Well, my little project is starting to take shape...
My good friend Scott McRoberts (McRoberts Machine Shop, Longmont (http://www.mcrobertsmachine.com/)) helps me with machine work...
ScottMcRoberts-MachineShop.jpg
Throttle body aimed rearward for cowl induction / cold air induction.
Plenum-Base-and-Fuel-Rail-1.jpg
I'm reworking the Rover fuel rail, and I plan to have it nickel plated.
Plenum-Base-and-Fuel-Rail-2.jpg
Wraparound plenum wall will be TIG welded to base. There will be a flat, removable lid...
Plenum-Wraparound-Wall.jpg
Now I have some details to work out... and I wondered if you guys might have tips.
I need to add ports for PCV, vacuum advance, and the vacuum sense line on the fuel pressure regulator... There isn't much room for threaded-in barbed hose nipples... should I use some sort of press-in, weld-in, or glue-in vacuum hose ports, and if so is there a good source for them?
What's the right physical location for the vacuum advance port? On the throttle body?
A question for people who have installed Rover EFI manifolds on Buick engines: what did you do about the thermostat bypass hose? Rover flipped it from left-side of the thermostat to the right-side. (Third photo.) Did you cut and weld, so you could use a straight hose? Source an S-shaped hose?
Curtis, Nice work. Scott and the camera person, "both" need safety glasses. Cheers, roverman.
Put one or two 3/8"npt holes and as many 1/4"npt holes as you have room for in the base plate and a few 1/8" as well. Better to have and not need them, you can get hex socket plugs to fill the spares. If you need to put any through the side wall, tig an internal boss for it.
Consider splitting your fuel rail in the front. It's a real pain to install and remove as one piece.
Jim
"Put one or two 3/8"npt holes and as many 1/4"npt holes as you have room for in the base plate and a few 1/8" as well..."
Alas, so very little room to work with, both in terms of surface area and also height!
VacuumPorts.jpg
Yep that can be a problem. You might want to put a couple spare ports on the side somewhere.
Jim
Wow, nice stuff! I don't have anything to add but I can't wait to see it in person. Looks like you're using a GM TPS and IAC on your throttle body. How big is the throttle body? Just curious.
"A question for people who have installed Rover EFI manifolds on Buick engines: what did you do about the thermostat bypass hose? Rover flipped it from left-side of the thermostat to the right-side. (Third photo.) Did you cut and weld, so you could use a straight hose? Source an S-shaped hose?"
When I put the EFI manifold on my 215 I capped the by pass in the front cover. You can still used the 215 thermostat housing on the EFI manifold. (I am using a Buick 300 housing that point straight ahead) The openiing for the bypass on the thermostat is below the water opening and doesn't need to be capped. I am currently using a Buick front cover on a Rover 3.9 with the capped off by pass. Seems to work OK without the bypass. If I'm missing something please let me know.
"Consider splitting your fuel rail in the front." Why bother. You don't need to take it off that often in my opinion.
Al Wulf
Thanks guys! Excellent feedback. I really appreciate the encouraging words.
QuoteConsider splitting your fuel rail in the front.
I thought long and hard about it, but decided adding flex hose or fittings to the middle of the steel rail looked like a lot more work. I think the rail is ready to be plated now.
QuoteHow big is the throttle body?
The Chevy throttle body bore measures 72mm in diameter, which seems generous compared to the Rover 3.9's bore of 65mm except that Chevy put a peculiar stamped steel extension on the back side of the butterfly to restrict airflow. Chevy evidently used various different restrictors to tailor the throttle body for different applications. I actually have two different junkyard butterflies, with two differently sized restrictors. Some tuners trim or remove the extensions.
QuoteI am currently using a Buick front cover on a Rover 3.9 with the capped off by pass. Seems to work OK without the bypass.
Whew! It would be a real pain to modify the Rover manifold to match a Buick thermostat cover. I'll move that to the very bottom of the to-do-if-time-permits list.
With regard to vacuum ports, I think the best way to make straight NPT-to-hose-barb fittings work might be to weld a couple little blocks onto the bottom of the plate and achieve 90 degree turns by drilling perpendicular pairs of holes. That's a couple more widgets to have machined before I visit the TIG welder.
What about the trumpets, are you leaving them the same length ? Std manifolds have them mounted at different length to compensate the difference in port length ? Also, what will be the distance between the trumpet and the top of the plenum, looks like there is very little space.
Yes, I'm aware of the trumpet issues... as reported here (http://www.britishv8.org/Articles/Rover-14CUX-EFI.htm):
QuoteThe inside diameter of each trumpet is 3.81cm. The cross-sections of the runners in the cylinder heads are significantly smaller, but in the intake manifold the runners quickly transition up to round cross-sections of about 3.8cm diameter each. The average total length of all eight runners measured from intake valve to open plenum is about 38.75cm. (Runners 1 and 8 are approximately 41cm long. Runners 2 and 7 are approximately 39cm long. Runners 3 thru 6 are approximately 37.5cm long.)
All eight trumpets will have to be trimmed, but how much they'll be trimmed hasn't been determined yet.
Equal length runners (from the port) aren't necessarily best, as McLaren demonstrated back in their Can Am days. McLaren had a lot of dyno time to play around in, and their conclusion was that deliberately varied runner length could widen the top of the torque curve. I wonder if there's a handy trick for individually optimizing them. Comparing exhaust gas temperature?
My machinist friend has suggested that we might roll out the tops of the trumpets to a larger radius and wider bell-mouth.
I've been reading a bit more about runners and trumpets. If anyone is interested, tmoss's posts <a href="http://eectuning.org/forums/viewforum.php?f=17">
here</a> might be interesting. In one of them he links to this document about plenum size and runner length: <a href="http://eectuning.org/forums/download/file.php?id=570">
Plenum Size and Runner Length</a> Interestingly, that document includes some runner volume suggestions. If you extrapolate to get from 302 Ford numbers down to Buick/Rover numbers, it seems to be saying that a good target for total runner volume (valve to trumpet tip) for a 3.5L engine might be between 342cc and 420cc per runner. Based on a quick estimate, I think the stock Rover/GEMS system comes in around 442cc/runner on average. I'm almost curious enough to cc them accurately.
I'm about ready to start installing vacuum ports. I realize now that I'd forgotten about the MAP sensor, so I'll add one more 1/8 NPT threaded hole for that.
Question: How much condensation accumulates in an EFI plenum? Crankcase vapor? Fuel fumes? Moreover, how concerned should I be about those contaminants affecting or damaging my MAP sensor? I'm thinking an extension to put the MAP port above the floor of the plenum might be a good precaution.
Curtis, I'm sure you've seen these...
http://www.actproducts.co.uk/shop/ti022a-superflare-trumpets-39-45l-p-350.html
Yep... and I saw the price tag too! (197 pounds = 311 dollars) A wise guy once told me: "There's a time and a place for everything." but I'm skeptical about the inside of an EFI plenum being the right place for carbon fiber bling.
Whoa!! I overlooked the price.... was searching and ran across it one day. Hadn't really looked at it close. I think other options would be in order.
Making slow but steady progress!
TestFit-FuelRailEtc.jpg
Fuel rail is complete and has been electroless-nickel plated. New air cleaner housing is complete. (Intake air temp sensor will enter from cabin side, under the dashboard.) TIG welding on plenum and intake manifold is ~90% complete. Welding inevitably causes a little warpage, so I brought the plenum back out to my very good friend Scott McRoberts (at McRoberts Machine Shop (http://www.mcrobertsmachine.com)) to have gasket surfaces (to Rover manifold and GM throttle body) fly-cut flat.
Next I'll take the plenum back to the expert welders at Peterson Fluid Systems and have them weld on bosses I've made for PCV, MAP, and vacuum advance connections. The sides of the plenum will then need to be cut down at least a half inch, and at the same time a ledge will be cut into the thick back wall. That will provide a sealing surface for the lid. Due to bonnet clearance issues, the lid can't be as tall in front as the throttle body is at the rear... Scott and I have a plan for a billet aluminum lid designed to match Buick-style valve covers. Trumpets will probably be tuned/trimmed/staggered/whatever... but I haven't decided on lengths yet.
Fuel injectors have been cleaned and flow tested, and should be back on Friday. Fuel tank with built-in high pressure pump is scheduled to arrive in January.
I guess it's time I start tracking down MegaSquirt parts.
It's looking great Curtis. You planning to have it done in time for BV8 2013?
Thanks. Yep... Omaha or bust!
Looking good Curtis. Just curious, what did you have to pay to get your injectors cleaned and tested?
Jim
$16 each. I chose this gentleman, and am very pleased with his responsiveness and excellent communication:
http://www.mrinjector.us/index.html
I realized you were building a test rig just hours after sending my injectors off!
That's OK Curtis, I'm not up and running with it yet. It looks like a good process he uses as far as I can tell and the ultrasonic cleaning is something I will add. How many data points did he give you on the testing? Was there a graph generated? What I've seen is that only testing for full flow is not enough, you need to know what it does at lower flow rates. After all, most of your driving is at part throttle.
Jim
I mailed them out Saturday and hope to have them back tomorrow. When they arrive, I'll scan and post the test report.
QuoteHi Curtis,
Your injectors arrived and have been serviced. There were several that were stuck closed but that isn't unusual for injectors that have sat with old fuel in them. They cleaned up well and are well balanced. I have found these Lucas disc injectors to be very well balanced when clean. Another benefit is they are very quiet. They hardly make any click at all. Your two spares are in a sealed package marked "spares" for storage. There is also a full report in the package.
I'm shipping them back USPS Priority. You will receive an email with conformation and tracking.
Thanks for the business! BIll
Test report:
Injectors.JPG
OK, so I would guess that maybe the mention of the dynamic tests that they run would be the cycling test which would apply to part throttle conditions. To my mind the partial flow is every bit as critical as full saturation flow. In fact, if the injectors are correctly sized they will never quite reach full saturation and will always be operating in the cycling mode, which means the pulse width is turning them on and off continuously with the pulse duration controlling the length of time they are switched on, thereby controlling the amount of fuel that is delivered.
In all probability your injectors perform uniformly in the cycling mode. However, I have seen sets of injectors that did not and in fact I have a set on the bench right now that preliminary tests indicated horrendous variations even though the full flow is pretty uniform. Hopefully a good cleaning will straighten them out but my fear is that the vendor may have made up a set of 8 by combining 2 sets of 4 from different engines and they may be quite different in their fuel delivery curves. A good cleaning followed by the cycling test should tell the tale.
Generating a fuel delivery curve is a good deal of work. You have to determine the flow rate at each point in the graph, which means a timed test and a fuel measurement for each injector. So for eight injectors and just a four point graph which is very rough, you have 32 measurements to make and enter into a program to generate the 8 individual curves, one for each injector. It also means having an injector driver that can deliver the appropriate pulses to all 8 injectors at the same time and is set up to allow you to select the right pulse width. That's a pretty sophisticated flow bench. Mine has the capability to do full saturation and a cycling test that gives about 1/4 to 1/3 full flow, although I really don't know what the pulse width is. The next upgrade will have selectable pulse widths, possibly with a digital pulse width readout but I won't be needing that right away. The cycling test does a pretty good job for now, and I can get back to improving that once my car is a little closer to being driveable.
OTOH, if I happen across a digital display that can give a pulse width measurement I just might have to do the upgrade. This is the one I'm using for the test interval:
http://www.ebay.com/itm/280892518436?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2649
Maybe they have one that can read in the millisecond range. Injector pulses run from about 2 milliseconds up to about 28 typically. It takes at least 1-2 ms for the injector to open (disc type injectors are a little faster so they have a little better dynamic range... at the cost of lower maximum flow capability but perfect for a 3.5) and beyond 28 ms we run out of injector time at high rpm. Once that happens no more fuel can be added without increasing rail pressure.
Curtis, you definitely did the right thing having your injectors cleaned and tested. The test results show that very clearly.
Jim
I have injectors off of two different Rover engines. They're all nominally "Lucas" injectors, but actually only one set has Lucas logos molded into them and the other set has Seimens logos. The Lucas ones had green stripes painted around their middle, whereas the Siemens injectors were all black except for a green plastic clip near the top. Exterior shape/dimensions were identical. Some of the Lucas injectors were banged up, so I had the Seimans injectors cleaned (plus one of the Lucas injectors as a spare.) What I think I've learned from your post, Jim, is that I shouldn't mix the two brands on my engine, even though they all flow 58cc/15s wide open because they might not all flow the same at partial throttle. Cool.
Next issue... how to "optimize" the length of the trumpets. The flared steel tubes all start out 100mm long. Rover installed them into recesses of two different depths... so the trumpets for cylinders 4,5,6,&7 appeared longer:
(http://www.britishv8.org/Articles/Images-V15-1/Rover-EFI-BB.jpg)
Why? Apparently to crudely equalize overall intake port length. Note how some ports in the intake manifold are longer than others because they cross over...
(http://www.britishv8.org/Articles/Images-V15-1/Rover-EFI-C.jpg)
The bottoms of my trumpets are flat against my Rover intake manifold, and my plan for some time has been to trim some of them. The longest trumpet(s) will be left 100mm.
Some time ago, I crudely shaped lengths of copper wire to follow the approximate center-line of each port and then I straightened and measured the wires. Based on that exercise, I expect the trumpets for cylinders 1 and 8 should end up ~55mm long and the trumpets for cylinders 2 and 7 will end up ~75mm long.
Would it be better to "cc" the ports instead, and equalize volume instead of length?
I don't have an extra set of Rover 4.0 heads around, so it's not convenient to cc their intake-side ports. Anyone know where to get cc numbers for them? (I guess I could cc my spare set of Buick 215 heads...) Suggestions? Opinions?
Curtis, the runners are very likely to be different lengths and also different volumes. The ports in the head should be pretty close to equal volume although some minor variation is likely. So your task is basically to equalize the overall volume of the ports and runners by trimming the trumpets. You should be OK in disregarding the ports when you do this but if you can check with Dan Jones he will have the exact answers you are looking for since he has done a lot of port work on these heads.
Sticking with the full set of injectors is the safest approach, short of more thorough testing and I think you will be fine doing that. Your flow test after cleaning gives sterling numbers and there is no reason to think partial flow would show any more spread. But when you start tuning just remember that it can happen and be aware of any inconsistencies from one side to the other.
Jim
Equalizing port/runner volumes wouldn't necessarily guarantee equal flow. If you're trying to qualify equal intake runner flow, wouldn't a flow bench test be you're best bet?....ie tune the trumpets for equal flow.
That would be ideal, if he could get set up with the same head he is using. Same valves would be even better. A change in the bell shape can also vary the flow, and it should be tuned at different flow rates to see if they can be equalized under part throttle conditions. He already has some pretty good tri-Y headers so exhaust flow should be pretty close to equal. Curtis, I think once you are finished with this you are going to be pleasantly surprised with the output of this engine.
Jim
I'm definitely a neophyte in all this... but my impression is that a flow bench is the right tool for optimizing and equalizing max flow rate but isn't all that useful for working out Helmholtz (pressure wave) resonance effects. (As the intake valve opens and closes very rapidly, pulses of pressure run through the air column in the intake track...) Am I right in that assessment? I certainly do wish I had access to a flow bench, and to someone who could show me how to use it! I'm sure it would be helpful, but I suspect it would only help with part of the puzzle.
My plenum top will be held on by just two cap screws, I have an extra set of Rover trumpets, and Rover trumpets are only ever held in by friction... so trial-and-error dyno testing is feasible. It would for example be easy to do a quick-and-dirty A-B comparison between staggered (equalized port volume) and max-length (100mm) trumpets in consecutive rolling-road dyno pulls. If conventional theories hold true, staggered trumpets should produce a higher peak torque whereas max-length trumpets might produce a flatter curve with peak torque at a lower engine speed.
My machinist friend and I have talked about spinning some full radius trumpets out of aluminum tube. I enjoy few things more than learning new manufacturing processes! If there's a ready market for them and if I could take some pre-orders...
Curtis, Bill on the SAOCA used equal length runners on a homemade intake. Shorter length moved powerband up,longer length helped low end torque.
Yeah, they had those BBC's (Can AM maybe) with the staggered length runners supposedly to broaden the torque curve? Seems I heard something about that.
Maybe as much worth considering as trumpet length is proximity. What I mean is those runners nearest in the firing order really should be separated in physical location. #5 and 7 are usually the biggest concern but where the runners come together in the center like these do the 4&3 pair and the 6&5 pair might be considered as well. The problem is one cylinder stealing the charge from the adjacent one and it was a real problem in the early flathead V8 engines, leading to extremely and unexpectedly low power outputs until the dual plane type intakes were developed, which explains where this seemingly odd V8 design came from. I do not really know how severe the problem is with EFI where you are not dealing with the fuel but the cylinder is sucking in air for close to 180 degrees and the runners are 90 degrees apart so it is at least worth looking at where the runners come out to see how much separation there is on those three pairs. I've never done this and I would expect them to be optimized but I've been surprised before.
On the resonances, again there will be some interplay between the runners and without having the valves slamming shut there is no way to evaluate it that I know of. Sometimes with IR intakes you could see liquid fuel at the top of the throttle bores where it had been carried by the reversion but without some extremely elaborate test equipment I'm not sure how you would measure this. Possibly some sort of orifice equipped accumulating manometer sensitive enough to measure a decrease in pressure in the runner, or possibly a differential from one end to the other. It would seem that a larger than average differential would indicate reversion. But then the question is, how much difference does it make and then what can be done about it? Without more information I don't have any answers for that.
Jim
QuoteNext I'll take the plenum back to the expert welders at Peterson Fluid Systems and have them weld on bosses I've made for PCV, MAP, and vacuum advance connections.
Adding-PCV-and-MAP-Ports.jpg
I can't say enough about how pleased I am with the job Peterson Fluid Systems did on the welding. Of course their main business is dry sump oil pumps, tanks, breathers, filters, lines, fittings, and other lubrication system components... but they're happy to take occasional welding odd-jobs too. Give them a call.
(http://www.britishracecar.com/v-PetersonFluidSystems.gif)
(Did I mention that they're a BritishRacecar.com advertiser? Tell them I sent you!)
Something else to report: I did a little cc'ing this morning to satisfy my curiosity. I only had rudimentary tools at hand but I measured everything twice and seemed to get consistent results.
The four intake ports on a spare Buick 215 head are indeed very nearly identical in volume. The number I got was ~90cc each.
The Rover EFI manifold really is symmetrical. Mine has been decked 1/2" on top. I measured: cylinders #1 and #8 ==> ~190cc, #2 and #7 ==> ~160cc, #3, #4, #5 & #6 ==> ~130cc.
The volume of a stock (100mm) trumpet is easy to calculate: just barely over 114cc.
Thus, if I trim trumpets to equalize volume, the four in the middle will end up ~25mm longer than the ones for cylinders #2 and #7 and ~50mm longer than the ones for cylinders #1 and #8.
Plenum-Lid-Trial-Fit.jpg
Now we have a very stout and tight fitting plenum lid. Scott McRoberts does handsome work, doesn't he! The photo above is just a trial installation. Tomorrow we'll radius the lid's outer edge. I'm not sure if we'll stop with that or machine-in a little bit more detail. I expect the lid will ultimately be powdercoated satin black to match the valve covers.
If it works as good as it looks you'll be taking orders before you know it!
That's pretty sharp Curtis. What does your clearance over your trumpets look like?
Jim
Thanks, guys!
We finished the lid this morning. After sleeping on it, I decided to cut a shallow 3/4" racing stripe down the top from front to rear, off-center to driver's side like the big white stripe on my roof. We flipped the cover over and cut out the bottom so it's 1/8" thick through the center.
So my next question is: "How much clearance between trumpets and lid do I need?" The front edge of cylinder number three's trumpet will be closest. Without trimming it's only got about 3/8" clearance. That probably isn't enough, so each trumpet will need trimming after all. Maybe 3/8" off each? (More get trimmed on #'s 1,2,7,&8 to equalize runner volumes, per previous posts.) For a reality check, I pulled out a Rover 3.9 plenum and measured its trumpet clearances. Stock, they had ~0.45" for cylinders 2&7, and ~0.55" for cylinders 3&5.
Best way to determine that would be with a flow bench and the lid is going to affect #3 somewhat. You could probably trim it a bit more than the others. The torroid formed between the lip of the bell mouth and the lid effectively extends the length of the tube, provided the cross section is correct. What you can do is determine the area of the cylinder from the lip up to the lid (taking into account the lid angle) and compare that to the area at the max diameter of the flare. Where the two match is going to be real close to the smoothest transition at minimum height, or maximum length of the tube. The angle will probably throw the calculations off a little but at least it'll get you close. See how that matches up to the Rover spec.
Jim
Newer photos:
RacingStripe.jpg
(decided to add a "racing stripe")
ThermostatBypass.jpg
(Rover intake manifold with thermostat bypass relocated to align with Buick front cover)
Ain't that purty!
That GT has come a long way since Annapolis in '98.
Thanks, Carl!
My MG was pretty well sorted by Annapolis... but your comment made me think back to really early test drives. Here's the earliest snapshot I have of it in "running" condition. (No air cleaner, hood pins, or license plate yet.) That's my momma and one of my sisters borrowing it for a hell ride. "SEPT 1991" was printed on the back of the photo by the developer.
TestDrive-Sept1991.jpg
I guess this topic has pretty much run its course. Hopefully soon I'll have something to post about plumbing, wiring, and electrical components.
Wait Curtis, we want to know your final solution for the trumpet lengths.
Your "racing stripe" looks really good btw. Nice touch. Very classy job all around.
Jim
Curtis, I'm impressed by the cleanness and the execution of your installation; great job A+
Curtis, regarding the photo, what about the tree ? Good to see brand loyalty. Cheers, roverman.
Curtis is there an update to this topic in particular what you ended up doing with trumpet lengths and manifold tracts?
Wow. It's been awhile already!
I ended up trimming # 1, 2, 7, and 8 to equalize volume as mentioned earlier in this thread:
QuoteThus, if I trim trumpets to equalize volume, the four in the middle will end up ~25mm longer than the ones for cylinders #2 and #7 and ~50mm longer than the ones for cylinders #1 and #8.
Here's a snapshot with the lid off, after trimming. I left the middle four 100mm long. You can see here that we milled the bottom side of the lid to get more clearance above them.
(http://www.britishv8.org/mg/CurtisJacobson/TunedTrumpets.jpg)
Re: manifold tracts... I may have knocked off a little casting flash. My heads aren't ported either. Perhaps one day I'll "flow" the whole system (or have it done.) Bigger/better exhaust headers sit higher on my to-do list.
Thank you for your interest!
thanks Curtis. I plan on showing this to my guru. I was looking at just buying from V8developments but they haven't replied just like ACT and SC-Power and V8 tuner just wasn't interested RPI has nothing in their Ebay shop either so I'll revert to DIY
Mine
20160222_104621_resizedSmall.jpg