Good morning! I'm having a very curious problem that at this point I'm considering to be related to the carb icing. So here are the details: stock ignition, wrapped header, weber 48 ida. When I first set all this up it ran great but after fixing a vacuum leak I've been struggling with keeping the car running longer then about 15 minutes. The weber is mounted on a nylon adapter that I believe is possibly isolating the carb too much and causing it to ice over. I haven't witnessed visable ice but while the engine is running the carb is very cold to the touch (my infrared thermometer says the carb is ambient tempurature (~75f ~82%rH ~70f dew point) which according to carb ice potential charts I'm in the "serious icing (glide power)" (these charts are for aircraft but I'd imagine the data is transferable)


Looking down the barrels I can see moisture droplets in the tops of the throttle plates but they are still in liquid form.

Other notes and troubleshooting:

• Previously, I've ran into some vacuum issues that I'm sure I have solved but it's left me a little afraid to attempt to drive the car because the quick loss of power that occurs. My vacuum testing procedure has been to connect a shop vac on each intake runner and then use an incense and move it around intake joints and other area to see where the smoke gets pulled in. I'm not sure how much vacuum the shop vac pulls but I figure it's likely a good enough analog for the type of vacuum the engine is pulling at idle. I have a clutch fan and I'm a little skiddish on holding it while the engine is running. I'm considering removing it until the issue is resolved though.
• I have also sprayed carb clean around the same joints while the engine is runnning and didn't notice any changes but I might repeat this test again for throughness.
• I'm doing this work in a north-facing un-conditioned but likely insulated garage with 14 foot ceilings. This might be keeping the tempurates a little lower and the humidity a little higher. I don't know I'm just trying to paint a picture.
• About the carb, I don't want to get into a weber configuration war but I beleive the carb is configured correctly. Float guages and tools were used. I am using modern (may contain up to 10% ethanol) fuel however, this hasn't been a problem in the past. I doubt it is.
  • Idle - 120/65
  • Main - 150 air / f3 / 170 fuel
  • venturi 37
  • float needle 300
  • This is the same configuration as I had on the previous engine (12a) now a s5 13b but again while it's running it runs well. I have a wideband that agrees with that sentiment too. Am I possibly leaving power on the table? Maybe.
• I'm not married to keeping the header wrapped but currently I'm not willing to take it off to see if the added heat will help.
  • Maybe I'll try a heat gun on the carb first...
• Timing has been verified multiple times and is currently stock.
• I have a fuel sight guage on the carb so if it's a fuel blockage its inside the carb.
  • I did just recently rebuild/clean the carb all passages should be free and clear.
  • I've rebuilt/cleaned it before so I don't expect any issues with the cleaning. But I don't have so much ego that I'll say it's impossible I've done something wrong.
• Fuel pressure is at the recommended 3.5 psi. (Or thereabouts)

I can take a video and/or collect more data if requested.

As always thank you for your time and help!

 

Look at the last thing you touched and maybe undo it to see what happens.

What vacuum leak did you "fix"? Maybe it wasn't a leak.

Some background on the reason you chased a vacuum leak to begin with. You stated it ran fine up to that point.

 

this is something that stood out to me. i'm not out to start a war, either, but it would be silly to ignore this. to reference what t_g_farrell suggested, maybe the leak you fixed was inadvertently serving a purpose because i don't doubt the car would run on a 12A configuration, but it stands to reason that that configuration should not consistenly run the exact same on a 13B with all 6 ports open all the time.

as far as your "potential icing" issue goes, i'm not sure how you would deal with that with an IDA. my Dell'Orto used to do the same thing. i never actually looked at what was happening inside the carb or manifold, but i would actually see ice form on the outside of the upper manifold, but when i got hip to what was happening, i just kept the choke engaged longer until the intake warmed up. before that, it was hell to drive the car in stop-n-go traffic within the first 15 to 20 minutes or so after starting it in the morning. i had to figure out creative ways to keep my foot on the accelerator pedal as much as i could.

with your IDA, choke is not an option since it has none. however, if you can figure out a way to safely keep the throttle cracked until the manifold (or carb) heats up, then it should be all good after that.

 

At the risk of sounding like clickbait; I'm a pilot and can be 99% sure it's NOT carb icing that's causing the problem. On your reference chart, the glide power and cruise power settings are the risk lines for pilots that are flying airplanes at altitude when the air temperature changes rapidly (*~2degF per 1,000 feet = as you climb, the air gets colder). At glide, the airplane engine isn't producing as much power burning fuel, so the risk of carb icing goes up, as the engine block and heads cool off rapidly with the cold ram air coming in the cowling. Under power, the airplane engine is firing and making more power and heat, which helps to keep the carb from icing up. Note that the ice forms in the venturi of the carb because that's the lowest area of pressure where the temp drop is the greatest. The presence of ice here will actually block the fuel jet, starve the engine of fuel because the ice prevents air from flowing past them, and the engine will stutter and stumble from running excessively LEAN.

All that to say that your car, sitting idle in your garage, at 75F, even with high humidity and condensation is extremely unlikely to build up ice in the venturi throat. If it did, you would literally see ice crystals formed on your fuel venturi and your fuel flow into the engine would stop, killing the engine. As your engine RPM increases, the pressure and temperature drop would be greater, and more heat would be created by combustion. There's plenty of heat in the engine bay to keep carb icing from forming, and your phenolic spacers aren't causing carb icing to form. You're simply not in the envelope where icing is likely.

That said, your performance issue is far more likely to be caused by a jet mismatch for air or fuel, or even an issue with fuel pressure or flow rate than to be caused by by carb icing. If you're really wanting to rule out carb ice, take a bread toaster and run it upside down on your intake funnels. You'll have air at 200F going down the throats and I'll bet you have the same problem. Airplane engines have a Carb Heat lever which moves an air door to draw intake air from around the exhaust manifold, which quickly melts any icing in the venturi. Once performance of the engine returns, you turn Carb Heat off because it results in a rich mixture and poor combustion performance. Be sure to shake out the toaster first so you don't get crumbs in your engine.

 

Cause then we'd have real problems for sure and a whole new thread to diagnose. Toasted bread crumbs can play havoc with apex seals.

 

When I had my Dellorto with the RacingBeat intake setup on my SA, it would ice up but only in the winter at around freezing temps. I lived in Pittsburgh at the time and when the weather and roads cleared in mid winter I would sometimes take the 7 to work. Like @diabolical1 mentioned, I just had to keep the choke pulled until the intake warmed up.

 

You are likely correct LongDuck I haven't seen any physical ice but the carb does condensate a bit. I was asking more as a sanity check.
I did find yet another small vacuum leak.
t_g_farrell and diabolical1, I wasn't super clear but the mating surface of the s5 LIM is a bit well loved. I've lapped it a bit using wet sanding glued to a granite block. I didn't take it down to dead flat across the whole surface I figured around the ports would be enough.
5th/6th ports are currently closed but will be operational in the future. I do expect to need to change the main jets at that point.

You've probably had a double take or two by now. I've been hiding a bit of information becuase I wanted to make ensure that the focus wasn't on that at first. I've been slowly developing an adapter so that we can mate a weber ida carb to the s4/s5 LIM. Eventually, I may go EFI but for now I wanted to keep the carb. I also wanted to retain the port accuators (this isn't a race engine) and ultimately I wanted a challenge and to be able to possibly give back to the community. After a while of mulling this over I finally put together a back log of ideas to justify buying a 3d printer. I still don't have all the time to dedicate to my backlog but eh.. I'll eventually do some project planning.

As for the part itself. It works. I'm having some printer issues, (and I may ultimately send a production run to a service to either print it in metal or figure out aluminum casting) and this last prototype had a little warping. The previous prototypes I had incorrect data and the ports were ill-sized. For those wondering I've been printing in PPA-CF and Polymide-Copa. Both are nylon and rated for the the tempuratures I'd expect to see. On the plus side the intake charge is cold but I don't have actual numbers.(Another backlog item is a weber CAI plenum)

As for the vacuum leaks I don't think the parts are warping in the engine bay but I'm not ruling that out. I'm still testing which way to tackle sealing; gasket, rtv, orings, so on. Sorry if anyone feels that I've misled them in the post that wasn't the intention.

Both of the following images are earlier prototypes but the overall design hasn't changed.

 

Might not be a leak, but rather the Nylon adaptor flexing under the heat and vacuum.


Can you take that nylon piece you made and use it to cast an aluminium piece? Aluminium is pretty easy to cast if you just want something that isn't going to support a load.

 

Qingdao, that is a possibility both flexing and casting aluminum. The carbon fibre additive is supposed to help with stablizing it though.

Have you any experience in casting aluminum? If so, any tips or gotchas to consider? I think it'd be fun to try.

 

Heat does funny things to polymers regardless of whether or not they have CF strands in them. Add gasoline in the mix then things get really crazy.

Do you weld? Thats a pretty easy shape to make out of steel tubing and a couple of pieces of flat bar.



I've cast more bronze than aluminium.

Get clean aluminum and you're cast will be easier. Less drauss

Take that piece you made and make it out of foam. Foam casting is VERY easy. Just comes out hella rough.

Get some mag salts if you want to lessen the chance of cracking.

Use a large sand box. Bigger the sand box the slower it will cool and the less chance of cracking.

 

Yeah, I understand. The material is supposed to be gasoline resistent and heat resistent to 440f. I had considered the radiant exhaust heat but wasn't sure what that temperature may be. Air is a decent insulator, and the engine itself doesn't get that hot, and other research has shown I'm probably more okay than not.

I discovered a little earlier today that the plugs were fouled. I've been being cheap and recycling old plugs. Replaced them and started running much better.

As for welding, I don't unfortunately. I probably should get one but I haven't yet. Getting the part welded was my original idea.

For casting I can print the model in lost cast material I'd assume that'll make the end results a little better but not sure. I suppose I'll cross that bridge later. Sounds fun! I'll make note of researching mag salts.

Thank you for taking the time to chat with me.

 

CouchBuilt on YouTube has a really good series where he puts a rotary engine into a BMW. He designed an intake and had it 3D printed in metal from a service.

Craft Cloud for cheap 3D metal printing in Aluminum, Stainless, Inconel and more. https://craftcloud3d.com/ 10% discount code off your first Craft Cloud order! REFZ10W5P18

You Tube

 

I wondered why you of all people would post what you did in the original post. Makes sense now.

Do you have a 6 port RacingBeat intake you can swap in to see if it makes a difference? Then you would know whether to focus on the new part or look else where.

Cool project. Its amazing that you can now 3d print a prototype part and send it out to get printed in metal. I know the NASCAR guys have been printing metal parts for a decade or so here in Charlotte. I remember my neighbor telling me about it (he was the gas man for Gnassi and also later did fab work for the team) and it was very cutting edge back then.

 

Hey Kansas! I hadn't seen that video yet but I was recently made aware of Craft Cloud. I want to say they were going to charge ~$50 for it printed in aluminum using SLS, IIRC. Thank you for both the code and for your early encouragement.

t_g_, haha, sorry for the misdirection! I think I mentioned in the original post (There were a few scrapped drafts!) that this is a junkyard engine so unknown history or build quality. S5 NA irons, T2 housings ((I'm not sure how to tell the difference between s4/s5) or at least the exhaust port is T2), s5 rotating assembly. However, in my uneducated opinion everything appeared fine, compression seems strong (but no numbers yet) Anyway to answer your question. I do not but I wouldn't mind borrowing one with deposit to remove more variables.

However, I think I might have the running issue sorted last night. After replacing the plugs the engine began running must better I discovered that it seems to need 65 idle jets instead of the 60 I previously had. After swapping that and adjusting the speed and mixture screws it's idling a lot better. Still a little higher than I'd like ~1000 but the 65 jets I have are actually f8s so they have an air hole in them. I'll leave it as is for now and solder them later if I'm unhappy with the results.

So at the end of the day I suspect the majority of my issues were as previously suggested jetting, which likely fouled the plugs, (it was indeed quite rich), and of course the vacuum leaks. But yeah, I've been pretty happy with the work and dedication that the 3d printing community has done so that I can come in and create a few dumb things. Locally, we've had a few entities print whole cars, houses, and other things that you'd never expect. Some of the work is private some is ORNL.

 

Just dropping this here for posterity. Who knows who it could help at some point; good thread!

 

Yeah a good reminder for sure! I plan on getting those jets and venturi soon.