aprilia billet valve cover with cam sensor port

[Seb@NLR]

Few more adjustments to make on some edges and wall thickness in a few areas and it will be on the machine this weekend. This is for a aprilia 450/550 sxv valve cover that has a port for a cam sensor to be used with a aftermarket ecm and run full sequential.

http://i103.photobucket.com/albums/m124/NLRLLC/valvecoveraprilia.jpg
http://i103.photobucket.com/albums/m124/NLRLLC/valvecoveraprilia2.jpg

[exFSAE]

Going to machine it, are you?

Hmm.. I predict..

a) Big up front cost for billet stock
b) A good amount of machine time just for some custom fixtures for it
c) At least 4, possibly 5 different setups for it
d) There are features on that which literally cannot be machined.. unless you fudged it with a TINY tool, in which case your run time is gonna go through the roof unless you have a 15000 (at least 10k) rpm spindle
e) There are some deep-reaching radii which are going to require some long tools, which may or may not chatter
f) Whole thing will really have to be clamped all over so it doesn't vibrate
g) Mucho cost passed on to the consumer.

Neglecting all the up-front time and cost, good couple hours of setup and runtime. Plus billet stock.. easily couple hundred bucks for the part.

Looks almost constant wall thickness. Why not have the patterns rapid prototyped and have the thing cast? Only a very few spots that have to be machined.

Or.. if the OEM piece is aluminum and the only change you have is that oblong shape stickin out the top that looks like it holds a Hall effect sensor.. sell a simple jig that lets the customer drill out the hole for the Hall effect sensor, and sell a small weld-on "bung" for the thing.

Sorry to sound like an ass.. but I spent wayyy too much time running a machine shop and having to do one-off crazy pieces like that for grad students and professors. Got to charge em out the ass though!

[Davidimurray]

Hey Seb

Looking good, when you've got some pricing sorted for those covers let me know as we would probably be interested in getting hold of one.

How's the engine coming along?

Cheers

Dave

[Seb@NLR]

ummmmm your right..Im going fishing.
You forgot the rest of the alphabet with reasons .
If you think a billet stock is expensive you've got to be straight out of your mind to suggest rapid prototype and having it cast . The mold itself would cost about 5-6k not to mention the qty you would have to do in a cast piece.

If the OE part was aluminum I would sell a weld on part but its not thus the valve cover in billet.

David..rods are out to get shot peened and polished. Pistons are finished, valves are on the way.

[exFSAE]

Coulda sworn you could get things prototype cast for less than that, if you make patterns with help of a 3D printer (though the ones with SLS are stupid expensive indeed).

Just saying. That design is silly to machine and is going to be a pain.

[Seb@NLR]

piece of cake....this is simple compared to some of the stuff we do.
We will see in the end.

[Homemade WRX]

You can rapid prototype and do a simple investment cast. Cheap and easy. Then machine finish the cast. It sill save tons of machine time and material cost.
Or get crazy and machine plastic or wax to make the mold off of(really only saving cost of material then and simplifying some of the maching challenges then). Does the piece even NEED to be metal (could a plastic piece serve the purpose and be lighter? (yes, I know we all love shiny pretty billet aluminum but...)
Just tossing out ideas.



Originally posted by Seb@NLR:
ummmmm your right..Im going fishing.
You forgot the rest of the alphabet with reasons .
If you think a billet stock is expensive you've got to be straight out of your mind to suggest rapid prototype and having it cast . The mold itself would cost about 5-6k not to mention the qty you would have to do in a cast piece.

If the OE part was aluminum I would sell a weld on part but its not thus the valve cover in billet.

David..rods are out to get shot peened and polished. Pistons are finished, valves are on the way.

[exFSAE]

Or here's one better! Don't even need to rapid prototype the pattern..

Probably already have the OEM part since you're measuring and reverse engineering it. I'd machine that sensor holder and put it on the OEM valve cover with adhesive.. maybe a little extra around the base to get that radius in there. BAM, pattern done on the cheap!

And then yea, investment or sand cast and finish machine FTW.

[Boston]

ex, what school did you go to again?

[exFSAE]

Not gonna say http://fsae.com/groupee_common/emoticons/icon_smile.gif

Don't think it's that important though.

[Garlic]

Originally posted by exFSAE:
Or here's one better! Don't even need to rapid prototype the pattern..

Probably already have the OEM part since you're measuring and reverse engineering it. I'd machine that sensor holder and put it on the OEM valve cover with adhesive.. maybe a little extra around the base to get that radius in there. BAM, pattern done on the cheap!

And then yea, investment or sand cast and finish machine FTW.

Nothing compares to billet for quality and accuracy. Epoxy or glue is a hack job in comparison.

That design is machinable by a professional shop. Easy? no. Some changes possible to make it easier? for sure, but at the expense of weight and bulk. There are hundreds of more complicated designs made in billet every day.

The cost of billet is not that expensive, casting is way less cost efficient for small production. Someone is talking out of their rear.

I find there are two kinds of machinists out there. Those that complain about everything that isn't a flat plate, and those that enjoy a challenge and work to accomplish what is needed. Love the latter.

Now, there are definitely designers that make parts unnecessarily complicated to machine, but I don't see it in this case. The application is what makes this aprt complicated. And it is a part that any good machine shop wouldn't blink an eye at.

[MalcolmG]

Personally I think everyone sucks for suggesting metallic alternatives: take a mould from the standard valve cover with the extra boss glued on and make the thing out of carbon fibre w/ a high temp epoxy or phenolic resin. Bling, light, and cheap http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

[exFSAE]

Challenges are fun. But cmon, that geometry screams "cast me!". Job shop will love you for it because they can charge you for lots of extra setup and tooling.

Nothing beats billet for dimensional control and quality.. if its a good design. I just see that one, with the thin walls and long reach for tools it will be more prone to chatter and cutter/part deflection.

In any event, like I said, there are ways of casting that (particularly with just making a pattern off the OEM part with the sensor thing glued on) which would be more cost effective and less pain in the butt.

[Superfast Matt McCoy]

This looks like a straightforward, two operation piece easily machinable with common tools by any qualified CNC tech.

I don't see the issue.

Also, exFSAE, why are you so reluctant to tell us what school you went to? Are you trying to be mysterious? Do you think it boosts your credibility? This doesn't help your claim that you are not, in fact, Bill Kunst posting under a different name.

[exFSAE]

I'm Steven Segal. Try my energy drink! (http://www.lightningdrink.com)

Not so sure that part could be done in 2 setups. You can hog out the underside, flip it and start doing the top.. but you'll notice the sides have radiused profiles which you'd have to stand up and run around with an EM.. unless you REALLY wanted to take the time and fudge it by surface machining everything. Could do it, would just take a long time.

And you will most certainly have to make custom fixturing for it and probably move clamps around so it doesn't rattle to hell and chatter when you do the 2nd side and start to get to thin walls and floors and not much support.

Beyond that, on that top picture in about the middle of the part there's a place where that edge fillet runs down and merges with a flat part. At that merge, there's no way you could get a radius cutter down in there to merge like that and go completely flat. You could again surface machine that, but to get that sharp transition youd have to fudge it by surfacing with tiny tools (1/16 ball EM? 1/32?) and then cleaning up with a flat EM.

Again.. its the perfect geometry for a cast part, especially if the pattern is 90% there already from the OEM thing. If it were a 1 off internal thing, then yea screw it you could cut it from billet. But selling it, with some non-zero amount of volume, I just think casting is the way to go and cheaper.

[Drew Price]

I haven't looked at the Aprillia engines up close for very long, but the cam cover on our 450 single is only about 5x5 inches square, which would make the really deep stuff on that cover maybe 3 inches deep - not all that long a tool needed.

Yes, there are some tricky transitions, but the thing's a cover, and the transitions are on the bottom. All that needs to do is:

a) Be oil proof.
b) Hold the sensor pickup in the right place.
c) Be light.

I'm with Matt Brown on this one, looks pretty straight forward.

Looks good Seb, like you said, there were some final tweaks to be made, but I imagine it went pretty smooth.

Best,
Drew

[exFSAE]

3" is very long for a tool that is small diameter. Just as an eyeball guess, I would imagine some of those deep features are close to 4 or 5x diameter in length for whatever ball endmill you'll run around in there. Even for a carbide tool that's pretty long.

In any event. I'm not trying to argue whether its machinable or not. It is, more or less (depending on what you do with that transition). It certainly isn't the most technical or silly piece I've seen.

And if you're just doing a 1-off part for yourself, or for a FSAE car, who cares how you make it. Given that almost every industry is going lean these days, and given that manufacturing is getting very competitive in that respect.. if there's a better, more cost-effective solution.. why not use it?!

There are places you can get parts cast in small batches for cheap. I know one that specializes in it, and gets parts turned around in 7 days.

That part geometry is much more condusive to casting than machining.

So why not? If you can get the same form, the same function, and close to the same dimensional accuracy.. why spend a penny more or have your part any more expensive than it needs to be?

But anyway. Not gonna argue the point http://fsae.com/groupee_common/emoticons/icon_smile.gif

[Seb@NLR]

final solidworks drawings. Material is in and time to start making chips.
http://i103.photobucket.com/albums/m124/NLRLLC/valvecoveraprilia1.jpg
http://i103.photobucket.com/albums/m124/NLRLLC/valvecoveraprilia2-1.jpg
http://i103.photobucket.com/albums/m124/NLRLLC/valvecoveraprilia3.jpg

[kmrobinson]

Just a thought from me: removing the stock water pump and replacing it with an electric one (worked on our 2008 car) allows access to the water pump output shaft. We put a simple cover over it with a hole for fluid flow. This shaft runs at cam speed, so a simple trigger wheel and sensor might work for a cam sensor.

This was just a concept we had, but running batch fire was fine for us.

[Wesley]

I think the biggest part of machining that that most people see is just waste.

Which, in FSAE, happens EVERYWHERE! It looks good, machine away!

If you were to make a thousand, then you might look into some casting, but theres no reason to for a typical FSAE-oneoff unless you know someone who is itching to do it for you.

[Davidimurray]

Just a thought from me: removing the stock water pump and replacing it with an electric one (worked on our 2008 car) allows access to the water pump output shaft. We put a simple cover over it with a hole for fluid flow. This shaft runs at cam speed, so a simple trigger wheel and sensor might work for a cam sensor.

Out of interest why did you swap out the stock pump?

Cheers

Dave Murray

[fart can]

so what is the link to this cheap casting company with a 7 day turn around?

[exFSAE]

If you google "prototype casting" it winds up being the first result..

http://www.protcast.com/

[Seb@NLR]

cheap LOL!!!

2500 tooling fee
fla- 100 dollar to 1000 depending on the z
400 per part.
at 15 units 200 per part
in aluminum

[exFSAE]

You can probably reduce that quote given that you could have the final part pattern already done yourself with the OE piece. Potentially wipe out some of the tooling cost as I believe they do patterns with SLA.. and that is a BIG part of the up front cost. SLA patterns are what like $100 per cubic inch? I could be wrong though.

Let me know how many hours your billet pieces take to run, plus the programming and setup and all.

Not sure how you bill your CNC time, but I'd assume somewhere in the range of the going rate.. $65-110/hr for machining, setup, programming...

Couple hours programming, maybe some time in fixtures, hour or two run and setup time per piece...

Plus billet stock price.. dont know your dims but if it were a 5x5x5 cube of 6061 maybe $40-60 bucks?

Adds up quick... machining billet is an expensive process. Except for in FSAE of course where all the costing is ridiculously make-believe http://fsae.com/groupee_common/emoticons/icon_smile.gif

[kmrobinson]

Originally posted by Davidimurray:

Out of interest why did you swap out the stock pump?

Cheers

Dave Murray

The thought was we were getting coolant infiltration in the engine oil (there was some discussion here about our milkshake oil that was seen on our 2008 car). The seals on that shaft are known to be problematic sometimes.

However, we have good reason to believe the entire engine case is a means of coolant infiltration, even into the transmission.

[Seb@NLR]

here is the billet valve cover with cam sensor port. This is the first part off the machine and some clean up machining is needed as well as some modifications to the internal dimensions for clearance of cylinder head items.

Most important is the ability of the sensor to pick up on the exhaust lobe for sequential operation. This worked perfectly .

Enjoy.

sebastian

http://i103.photobucket.com/albums/m124/NLRLLC/HPIM1096.jpg
http://i103.photobucket.com/albums/m124/NLRLLC/HPIM1101.jpg

[John Grego]

Wow that's impressive

[mjdavidson]

That's pretty cool. Looks great.

[Erich Ohlde]

have you run the engine with that valve cover? I'm thinking you are going to run into some false triggering at low rpm due to the compression release...things (for lack of a better word). Why not trigger off of the intake lobe?

[Seb@NLR]

engine has not run with this valve cover. Im not sure Im following what your saying. How is compression release going to create false triggering of a cam sensor? You lost me

Triggering off the intake lobe will not work.

[Brian Schien]

Are you running this on a Motec.

We are looking for a simpler solution so we can switch from stock ecu with a tuneboy (sucks) to a M400. We were thinking of machining the stock cover and using some sort of mechanical fasteners or welding a bung on. Yes I now it magnesium we have a very competent weldor who has magnesium experiences so not an issue.

Looks good keep up the good work if you can make them for a reasonable price you might be able to sell them.

[Erich Ohlde]

the decompression 'actuator' on the exhaust lobe.. the thing that actuates the exhaust valves on the compression stroke so the engine can start. that might false trigger below ~2000rpm. why can't you trigger off the intake lobe?

[Erich Ohlde]

we are running the aprilia off of an AEM. it is working well for us. but we are thinking of switching ecus for next year. something with traction control would be nice.

[Seb@NLR]

This sensor picks up the tallest edge of the exhaust lobe at which point it is less than .070
Once the lobe passes there is over .200 of distance to the sensor.
The sensor needs a ferrous metal trigger within .070 to operate and the actuator is too far to be used as a trigger.

False triggering will not be a problem using this sensor combination and distance it has been set up at.

The intake lobe is not in the right position plus the cover would have to be another 1.400 taller.

Yes this was designed for a Motec ecu but can be used with any ecu that requires a cam sensor. We looked into another mounting solution but building a billet valve cover was the best solution for us.

[James Kucinskas]

After all the discussion on the first part of this thread about the best way to manufacture this part, I'm curious to know a little more about how you ended going about machining it. How many operations did you end up doing? Any special fixtures/jigs? And of course...how long was the prep/cycle time?

Looks very nice.

[Seb@NLR]

2 ops. , no special jigs. 90 minutes per side.

[Davidimurray]

Hi Seb

Fantastic looking job. How are you getting on with your engine?

Any thoughts/prices on making more billet covers?

Cheers

Dave

[Seb@NLR]

coming along. Turbo manifold exhaust flanges and spigots are off the machine and ready to go.

Billet valve cover with sensor 500.00 including sensor and oring.

[SArgo@EPM]

Hi

we're looking to install a cam sensor on a an Aprilia SXV550 with a Walbro ECU - A (FSAE Team).

I found that the ECU is expecting a sine wave with a difference in the signal where there's a missing tooth. Am I wrong?

I can't find a way to produce that kind of signal with the cam lobe.

I'd like to have more information about it plz.




Originally posted by Seb@NLR:
This sensor picks up the tallest edge of the exhaust lobe at which point it is less than .070
Once the lobe passes there is over .200 of distance to the sensor.
The sensor needs a ferrous metal trigger within .070 to operate and the actuator is too far to be used as a trigger.

False triggering will not be a problem using this sensor combination and distance it has been set up at.

The intake lobe is not in the right position plus the cover would have to be another 1.400 taller.

Yes this was designed for a Motec ecu but can be used with any ecu that requires a cam sensor. We looked into another mounting solution but building a billet valve cover was the best solution for us.

[AustinS]

We used this same setup for a couple of years while running the Aprilia.

In this setup, the ECU is looking for a sine wave with a change in signal for missing teeth from the crank (REF) sensor.

We used a Honeywell GT101 hall effect sensor for the cam (with Seb's billet valve cover), which will send back a digital 0-5V signal to the ECU. With the sensor around .07" from the tallest edge of the lobe, we didn't have any issues with our ECU picking up a cam signal.

[B Lewis @ PE Engine Management]

We had a team this year trigger off the cam lobe itself. It seemed sensitive to sensor location because it could throw off the relationship with the crank sensor but as far as I know it is still working with our new EDGE ECU. I think that in the near future we will offer a MAP based sync to use like the factory does for this engine. That way you won't need to add a cam sensor.

[Hugh]

We are planning to use a cam sensor with the SXV550 this year and from what I can see the intake lobe would provide the best trigger. Its been said a few times that the intake lobe wouldn't work and the exhaust lobe is the way to go, why is that? I know its just been said that it can throw off the relationship with the crank sensor but why? Any help with this would be much appreciated.