Thanks, I'll definitely keep posting. Regarding HP, that's a bit of a long answer, with a lot of guessing / conjecture... I believe HP came in around 300HP for the EFI conversion with stock intake (see previous post about dyno issues that day). I get the sense that a stock BBi typically comes in around 270HP (or less...). I am hoping that the ITB themselves help with breathing, but at the end of the day, it's still a small 35mm inlet port on the head (at the widest). If you go by Jenvey's math for inlet-to-HP max, then 35mm would translate to a max 360HP. I'm hoping for 330HP; which is essentially attributing another 30HP to the ITBs over the EFI conversion. Essentially 270 stock + 30 EFI + 30 ITB = 330 HP (total guestimate). From there, it's head porting, cams, and higher compression. I don't want to go overboard with it (it's still a street car). So I'm shooting for another 100HP; putting the total at 430HP. EFI and ITBs essentially remove all the barriers to getting proper air and fuel into the engine. Seeing the HP numbers some of the BBLM cars got with just modest upgrades, 430HP should be a relatively easy target. At the end of the day, if the car ends up in north of 385HP, I'd be happy, but there's a certain psychological accomplishment to breaking 400HP. I honestly don't think it should be hard to attain. It'll end up being a lot of money to get there, but there's a lot of value to me in the whole package (not just the HP number). The sound, aesthetics, feel, and reliability are all factors as well.
Status update. Mostly distractions, but not all. PDM (distraction #1) I've been a bit distracted with the cabin PDM project, but that's wrapping up. I'll start a new thread with all the details and pictures, but in-short, I've piggy-backed off the existing fuse and relay wiring as ways of distributing power from the PDM. All the fuses have been replaced with 30A ceramic, and the PDM is now responsible for everything fuse related. Additionally, many of the switches have been converted to digital inputs that I can use in PDM functions. Temporarily, I have a CAN keypad that allows me to quickly debug. The nice thing is that it's all easily reversible as all of the physical changes are in the foot well and glove box, and consist of simply swapping spade connectors from stock to the PDM outputs. I even purchased the same white fuse panel connectors: TE Faston 280232 Since the outputs are now digitally controlled, I can easily change how the stock interactions work. Eg, the glove box now opens when the key is on accessory. I can manually turn on/off fuel pumps without having to jump the relays. I can have the low beams turn on when the headlamps are open, instead of having to move the lever down (which I often bump on my leg when driving which flickers the light). I can turn on each of the radiator fans manually, or use the ECU data to turn them on automatically when a given temp is hit (eg, replicating the stock behavior). I can have the flash-to-pass lights function as a fog light (instead of just momentary flashing). Basically, I can tweak the car's behavior without actually changing any of the wiring. CAN-to-Analog (distraction #2) Initially, I figured all of the gauges in the dash (except the tach) would be fed from the original analog sensors. This meant the ECU wouldn't know about the oil pressure or oil temp, but that's fine because I had the old-school gauges. Honestly, this is probably where I should have drawn the line, and just accepted that the ECU doesn't see everything. This has been nagging at me though, as it would be nice to have that data in the ECU. As far as I can tell, there are two main approaches other "restomod" cars typically take: Option 1 is that you replace the gauges with a digital dash. All sensors feed into the ECU, and then you send all that data via CAN bus to a digital dash that can be programmed. Fine for modern motorsport, but no way I am replacing those Veglia gauges for a screen. Option 2 is that you duplicate all the sensors. The original analog sensor goes to the dash, and the new digital sensor goes to the ECU. This is basically what I did for the coolant, as the stock sensor was left in place, and I reused the unused CIS coolant sensor port for the ECU. That gets a bit more tricky when it comes to oil though. I'd like to try option #3 (far less common), which is to send the sensor data to the ECU, and then use the CAN data to send analog signals back to the gauges. This is similar to option 1, but there's extra effort involved to convert the CAN data into analog signals the original gauges expect. There are very few off-the-shelf options out there to do this, but I am going to experiment with one from Lingenfelter: Lingenfelter CAN to Analog Output Module This requires sniffing/parsing the Motec values from the CAN data, then mapping/calibrating those values to match what the gauges expects. Again, this isn't strictly necessary. The current setup works fine, and I'm no worse off than I was prior to the conversion. So as I get into this, if I'm finding that the juice just isn't worth the squeeze, I'll mothball the idea. It's not like I don't have other projects to attend to. Tuning This is probably why I'm letting other things distract me, because the tuning is in a good spot. The car starts right up, idles great, drives great, sounds great. I don't even have closed-loop-fuel trimming enabled. The aluminum 3D printed throttle crank arrives this week. I also received a pressure damper from my Motec dealer; which he says would allow me to switch back to using MAP / Volumetric Efficiency based tuning. I'm not sure I really care since the TPS based tune I have now is working well, but I'll try it. After that though, I think I'm good to call the ITB conversion done! Side note, I am noticing a dip in fuel pressure on the RH bank under hard acceleration; which can lean out that bank and lead to knock (between 4.5k-6k RPM). I think it's due to a dirty fuel filter. If I compensate by giving that bank an extra 10 kpa via the fuel pressure regulator, the issue goes away. Just goes to show that without these sensors, you really don't know what's going on in your engine. I'll report back on this after I replace the filter. Scoops So I did my first batch of 3D prints. Each one takes between 5 and 11 hours, and I have to do 9 prints for each scoop. I only did half just so I could get a sense if I was on the right track. It basically takes a full week of printing to get half a scoop; which then requires using your imagination to see how it'll look. Honestly, I think it looks better than I expected, so I'm a bit more excited for the scoop concept than I was before. There's a lot that I now need to tweak. It's too short, the supports are too high (making the skirt touch the engine lid), the front inlet should extend forward, and the rear probably needs to come up to make sure the rear most cylinders are getting enough air. I'll need another week to print another revision. Hopefully it only takes one or two more revisions before moving on to the final pieces. Image Unavailable, Please Login Image Unavailable, Please Login Misc. distractions I'm working through some of the details on the electric AC from Classic Retrofit. I have to do something, as I'm dying in this Miami heat/humidity without AC. Also, I made a call to Ron Davis about getting one of their radiators. It's my understanding that the radiator itself will not perform better per-se, but it's lighter. The better performance comes from improved fans and shrouds that prevent the air from taking a path of least resistance. So basically, if you wanted better performance on the stock radiator, just get better fans and seal them properly against the radiator. By going with a Ron Davis radiator with fans/shroud combo, I get a bolt-in replacement that is lighter and performs better.
Good day Robert, Why not do option "4"? Simply tap into the existing analog sensors, add some appropriate analog buffers (can be electrically isolated if desired), and then feed the buffered analog signals to your ECU... This is far cleaner along with no messing/affecting the original analog system... and would be a lot easier than your option 3 provided you can or have access to someone that has some basic analog electronics background. Cheers, Sam
You're probably right. The ECU has analog inputs with 100k input impedance, so I'm not even sure I'd need the buffers. It might literally be as simple as tapping into the signals, and sending that into the ECU. Considering the other projects I have going on, that's probably the more practical route. I liked the CAN-to-Analog option because it seemed to give me flexibility to do more interesting things. For example, I'd be able to use modern digital sensors with better accuracy (not sure it'd be consequential). Or maybe triggering the oil pressure dummy light for any warnings from the ECU, not just oil pressure (eg, knock could blink the light). Or maybe using CAN to activate a relay. That too could be done somewhat easily the old-school way. It seemed like there were a few interesting use-cases for converting CAN to analog, but I'm probably over-engineering this particular solution.
Good day Robert, Indeed, you may not need to buffer the signals... but it all depends upon the input section of the MCU and how it was designed. Input impedance is one thing, but it could also include internal biasing (voltage or current) as well and so this could back feed into the original analog circuitry. Since I am bit "particular" design-wise, I tend to err on the sign of caution and so I would personally buffer the signals to be safe... but also buffering would be trivial to do with a handful of inexpensive comments (assuming full galvanic isolation is not implemented). As for CAN bus... tapping into the existing analog network would not prevent you from implementing a CAN network. In fact, you could even do a CAN to bluetooth and have all of the sensor date sent to your smart phone,tablet, or laptop...all without cables. Cheers, Sam
Status update. Throttle Crank My aluminum throttle crank arrived (3D printed aluminum is amazing). I installed it without the bushing just out of curiosity. I'll run it that way for a few weeks, and then drill out the center for the bushing. Honestly, I don't feel any grittiness in the throttle at all, so I'm not really sure how much of a difference it'll make. Side note, without the ball studs installed, the aluminum throttle crank is 70g vs the stock's 110g. – despite being larger to accommodate the ITB geometry. Tuning After installing the throttle crank, I quickly rebalanced the banks to make sure everything was even. While I was there, I added a 1% TPS row to the tune. I now have 0%, 1%, 2%, 5%, 10%, 20%, etc.. The 1% row allows me to refine the tip-in; which was often just a tad lean (ITBs are very non-linear). I'm about ready to call this tune done; which means I can dyno the car to see where this all netted out. I'd like to do a few more small bits first though (eg, replace the fuel filter on the bank that has a pressure dip under hard acceleration). Scoops I started focusing more on the front corner of the scoop. The middle and rear are much more straightforward, and just add time to the feedback loop (which can take a whole week if I print the whole scoop). I lowered the print quality settings as well. It requires more imagination, but I think it's worth it. I'm mostly settled on the exterior design; which is inspired by the Le Mans and Koenig scoops, but it's not a direct copy. I added the skirts back based on the stock lids, and rounded the corners using the same radius as the stock lids. Similar to the stock lid, the skirts taper towards the front. The scoops have inner flanges that bolt to the aluminum bracket and create a seal – so there's no visible hardware. I'll need another few tweaks to get the clearances right. I'd like a large opening, but I also don't want it to bang on the roof when I close the lid, etc.. Image Unavailable, Please Login Image Unavailable, Please Login Just a reminder, the sausage filters are temporary. I didn't feel like removing them every time I wanted to test fit the 3D print. I need to figure out a solution for the filters though. I can do something inline that mounts to the inside of the scoop, or maybe just a shorter ITG filter that is offset to remove the overhang (much simpler). I also need to settle on how to make the final part. One option is to simply skin the 3D print in carbon fiber; which I would then paint satin black (not interested in a modern CF aesthetic). This would add strength and impact resistance; without requiring a plug and mold. I would probably outsource a larger / high-quality print of the scoops first (my printer is limited in the print volume) so there are fewer seams to deal with. It is my understanding that carbon fiber bonds well with ASA, and would add less than a millimeter of thickness. This allows me to have an inner shell that is more complex – as opposed to doing the entire scoop in CF, kevlar, fiberglass, etc.. I need to do more research on this and maybe a few tests. I think those sausage filters are going to be on for a while... Misc. (PDM) The cabin PDM project is mostly finished. All fused accessories and most of the relays now run off the PDM. Was it necessary? Probably not, but I do appreciate the configurability, observability, and the robustness. An added bonus is that since I am pulling from the shunt, the ammeter still works. It's also easily reversible as the only substantive changes were done in the footwell and glove box. It's all hidden, so the cabin still feels completely stock. With most of these projects, I tried to follow the "POLA" principle (Principle of Least Astonishment). Hopefully if/when someone sees this work, they'll agree. Misc (AC / Cooling) I'm also making progress on two other projects; which I may do at the same time since they're somewhat related: 1. Electronic AC. I ordered the compressor; which is made to order (takes a few weeks). This will go in the frunk; which removes the rear compressor weight, and shortens the hose routing. I'll need a larger alternator; which will likely be a modern six phase / denso style. Probably the 175A option. I'll upgrade the alternator b+/gnd to larger motorsport battery cables as well. Not sure if I'll do both alternators, or just the one that's easier to get to. My hope is to retain all other bits of the AC system to keep things less complicated. We'll see how that works out. 2. Ron Davis radiator and fans. I've spoken to Ron a few times, and I'm likely to pull the trigger on this. He has drawings for the BBi radiator already; which we confirmed are correct over the phone (he had it in his notes as an early 80s "Ferrari 512i"). The total depth is the same as the stock setup (about 7.5" / 190mm). There'd be two 13" SPAL fans fully shrouded that take about 22A each. So I'd be going from 3 (essentially un-shrouded fans) with a total of 25A (7A + 11A + 7A) to 2 fans with a total of 44A. Ron assures that cooling effectiveness is for the most part directly proportionate to current (provided proper shrouding). I have 60A reserved for the fans coming from the PDM, so the higher draw is easy to accommodate (but might further benefit from the larger alternator). I know many feel the stock radiator is adequate, but that hasn't been my experience. The combination of pushing the car hard + Miami heat + Miami traffic leads to a slow overheat. Also, I am considering an auxiliary electric water pump when I do the radiator swap. This would allow me to move water at idle and after shutdown (limiting heat soak). This would be controlled via the ECU and work alongside the mechanical setup. Still deciding if this is truly necessary, but it's not too expensive, and it's easy to add if you already have an ECU that can PWM based on criteria (eg, coolant temp, RPM, throttle position, etc.). I'd only run the pump when the thermostat is open, and at idle (or after shutdown). Moving forward, I'll share progress on dedicated threads so this one stays EFI/ITB focused.
Status update. PDM-to-ECU comms Now that the PDM is nearly finished (needs some wire tidying), I'd like to connect the ECU and PDM via CAN. Up till now, all of the EFI / ITB work was isolated to the engine bay. I really only need a few wires: an ethernet cable for diagnostics and a CAN bus (typically, two stranded 22 awg yellow and green). The CAN bus wiring will allow the ECU in the engine bay to communicate with the PDM in the footwell (eg, the PDM can read sensor data from the ECU, and the ECU can read output current data from the PDM). Cabin / Firewall access My original thought was to sneak the wires through the main firewall hole directly above the tunnel, following the center console to the footwell (just like the factory wires do). Yeah, that's not happening. Ferrari left zero room for any wires other than the factory bundle. Luckily, I had taken the seats out to dismantle the center console (pursuing the initial option), and I noticed a dead speaker wire that went to a 35mm grommet in the rear of the cabin. Not sure what it was doing for that speaker wire, but it works really well for my use-case. It's near the LH fuel pump, so I can run the wires over the wheel well, down towards the pump, then up into cabin. So I lucked out. Scoops I had family visiting this week, and they stayed in the room with the 3D printer. So not much progress was made on that. I'll be getting back on that soon. Dyno I definitely want to see where the ITB setup got me. I think there's a real possibility that the intake restriction at the head is still the bottleneck, and that power will not have increased considerably over the EFI upgrades, but I am curious to find out. Even if power isn't up, the aesthetic, sound, throttle response, and weight savings were totally worth it. Plus, it unblocks the engine internal upgrades (cams and higher compression). My plan is to do the radiator swap first, then dyno the car. Keeping cars cool while doing stationary hard pulls in a Miami Summer is tricky even for cars that have proper cooling setups.
Good update! They say to really feel a power difference you need a good 25-30hp more to know anything happened. Heck I could drive a well-tuned early carb 308 and compare it to a good 328 and can barely tell the difference, but I know the 328 must be moving a touch quicker, it's just tough to tell. You're surely flowing a good bit more air with ITBs!
The car *feels* noticeably quicker. The problem is that I don't know if it's due to more power, or just better throttle response from the non-linear rush of air you get with ITBs. I guess there's really only one way to know for sure!
Get it on the dyno. Like I said, my 82 did not make the claimed 340 crank hp on a Mustang dyno and I think we even used 18% as a factor to get crank hp. I left the dyno shop with an upside down smile.
More/better throttle response means more power in my book. The engine has more vacuum hitting the intake port due to a stronger more positive signal. This is really true with carburetors, you can feel more compression under your foot in how the engine responds, the pedal is a bit stiffer and just more responsive overall. Small 911 engines going to twin plug, high compression, and a set of hot cams make those things have a lightning response with how light the crank already is. Can't imagine how heavy a boxer crank is, then again it does have 12 pots moving the thing around.
Status update. I've mostly been experimenting with the scoops. There are a few ways to go about this, but I am leaning towards plastic (ASA). I may wrap / skin in carbon fiber (then paint satin black). It's not for the look of CF, but rather the low weight and added strength. I've tried a few experiments, and each takes a day or two. Not to mention the multiple days it takes to print all 12 pieces. So it's a bit of a slog. A few findings: - ASA (similar to ABS) seems to a good plastic to use for this. - My printer is limited in size, so I have to bond the pieces together. - The prints can sometimes warp a little which makes bonding tricky. I can have them printed as one piece, but that's quite expensive. - I've tried a few bonding / strengthening agents, but the top contenders were super glue, ABS cement (solvent), plastic welding, and epoxy resin. I actually installed the prototype scoop and went for a spirited drive. A few notes: - Intake temps were between 15C and 20C lower (compared to no scoops), and that's without any heat shields around the trumpet inlets. - I did not get the sense that I was getting hot radiator air. - The sound is slightly louder above 4k RPM. The trumpets essentially reverberate in the scoop. I'm not complaining! - The ABS cement alone was not strong enough to survive being thrashed around (speed bumps, heat, closing the lid, etc.). - The plastic welding works pretty well. - The plastic welding with resin layer works best (very strong, no cracks). Although it's slow going, I'm feeling pretty good about where it's ending up and it seems the scoops actually do their job! I need to adjust the dimensions a little bit and do some more bonding experiments. For now, I think the plan is to print the pieces in ASA, plastic weld, then coat in epoxy resin before painting. I haven't decided on whether to add a carbon fiber layer. Attached are a couple pics of the prototype. It's just resting on the lid (not on the bracket); which is why it's not floating like the factory one. Don't judge too harshly, it's just for various experiments to see what breaks, and for me to practice the plastic welding, etc.. It's literally trash. Image Unavailable, Please Login Image Unavailable, Please Login
Quick update. I added a plate that seals the scoop to the trumpets when the lid is closed. The intake temp sensor showed 40C while driving. Ambient is currently 32C in Miami. So only 8C above ambient! That's a clear winner of all the options I've tried so far. Sitting in traffic, the temps got up to 55C, but quickly came back down once moving. While the look of the scoops may not be everyone's cup of tea, it's the best I've found so far for lowering intake temps. Personally, the look has grown on me. Also, rear visibility isn't as bad as I thought it would be.
where you meassure the temperature? and with what kind of sensor? you meassure on all 12 trumpets or only at the inlets of the scoops?
I used a RaceGrade (Motec) air temp sensor; which sends the signal to the Motec ECU. The temp (IAT) is actually part of the fuel calculation. I placed the sensor between cylinder 1 and cylinder 2; right next to the trumpet inlet. Prior to the ITB install, I had the same sensor installed in the plenum where the CIS cold start injector was. You can see the hole for the sensor in the attached pic. Image Unavailable, Please Login The design is loosely inspired by the BBLM (of course). I have steel mesh arriving today; which has 69% "open space". I plan on installing across the trumpets as a precaution now that the ITG filter is removed. It won't protect against dust, but that's a problem for another day. I have some ideas.
thank you for explaining. it would be interesting to know the air temperature at both front cylinders and both rear cylinders. I think there is a difference from some degrees? the air temperature sensor is a PT 100, PT 1000 or a thermocouple or a normal NTC? in my competiton I have mostly thermocouples, they react very fast. the black air hose is for cooling the generator?
I could try it. NTC. That pic is of a BBLM (35525). I'm not sure, but I think it's an air hose for an alternator that was moved to where the AC compressor would have been. Maybe they swapped the two lower hanging alternators for a single one mounted higher?
Status update. I've been doing more experiments / prototypes for the scoops and intakes. Reiterating the above, this is repeatedly the coldest air I've been able to get on any of the locations I've tried, and it doesn't seem to be impacted by hot radiator air. It also helps that the aesthetics of the scoops have grown on me, and they slightly amplify the sound of the ITBs. For the v2 scoop, I stuck with ASA and simply "welded" the parts together and rough sanded. These prototypes are throw-away, so I don't want to spend hours and hours hiding seams. Mostly checking for fitment and durability. For v3, I need to make some minor tweaks, and I'd like to incorporate a middle spine that should provide some rigidity. A later design (v4?) will tweak the design to mount an air filter as the air passes from the "dirty side" to the clean side. Image Unavailable, Please Login Image Unavailable, Please Login I 3D printed a modified trumpet design (still 150mm), as well as the top plates / shield. Even though it's only gray plastic, I like the direction it's going in. I have two test trumpets being made, one in Ultem 1010, and one in aluminum. Ultem 1010 is very expensive, but it has a kind of early 70s translucent fiberglass trumpet look (eg, 512M). Curious how that'll look, even though I've never seen any BBLM with fiberglass trumpets. I'll have the top plate / shield made in aluminum. The top plate seals (using that term loosely) up with the scoop bracket. Similar to the BBLM, I have mesh across the tops of the trumpets. I have new brackets being cut / bent. The ones shown in the pic are the ones used for the ITG filters (hence the holes). I widened the primary opening as well as added mounting holes for the scoops (they mount from inside the scoop). Image Unavailable, Please Login Image Unavailable, Please Login
Looks good, why not make trumpets out of plain fiberglass ,like early Ferrari race car stuff? Supposedly John Surtees introduced GRP to Ferrari racing team in the early 60s.
Mostly because I'm just fascinated by Ultem 1010, and I'm looking for an excuse to use it on something. It's also easier to have one trumpet printed in Ultem than it is to create one in fiberglass to see if I like the look. Image Unavailable, Please Login Ferrari definitely used materials other than aluminum for their velocity stacks. Here's a pic of the 512 S. Image Unavailable, Please Login It's mostly just an experiment. I'm assuming I'll go with aluminum at the end of the day.
Quick update on the scoops. I printed v3 in ABS-GF (bits of embedded fiberglass); which replaced v1 on the RH side (I'm alternating sides for each iteration). I came across ABS-GF after doing some research on filaments for larger prints. ABS-GF seems to be recommended for larger parts as it doesn't deform / warp. After about a week of printing, I can confirm that it works much better than pure ASA. The parts required very little work to line up. The lack of warping combined with the "spine" seems to have solved the butt cheek issue I was getting in the rear (solvable with bondo, but I wanted to avoid that if I could). I am also going to try ASA-CF. It's a relatively new filament, but I am curious if it'll be even better than the ABS-GF. I am in the process of printing another v3 for the LH side using ASA-CF, and I will compare the two. A full length aluminum "spine" should arrive from SendCutSend tomorrow. The ones shown were printed in ABS-GF. Depending on how the ASA-CF version comes out, I may convert these to real parts with proper sanding / paint. Image Unavailable, Please Login Image Unavailable, Please Login Image Unavailable, Please Login Image Unavailable, Please Login Image Unavailable, Please Login
A bit late for this comment probably but I would integrate the scoops into the spoiler. (Alettone on diagram)
I considered something like that, but ultimately decided to go with something more reminiscent of the early BBLM (pre "silhouette" rules) and Koenig cars. Basically, the scoops are Series 1 BBLM/ Koenig, the ITBs are Series 2 / Series 3 BBLM inspired, and I'm trying to marry the two in a clean way. I tried to retain the floating look of the intake lids, but also the tapered lip around the bottom. A lot of the Koenig scoops I've seen skipped that, and I don't think it looks as nice without it. Image Unavailable, Please Login Image Unavailable, Please Login Image Unavailable, Please Login
