optimised map. Given only 5psi, none of us saw any point. By the time you take out some power for the 'poor' map, and the drive costs of the blower, it's about right. Also, I'm using a 3rd gen used blower, not one of the new ones, so there's some more losses to take into account. AFR was around 13:1 falling to about 12:1 at the top. Tad rich but not power-sapping. Timing very safe, deliberately so.

My main question now is how fast should I spin it, based on these results for this setup, to get some useful boost. I don't think there is anything fundamentally wrong that's worth changing, other than drive ratio. Brand new top-spec M62 would help of course, but that ain't going to happen!! It works as it is now (more or less) or I give up on it. if I can get 8psi or so then I'll give it another go. Otherwise, I'll not bother.
 

·
Registered
Joined
·
2,984 Posts
Just for the heck of it, did you see what the PSI was just out of the blower before any other bits?
if you have not done this it might be a good sanity check.
 

·
Registered
Joined
·
2,984 Posts
one other thing. check for rags in the IC. This may seem silly but it has happened before. where a rag was stuffed in there to keep out the dirt and forgot.
 

·
Registered
Joined
·
4,075 Posts
I don't think there is a serious problem here, I think there is a combination of smaller problems all of which can be easily solved.

OK, this one is going to be long.

First, let's talk about pulley drive ratios. In my book I have a list of required drive ratios for various supercharger and engine sizes. (I just realized I don't have them for the M90 on a 3.0). I used Magnuson's calculator to come up with those, and that's what I normally use. That formula is in my book, and if you look at it you will notice it DOES not account for the changes in volumetric efficiency throughout the blower's rpm range. It really only accounts for supercharger and engine displacement. I think that the formula assumes that you will be spinning the supercharger fast enough so that it gets into the area of high volumetric efficiency. I don't have a map for that blower, so let's use an MP62 as an example. At the blowers rpm limit with 5psi it has a volumetric efficiency of an amazing 98%. With a 1.6:1 drive ratio at 6000 engine rpm (9,600 blower rpm) it only has 88%. Now it's safe to say that if a formula predicted 5 psi at 98% v.e. and you only had 88% v.e. boost would be below 5 psi.

So I think that's a big part of the problem here, the supercharger just isn't operating in it's high volumetric efficiency range, and thus boost is low relative to the predicted number.

It's also important to understand that while v.e. goes up with supercharger rpm, it goes down with boost. This makes it very difficult to predict boost exactly because more blower speed increases boost and v.e., but the higher boost lowers the v.e. which lowers the boost.

Second, don't underestimate the ability of boost to escape through tiny leaks and lower the pressure in the plenum. It's easily possible you have 1 or 2 psi worth of loss due to leakage. Here are a few thoughts: Are you sure the supercharger discharge port is sealed to the supercharger? I can see there was welding on that port. If it wasn't milled flat after the welding, it's likely that it's slightly warped. The two silicone couplers between the blower and the intercooler have worm and roller clamps. There is almost no way those are not leaking. T-bolt clamps are almost essential here. The couplers on the plenum runners have clamps that look suspect as well.

This leakage issue is really under rated. I think it's because the turbo guys usually don't notice it. A turbo (within its limits) will increase its rpm until the desired boost is seen in the plenum. If the couplers on a turbo system have a lot of leakage, the turbo will have to spin faster than it should, which is bad, but the driver will never notice a loss of boost because boost in the plenum will appear normal. With a supercharger, any leak shows up as a loss of boost on the boost gauge.

Let's review where we are. You have 5.25 psi. Leakage is probably about 1.25 psi (maybe more, maybe less). Fixing that would get it up to 6.5, then figure it's about 11% low due to the v.e. issue, so that alone would make it more like 7.15. Now the difference between 7.15 and the 8.8 the formula shows you should have could easily be cams and intercooler.

The relative lack of power is another issue. I think a large part of the issue is ignition timing. These cars like a lot of advance. PM me your advance data, initial advance, max advance, and when in the rpm range does it reach max advance. I will PM some numbers back to you. I have found that 5 degrees or retardation can easily knock of 20rwhp or more, so this is a big issue.

If the timing was advanced and there was too much water, power could have been reduced dramatically. Even with 5psi you should have at least 25% more power than when normally aspirated, and probably a lot more.

I agree with ToonRboy that it's a good idea to make sure everything else is OK before cranking up the boost. Slyalfa's points are excellent. A blockage would cause this problem, and it happens.

Here are some MP90 4th gen numbers from the Magnuson calculator on a 3.0 liter. All numbers assume a 6 inch crank pulley:
3.55sc pulley=10psi, 3.29sc pulley=12psi, 2.95" pulley=15psi.

Greg
 

·
Registered
Joined
·
400 Posts
Discussion Starter #73
Greg, chaps,

Thank you for the thoughts & ideas. I'll certainly look at leakage as a possibility. Although, it produces the same manifold depression at idle as it always did, at the same rpm and with the same IACV setting. This led me to think that if it can hold -8psi, then it can hold +8 and so leakage wasn't a problem. I appreciate that positive pressure will work at the joints differently than negative, but would expect to see a problem holding vacuum if there was an issue?? Also, I was careful to check the flatness of the outlet (although it wasn't machined post-welding) and have used sealant on all bolt-up faces. All this said, I didn't actually look for leaks on the day. Absolutely certain there are no blockages - I was very careful about this during assembly.

I'm sure the main issue is that it's simply spinning too slowly and this is putting it way down at it's least efficient area. Greg - you always had reservations about using this blower size on this engine, at this boost level, for this very reason. Can't say you didn't warn me!

The low charge temps are also an indication, I think?, that there really isn't much compressing going on. All points to the pump not pumping much.

Power-wise, I'm really not concerned about the actual numbers at the moment - timing was backed way off just so we could run it up and test the boost. The final map was quick & dirty as I said, just so it's safe to drive. I was on the verge of pulling off the blower belt and going back to aspirated but was persuaded to persevere. None of us wanted to potentially waste any more time as it was, it was late, and I had to drive it home. Even if it was optimised, and we got an optimistic 30% more power, that's only another 60bhp or so. Take off the power to drive the blower, say 20bhp??, so +40bhp. Given all the extra weight I've bolted on, the net result would likely be marginal gain....which brought me to the thought that unless it can make and use 7-8psi or so, I don't think I'll persist.

..so back to the $64000 question (or almost £50000 at current exchange rates), how fast should I spin it. I'm leaning toward 1.9:1 based on what I've seen for this particular setup.


I will look for leaks, that's easy to do and might just be all that's basically wrong. After that, it's a smaller pulley and see what it does. Those Magnuson numbers look a lot like the basic volume ratio I used to come up with 1.6:1 (or rather, since that was what the pulleys turned out to give purely by chance, what I decided to try). Clearly I'm very low on those.


I should have taken a selection of pulleys with me, but it never crossed my mind that it would be so low on the prediction until very late in the day.

Again, sincere thanks for the thoughts....more anon once I've had a chance to have a look.
 

·
Registered
Joined
·
400 Posts
Discussion Starter #74
Just for the heck of it, did you see what the PSI was just out of the blower before any other bits?
if you have not done this it might be a good sanity check.
We didn't, but it did cross my mind it would be nice to see both boost & air temps at that point. I might consider putting monitoring points there if I end up taking parts off to fix leaks.

At least I can now drive the car with the blower, so logging this sort of info is now simple if I add the sensors. I can just use the ECUs own logging tool for the job. The map/ECU configuration is now such that it will manage more boost automatically (up to a point) as there is MAP correction of fuel and timing. Base map is very safe, so this should work fine as long as I am careful. Ultimately, if I can get 7-8psi, then I'll take it back and we'll do a proper power-optimised map, plus a safe one for when the water/meth runs out.
 

·
Registered
Joined
·
4,075 Posts
Just because something holds vacuum does not mean it will hold pressure. In fact it's quite common for some thing to seal under vacuum and leak like crazy under boost. Once you have it up to about 7-8 psi it will have a good 50% more power and you will never consider returning it to normally aspirated.

Greg
 

·
Registered
Joined
·
400 Posts
Discussion Starter #76
Thanks Greg. I do appreciate that +ve pressure is going to tend to open up joints where -ve will tend to close them, especially on something like a soft hose...just being naive I guess! I'll try a squirt of soapy water over all the joints and give it a quick blast to see if it blows any bubbles. Funny enough, the only place I thought might be an issue was the water injectors. They aren't absolutely tight so rely on the O ring somewhat. I half expected to have to tighten them a bit more. Think I'll do this anyway, then test.

If I'm honest, 50% was the sort of figure I'd consider a result. Much less and I'll question the value of it in my particular application.
 

·
Registered
Joined
·
4,075 Posts
If 50% is the minimum desired gain, more boost is the answer! I think your kit has the potential to make whatever amount of power you want to risk. My primary concern is the cast pistons, but if those can take it, cranking it up to 10psi will make is crazy fast.

Greg
 

·
Registered
Joined
·
400 Posts
Discussion Starter #78
Hi Greg,

I too am concerned about the pistons to be honest - it's why I went to all that bother and expense with the WI setup. Just hope they can take the strain.

Just taken the pulley off the blower to take measurements so I can get a smaller one made. While doing this, I discovered where some of the expected boost got to...much to my shame, I found that the bypass valve wasn't totally shut when it should be. Don't know how much of an impact this would have as it wasn't by much, maybe 0.25mm gap at top & bottom of the butterfly...but it won't be a good thing! Rather annoyed at myself really - it would have been so simple to triple-check this at the RR....but I thought it was already correct. Something must have moved since I originally set it up as I KNOW it was right before. Another vary basic error like the lack of full throttle. Just hope it's as insignificant, or I'll turn up with 12psi next time if I'm not careful!!

No signs of any 'proper' leaks on the bits I took apart as yet. Blower outlet was tight sealed all the way round.

Thinking about a 70mm/130mm pulley ratio at the moment.......I think that makes it +43% plus a bit more for better VE, less a bit for higher temps...so somewhere closer to 8psi perhaps now I found the 'leak'????? Starting to think a couple of pulleys might be a smart move. A 70 and a 75 to go with the stock 80. Shame it's such a pig to swap the pulley - have to remove the blower (or at least 2 of the bolts) and of course, this means removing the blower outlet and bypass along with it in order to get at one of the bolts. Might see if it's possible to do something about that while it's in bits again.

Wish I could get something right first time just once!
 

·
Registered
Joined
·
325 Posts
Chris,

I think you still may not be heeding Greg's advice well enough...I have alot of experience with the Eaton MP-62 supercharger on my race car (components and advice from Greg made it possible).

You really, really, really need to check for boost leaks. I bought a rubber 4 inch pipe cap at Home Depot and put a wheel air valve in center (will need aircraft clamps on both sides of valve). I then use a large clamp to secure to the supercharger intake, and use compressed air. This should get you 3 to 5 psi, depending on the capacity of your compressor. Then spray a water and soap solution in a small area and check for bubbles. Repeat anywhere there could be a leak. Even a very, very small leak will substantially reduce boost.

I had a small leak at the seal in a weld of my outlet port that barely made bubbles, but reduced boost by about a pound!

Also, engines with very good flowing exhausts will tend to have a bit less boost showing up, though the air/fuel mixture is flowing.

I'm currently running 7.5 pounds boost with 10 to 1 pistons, but use 112 octance race gas.

You are smart for doing water injection at all cylinders. I tried just 1 nozzle in the intake and ended up destroying a new engine!!!
 
61 - 80 of 100 Posts
Top