Alfa Romeo Forums banner

41 - 60 of 100 Posts

·
Registered
Joined
·
2,269 Posts
Hey, thanks for pointing that out Greg. I did read the legend upside/down. Thanks for making my point that much more succinct. That is not to say this is the unit that Chris is running. He didn't answer that question. I sure would love to have his torque curve (+ 50ft.lbs the sc brings on the whole curve)!
 

·
Registered
Joined
·
4,075 Posts
Hi ToonRboy, Chris is actually running an M90 supercharger. The M90's inlet flow chart looks just like the M62's chart you posted, just with larger flow numbers.

Don't misunderstand this chart, at an Eaton M90 will put far more air into an Alfa 3.0 at 10psi than it will at 5psi.

Greg
 

·
Registered
Joined
·
2,269 Posts
Hmmm, M90 seems a bit big for just 3Liters. For a target of 250hp and 250-270ft lbs of torque, I would think 350-400cfm should do.

BTW - I got that chart from RB Racing. http://www.rbracing-rsr.com/turbo/TurboMaps/M90flow.gif

I think the lesson here is that more isn't necessarily more. Everyone talks in terms of boost and as you said, the flow increases with rpm (at a given boost). However, the charts show that with these types of blowers, flow gets impeded by increase in boost pressure! The fact that it lost only 10-20 cfm is admirable, but by doubling the boost all you did was create heat. And that is my concern here when Chris is already starting with 9.5:1 static compression. At 5psi boost at ~100ft altitude ASL, your effective compression ratio would be ~13:1. At 8.5lbs boost, you'll be at ~15:1.

So I ask, would you even think of running pump gas if you had a NA engine with compression ratio's like this?
 

·
Registered
Joined
·
4,075 Posts
Ok, I think we got lost in the editing. I was editing my post before I saw your response, then you edited your response to my post. I never saw your original response.

Anyway...This is getting complex. To keep things organized I will address the effective compression ratio first.

There are actually a few ways to calculate effective compression, and the charts out there are not all in agreement with each other. I use Jim Steck's chart.

On Jim's chart a 9.5:1 engine with 5psi will have an effective compression ratio of about 10.8:1. That's been well proven to be a usable ratio on pump gas (other charts may show different effective ratios but that boost value and compression ratio are knows to work so the reference number is still a valid baseline). With 10psi this goes up to about 12.3:1 and this become a little questionable, but not impossible.

Greg
 

·
Registered
Joined
·
4,075 Posts
Regarding the airflow issue. I really think you are misunderstanding that chart. A lot of people do, and that's probably why Eaton has replaced this chart with more conventional compressor maps. I hope I am not coming off as being hostel or snooty, I am just trying to clear this up.

With an increase in boost pressure from 5psi to 10psi on an Alfa V6, air flow out of a M62 or M90 will INCREASE by about 25%. The increase in heat offsets a little of this, but not that much. More IS more in this case as it pertains to airflow. The airflow increase happens because in order to increase boost you increase the supercharger's speed. This chart doesn't clearly show that. It's not designed to.

Here are some approximate numbers. If 6000 supercharger rpm gives 5psi on a give engine, that engine will probably need about 7500 supercharger rpm at the same engine speed to get 10psi. Compare the 6000rpm 5psi flow numbers with the 7500rpm 10psi flow numbers for a more valid comparison of airflow.

Think of that chart this way, that it'is showing an increase in boost pressure resulting from a decrease in engine displacement! NOT from increasing boost pressure from the supercharger on a give engine.

Oh, and one more thing, flow out of ANY forced induction device gets impeded as the result of increasing boost pressure. With a positive displacement pump this is actually pretty easy to measure because it results in a decrease in volumetric efficiency. With a centrifugal pump, the effect is essentially the same, it's just tougher to measure because centrifugals are not rated in volumetric efficiency.

Greg
 

·
Registered
Joined
·
4,075 Posts
OK, I really want this to be clear, so I went through the trouble to plot the data on the M62 compressor map. The top line is for a 2.5 with 10psi at 6500rpm, the bottom is for a 2.5 with 5psi. At the bottom of the chart you can see the flow numbers. At a given engine rpm, they are not even close about 630 vs. 825 at redline. At 3250 rpm (half way to redline) it's about 420 vs. 530.

You can also see that the supercharger stays in about the same adiabatic range in either case, with the 10psi car having an adiabatic advantage at lower rpm.

I really hope this clears all this up...

Greg
 

Attachments

·
Registered
Joined
·
2,269 Posts
Well, confused is how I would describe the individual who goes through all the data and then creates a graph, which is inaccurate. This last compressor map essentially says the same thing as the simpler one – except that the Z-axis got turned around.

I know what I’ve submitted is sound advice. As this project moves forward, I hope that you’ll be there to provide your expertise & that Chris will post the results of his project and all the parameters – good or bad. Ultimately, it’s his baby.
--------------------
For the edification of non-engineer members (like me), I thought I’d break down this compressor map so that the big numbers don’t confuse. The Y-axis is atmospheric pressure and 2.0 represents 1-BAR or 14.695psi. The blue range indicates adiabatic efficiency (a measure of a compressors ability to compress air without adding a lot of heat in the process). M^3/hr is a measurement most laymen wouldn’t understand (X-axis). So I looked it up. Divide by 1.699 to get CFM.

OK. So
430 M^3/hr = 247cfm @5lbs boost, and
530 M^3/hr = 312cfm @10lbs (3250rpm). And
630 M^3/hr = 370cfm @5lbs boost and
825 M^3/hr = 485cfm @10 lbs boost (at engine redline/6500rpm). The RBR chart I posted displayed about 420cfm at the same rpm. Which sounds more realistic for that unit. Not worth squabbling over 60cfm at max boost @redline is it? IMO 485cfm should support just over 300 HP. Unless that’s a lot of hot air.

Obviously, the RBR table leaves some variables out of the equation and can be misleading. So, according to your graph, the faster you spin the compressor, the more boost. The more boost, the more flow. Adiabatic efficiency starts at 66%, but never gets worse than 54%. Thanks for clarifying and I hope I got that right.

Best regards,
 

·
Registered
Joined
·
4,075 Posts
Well, confused is how I would describe the individual who goes through all the data and then creates a graph, which is inaccurate.
What exactly is inaccurate in the graph I posted? It's the actual compressor Map from Eaton. The lines I plotted on it are taken from actual numbers at 5psi and 10psi with this supercharger installed on an Alfa V6.

The graph you posted from the RBR site is also a graph from Eaton, although it's at least ten years old. I think it's from the 3rd gen blower so it's numbers may be a little different.

Greg
 

·
Registered
Joined
·
400 Posts
Discussion Starter #49 (Edited)
Hi guys. Not been online recently, so have missed a lot of this...

Anyway, I'll refresh a few of the basics of the installation. As Greg says, I'm using an M90. It's an early one, and is not the ideal size for the boost/capacity, but it should work. I am running it at 1.6 x crank speed. Fuel, timing, water/meth injection are all fully controlled by the Emerald ECU, with the choice of 3 on-the-fly switchable maps. I intend using a float switch in the water/meth tank to invoke a very safe map in case it runs out while I'm driving. MAP sensor will be primary load sensor. Charge cooler is the core of a Jaguar unit with a VW charge cooler pump and the radiator from some sort of 1000cc superbike. I'll be using 6 water/meth nozzles and an HSV for control. Well, I will if the darn things ever clear UK customs.....

I will indeed report the results when I finally get to them. I hoping I can report success of course, but I will tell it like it is when it happens - good or bad. I will be able to post up various plots from the dyno session, so they should hopefully provide some useful data.
 

·
Registered
Joined
·
2,269 Posts
I wasn't referring to you Greg. Heck you're the foremost expert on this forum with this type of blower on these vehicles. I got no wind in my sails! I just find it interesting that the two graphs contradict each other with respect to flow at the given boost. As you suggest, it's probably different point-in-times of the M62's lifecycle.

-I agree that the compressor map shows better adabiatic (and flow) @10lbs at lower rpms.
-I would agree that 115cfm is a big deal (flow diff. at redline @5 vs 10lbs).

I haven't located the formula, but what would be the charge temperature difference at 68* ambient 73% relative humidity (London right now)?

Add heat soak from the engine compartment. All I'm saying to Chris is start at 5, and think about using an IAT sensor hooked-up to a gauge in the ****pit. Work from there.
 

·
Registered
Joined
·
400 Posts
Discussion Starter #51
Appreciate the input as always!

There is an IAT sensor at the inlet, downstream of the cooler, and this will be used as part of the mapping process, and as a safety net to catch very high charge temps. The ECU offers many ways to modify timing & fuel and if the car can do it, boost control. I'm still not sure how it's going to behave, and there are a number of ways I can control the water/methanol injection at this point. One modifier on the map I was going to control the WI with is IAT. I did consider trying to rig up ECU controlled boost, but couldn't think out a sensible way to do it. The blower bypass on idle/light throttle I can understand - there isn't much air flowing in the system, so the bypass can handle it. At WOT and full revs, that's a lot of air to dump isn't it?...I didn't see how the bypass could handle it. If it can, then I can do a quick re-jig and use the boost control map on the ECU to do just what it was designed for. This would be a brilliant way to do it, so I'm open to suggestions on whether the 40mm bike TB I'm using would be up to the job. An actuator isn't too hard to figure out. This would give me boost control from the dashboard and complete failsafe boost management. Downside would potentially be a lot of unnecessary heat going into the charge(?).

Like I say...I'm struggling a bit here - could easily have this all very wrong in my mind.
 

·
Registered
Joined
·
4,075 Posts
ToonR: I am in Total agreement with almost everything you have said, and you have provided some good input. Too few people out there bother to read anything and try and learn, and it's obvious you have put in some time on this subject.

I especially agree that 5psi is a good starting point and to work up from there. You don't want to start with 10psi and then blow a hole in a piston before you even get it tuned!

I am away from home and my list of formulas. However I the Eaton "Delta T" chart for the 4th gen M62 shows that in this case at 5psi and at 10psi the temp increases, PRE intercooler would be about 113F and 192F respectively.

Add in heat soak and a failure in the intercooling system and things can get ugly fast!

Greg
 

·
Registered
Joined
·
4,075 Posts
Chris, I know you have an M90, we just got sidetracked in our own conversation, sorry about that.

Anyway, back to your car.

A 40mm hole is plenty big as a bypass valve. For reference a Porsche 993 turbo has a 25mm valve, and it doesn't open very far. When that valve opens up with 10psi in the system, it discharges air with an amount of violence you wouldn't believe. Of course on a supercharged car it only does it for a fraction of a second because the when the throttle closes boost stops pretty quickly.

If you are going to rig up such a valve in an effort to have adjustable boost from the ****pit, I am sure a 40mm valve will do the job, but it will really have to be secure, and have secure plumbing back to the blower's intake side. You will have to dump the air overboard to prevent heat from going back into the blower and driving temps up. This will make it really loud. I rigged up such a valve on my Grey GTV6. I used a 15mm cable operated valve. When fully opened it knocked off just under 2psi for extra safety on really hot days. As for controlling this thing with an ECU, I have no idea about doing that. I like the cable control method.

Not a lot of people with superchargers go with ****pit operated boost control system. It's far more common and a little more efficient to change pulleys. I would definitely plan on using a lot of water in case the boost control system fails in the wrong direction.

Greg
 

·
Registered
Joined
·
3,771 Posts
Nice Ford Thunderbird SC! Junkyards in California are littered with them.
 

·
Registered
Joined
·
400 Posts
Discussion Starter #55
Greg, WRT using the bypass as boost control, what you are saying confirms my thoughts - it's not a great idea! Possibly the single most desirable feature of a turbo is the ease of controlling boost (within certain limits). Far more awkward with a blower - more a case of getting it right to start with. I do appreciate that changing pulleys is the way to go - I was just pondering out loud in case you said 'great idea'. Driving a bypass valve from the ECU would be simple enough - I was thinking about this last night - a modified IACV would work pretty well to actuate the butterfly and driving it with a PWM signal from the ECU is how the system controls boost on a turbo car anyway, so it's all ready to go. Shame the basic idea is flawed really.

Anyone know where I could find a larger than stock pulley for an old M90?? None of the aftermarket types seem to be available for it. I saw a nice quick change kit but it was based on a press-on centre and only suitable for the M112 apparently. Suppose I could always get something made....but I was hoping to avoid yet more costs....

Realistically, I think I'll get 7psi of fairly cool charge. I'm hoping that will be OK with all the precautions I'm taking. I'm relying on the ears of the RR operator to detect det, although having said that, I made a det monitor amp for a guy recently and have had a request for another, so might make one for myself!

Grant, that's where my M90 came from ;) It's old, but seems in good order. I replaced the drive coupler and the shaft seal. Bearings sound good.
 

·
Registered
Joined
·
4,075 Posts
Hi Chris,

I don't totally dislike the idea of adjustable boost from the ****pit. I just think it should be a simple cable controlled open or closed valve with the closed position offering normal boost and the open, low boost for really hot days, diagnosing running problems etc.

If you have the system set so it can run abnormally high boost, and it's computer controlled, I think there is an increased potential for disaster. For example, if boost spikes to 10psi AND the water injection system quits, I don't think it will hold up for long. So it's best to set the pulley size for max safe boost and use the valve to lower it when needed. With 7psi, an intercooler, good engine managment, and good water injection you will have multiple layers of safety. For example, if the water injection quits, it should be OK. Same if the mixture goes a little lean or if the intercooler pump quits. However two problems at the same time would be bad, so I like to keep everything as simple as possible. I suspect with 10psi and cast pistons, and single failure would prove costly. Having the ECU control fuel, spark, water injection, and keeping the boost at a safe level, is really putting a lot of eggs in one basket.

Pulleys for the M90 are not a problem. PM me the size you have and the size you want and I will let you know where to get it.

Greg
 

·
Registered
Joined
·
400 Posts
Discussion Starter #57
To be honest Greg, I was mostly musing out loud. I really can't see the current setup giving me more than 7psi. My thinking of using boost control was as you suggest, to provide a way of reducing boost from a nominal safe level in the event of a problem - say charge temps going high due to cooler sytem failure or just high ambient. I can make it safe in other ways using the facilities of the ECU to limit advance, and chuck in more fuel to try and cool things down if it all goes wrong. Quite agree that having the basics right is the way to go.

I'm quite happy about having the ECU doing everything though - it's got plenty of 'traps' to make sure things stay safe. I don't think I need to, but I can even fit an EGT sensor if I want.

I'll shout if I need a bigger pulley - thank you for the offer.

Still waiting for the nozzles etc to clear customs.....it's going to be slow I think.....
 

·
Registered
Joined
·
1,733 Posts
Brilliant :D!
I love the project and the personal involvement. Too many cheque book builders these days, if you know what I'm saying ;).

I agree that 1 system should be used to control the whole shooting match (where possible). Any aftermarket computer worth it's salt should be totally reliable and able to run what ever is programmed into it, repeatably and consistantly.
But I had to laugh when you stated that you were having problems tuning the idle with your 360cc injectors :p.
I do appriciate your patriatism by using the Emerald ECU, but down here, that sort of thing was delt with years ago by the Australian after market ECU producers (550cc and up with stable control of idle mixtures ;)).

Keep up the great work, inspirational stuff :cool:
 

·
Registered
Joined
·
400 Posts
Discussion Starter #59
Hi Duk,

I guess physics is different for you antipodeans ;) The issue I had originally was simply that the injectors were running pulse widths close to their rise time at idle...so very hard indeed to expect them to be repeatable when the error is equal to the data, IYSWIM. To be fair, it was OK...but not brilliant. Converting to sequential solved it. Given I was using injectors twice stock (almost), I reckon it wasn't that bad.

.....but you're dead right - I am loyal to Emerald, both as a product and a bunch of guys. Can't think of anything that would make me change brands. I'd love to visit Oz...but not to map the car :D

Still waiting for parts to arrive...waiting for a friend to TIG the mounting bosses for the water nozzles to the inlet runners...waiting for the cash to get it re-mapped.....waiting for enthusiasm to re-charge......probably going to be a long wait......
 

·
Registered
Joined
·
400 Posts
Discussion Starter #60
How time flies....

All the assembly work is now complete, just waiting for a vacancy at the rolling road now...and just possibly a custom mod to the ECU just for me! The designer is possibly going to amend an existing 'free' PWM table so it becomes a water/meth injection control map. This would greatly simplify mapping for sure as the plan is to input nozzle flow rate, then have the ECU calculate the required PWM duty cycle to achieve the desired water/fuel ratio called for by the map. Map is boost vs. revs. so would allow proper control of the ratio at all points of engine operation. All depends if he has the time and if he considers it worth the effort. If not, then it'll just be a case of noting fuel for a given cell, then setting the PWM %age to achieve the desired ration by hand. Not hard, but way, way better if the ECU does it on the fly.

Car is currently running with no blower belt. Feels somewhat strangled at higher revs/loads as you'd expect, but given I am using a 40mm blower bypass, it drives well enough to get me safely to the session. Figure a hole that big is good for about 140bhp, plus it can suck a bit more past the free-wheeling blower.

Must take a few pictures now....
 
41 - 60 of 100 Posts
Top