LSR Karmann Ghia Blog, September 12, 2009

The following was sent to the Karmann Ghia Club of North America's email list on September 12th, 2009:

Hello Everyone,

I awoke with a start at about 2:30 AM today to the sound of rain on the roof and darted outside to protect our Land Speed Record Entry from the unexpected shower - the forecast said 10% chance! Pieces were "everywhere" - and tools - but thankfully, there's lots of cover (sheds, awning, etc) for such circumstances...

And now it's raining, so I'm taking a moment to update you on our progress.

The last update was on Sept 5. Since then there's been so much happening I'm sure I'll miss quite a few things. On the 5th, Britt Grannis was here helping with the paint - that thorn in my side if ever there was one. Earlier, Dave Whitaker had helped prepare the paint to fix some serious runs but the paint was just too thin. So Britt sanded the whole car down, right through various layers of paint to "straighten" the body and help prevent the paint from ending up very thick in preparation for yet another laying on of paint. The paint was shot on Saturday, the 5th, very early in the AM, as I recall. The next morning, I found there were some details along the rockers and under the nose, and one spot where an air hose had touched the car, and so I then added a second coat to Britt's one - making two coats (pretty typical). Thankfully, this time the painting effort worked out - there will still be some polishing to do, but no new paint will have to be shot.

I then turned my attention to getting the engine finished and tuned. On the 4th, I had finished making the tin, so now there were only three areas left that needed work: the exhaust system, the fuel line, and the throttle linkage.

The exhaust system work was straight forward if time consuming. The most time consuming part was working on the flanges that bolt to the heads. Each must be flat, transition from round tubing to a rectangular port on the head, and the pipe must be fitted properly both in terms of length / depth and rotation such that there is absolutely zero stress on the parts when installed. The engine grows wider as it warms, after all, and I wasn't fitting the exhaust with a hot engine, so any stresses in the parts when installed cold tend to tear the exhaust apart when warm unless lucky enough that they are relieved as the engine grows wider. ... Another time consuming factor was the installation of the forward end of the heater boxes on the J-tubes. (I have put images up on the web site, showing the manufacture of the J-tubes.)

On the 6th, I spent a lot of time working on the fuel line and linkage, and finished them the next day. The fuel line design was inspired by the stock Denzel fuel line, the Okrasa fuel line and the Porsche design, too: All three use all-metal lines from pump to both carburetors and all use "banjo" fittings to attach to the carburetors. The original Denzel fuel line I had on hand was designed for the Solex 32 PBIC or PBJ which have the banjo on the forward side - near the firewall. The Solex 40PIIs that this Denzel runs have their banjos fitted on the outside, opposite the fan housing, just as the Porsche 356 does. Further, as we had chosen a Porsche fan housing with its "over the top" throttle linkage, our strategy should be similar in design as it has the same issues of clearance. The Porsche 356 / 912 system differs in a few ways, however, importantly including the height of the carburetor (the Denzel's carbs sit lower) and both the position of the fuel pump's outlet and its attachment strategy.

Work progressed starting from the fuel pump. The keys on the first leg were the height at which to become horizontal and I chose a two stepped strategy with a roughly 45 degree intermediate bend. The height chosen was effectively the surface level of the fuel in the float bowl; this would be below the bottom of the banjo fittings by a tiny bit, hopefully enough to prevent fuel from running out of the line whenever a banjo is removed. Next was the rather simple matter of "turning the corner" to avoid the throttle linkage's range of motion - I say simple, but really I was just lucky as I had planned to make that corner as complex as necessary, including a possible dogleg, to avoid the throttle but it turned out that just a simple, gentle enough radius was all that was needed.

I then used some bits salvaged from several abandoned Porsche 356 fuel lines, including a round junction block and two banjos. However, the height difference prevented me from using more of the Porsche parts. I did both the right and left simultaneously so they'd fit perfectly. First, I studied the 356 design to see how it is supposed to clearance the throttle linkage and be supported from vibration. The Okrasa and Denzel designs do nothing of the kind - they both use a through-the-fan-shroud throttle technique and have nothing to clearance in the back and both are therefore rather haphazard in their placement - and obstructing the fan intake was not a concern for either. (It is, however, for me!) From the Porsche I observed that there's a chance to support the line from one of the two M6 bolts that project through the fan shroud to support the oil filter on the opposite side. The key observation was the depth of the metal line's protrusion through a flexible portion of hose that handles some of the expansion and contraction of the engine and prevents vibration induced cracking. In short, they have a lot of hose with only a little gap, and the clamp to the fan shroud goes directly on the hose which is supported by metal inside - thus, no rattles, no extra grommet, etc. Nice design. I copied it!

This made a design including four metal lines, two banjos, one 8mm ferrule fitting, one cloth braided flexible hose section, two metal caps that terminate the flexible hose, one round, flat chunk of metal that serves as a junction block, and a sheet metal strip with two holes that constitute a clamp for the line to the fan shroud.

Once satisfied with the fit, I soldered the metal lines to their respective locations (both banjos and three metal lines to the junction block). I used silver/tin alloy. It seemed to go quite well and fit was perfect, including the fiber washers used on the banjos - another no-stress-in-the-system situation.

Meanwhile, after great frustration trying to select the proper spark plug, I called Bosch's technical assistance line and spent a lot of time with one of their techs. He worked out from their computer what the most likely spark plug would be and he offered that they don't have this engine in their computer, and that they want to include _every_ internal combustion engine in their database, so they are willing to loan me use of an incredibly expensive tool to determine an optimum spark plug! For this, the fellow hooked me up with four groups in their company, including the racing team, Research and Development, "Motorsports" and a distribution branch, all in search of one of their several instances of this tool to determine how most efficiently we could get the tool and my engine together to determine that best plug! Talk About Support! AND, they offered to sponsor our LSR run! Unfortunately, there was no time for their bureaucracy to do its thing for sponsorship this year. ...Anyway, I did get a recommendation for the perfect plug.

Next up, the throttle linkage. I love the Porsche over-the-top solution because it's elegant, efficient, reliable, easy to service, and is very out-of-the-way for accessing other engine elements for service purposes. It also resolves the engine expansion issue so both carbs remain in sync no matter the temperature of the engine. However, hooking this to the VW throttle linkage is always an annoyance. I've made such solutions on quite a few occasions and am aware of the shortcomings of all of them, and all of them require some modification of the forward (firewall) tin. Not wanting to repeat that work again, I decided to design a new, better solution.

After much consternation and thought, I ended up making a new bellcrank that replaces the stock Porsche firewall side bellcrank (that mounts to the left side of the fan shroud), yet retain the stock VW throttle passage through the forward shroud. To accomplish this, I created a support on the right-hand side, at the crankcase center, for the very long shaft of this bellcrank. I had a lot of trouble finding 6mm rod / dowel to use for the fulcrum shaft of this bellcrank, and ended up using something smaller - 5.5mm. The Porsche end of the bell crank was easy - copy the stock 356 fitting. The other end, however, required some ingenuity. One issue is that it has to be able to be disassembled, so no welding it in place! This means that this arm of the bellcrank has to be removable yet must also have a very good rotational position attachment as any shift in rotation would ruin the driver's control of the engine. I took a piece of 5.5mm rod and, using the mill, cut two flat faces, parallel to each other, on opposite sides of the shaft, to a dimension that left 4mm. I then got the rotation right by doing the welding of the Porsche side only when both arms were ready and in their center of motion positions. For the VW side's arm, I welded a bit of standard VW barrel nut-based throttle linkage (standard on every Ghia starting sometime in the 40hp era), to another linkage part I found in my VW parts bin and now this arm connects via a sturdy M6 bolt with nut that tightens the arm on to the shaft - rotation is fixed via the parallel flats, so the clamp only keeps it in position from right to left and at that, there's not a lot of side-loading to worry about, so it's a robust assembly. The real key, though, was to fabricate a support for this shaft at the center of the engine!

The result was a very natural looking, very effective throttle linkage design that I'll probably use on all my Porsche-to-VW applications in the future. (Photos on the web site.)

Meanwhile, enthusiast Aaron Castillo joined the team. His first actions were to secure the racing fuel we need and the spark plugs that Bosch had recommended - and a few other unsundries that helped keep me working and not out hunting down parts at some store. He then took over the blasting cabinet, preparing five rims for us. It took him _days_ to clean off the rims!

Eventually, the engine was essentially finished - none too soon - but not quite done, either, just ready for testing. The next day I moved the engine to the dynamometer as I obviously wanted to start it. Not so fast! The full-flow oil system had to be provided for. And, I haven't used the dyno much yet - just got it going only months before - and so I had to make wiring leads for the tachometer, oil temperature, and so forth. Both Chris Morrice and Aaron Castillo happened to be on hand to help me mount the forward shroud, and it took some doing. Someday I'll make some modifications to it to make mounting it a lot easier, but no time for that now! I then routed the fuel pump inlet line and performed other basic tasks to get it ready. Late in the day I thought maybe I had it ready, and anxious to get it started, I hooked up the fuel and turned on the fuel pump - carburetor sight-glass installed to set the float level - and found my 6v fuel pump had died! DAMN. I then tried my 12v electric pump and it promptly overflowed the carbs - way too much pressure, though I don't understand how it got by the regulator... And so, it became the next day.

The next AM, I ignored the fuel pump issue for the moment and tried cranking the engine, but it refused to even turn over! Long story short, it turned out to be two problems: weak battery and a very slight interference fit of the crankshaft pulley on the crankcase. I resolved the latter by removing the pulley and then installing it a little at a time, repeatedly hitting the starter to turn the engine and thereby slightly machine the pulley into place, and then pushing the pulley on a little deeper and repeating the process! This took _hours_ - it worked but took a huge toll on both me and the starter, not to mention the battery...

It was now September 9th, sister's birthday, and I had to get the car registered. Previously I had spent many hours looking for the paperwork and I could not find it _anywhere!_ So I headed off to the DMV - just up the street. I had the bad luck that the official who came out to do the verification was a bit of an ahole and had a serious attitude problem and rudely refused to do verification of vin! The lady said, "Go to the CHP - they verify shells." ... Grrr.... ....I went home, found the necessary forms on the DMV web site, printed them, and showed up at the CHP site. The first guy said I had to have an appointment and they were booked up for the next FOUR days! -Gasp!- I stammered, appointment for a 10 minute form?! The officer said no, it takes about 45 minutes or more. For what? To install a NEW VIN, of course! WHAT!

-heavy-sigh-

I calmly indicated that there must be some mistake - that's not the procedure the DMV asked for, just to verify the vin that's already on the car. He wouldn't deal with it, but, after I gestured that it was just the little Karmann Ghia visible through his window, and showed that I had the forms already, his office-mate said he'd take a look. This guy didn't want to do it either, but after much gentle persuasion, within about an hour total, I walked out with my completed form. ...Then, to the DMV - a different one! Huge line, but friendly people! Just as my turn was bout to happen, a female rent-a-cop came slowly wandering through, looking like she was lost. I was in conversation with two others and mentioned that it was unusual, and interrupted our conversation to ask the security officer what she was doing. ...It turned out that she was looking for _me!_ Why? Someone complained that while I was not parked in an obviously illegal spot, someone in a too-large vehicle didn't like having to seesaw their way out of their parking place to avoid Babe and the Ghia! She said, well, they were going to tow you, but I thought I'd just try and find you and see if we can find a new parking place for you! GREAT! Longer story shorter, she actually found me a new parking place and I just barely made it back in time to have my number called - and then I flew through the reg process and returned with new plates and tags! I _immediately_ called my insurance company and got the insurance turned on!

Next, Aaron and I completed the painting of the rims and went to lunch, after which we went together to get the tires. They're rated to 200 mph! Getting them mounted and balanced took the afternoon. And I helped Aaron with a few other things, too as he began work on the car while I worked on the engine. That evening, if I recall correctly, Chris Morrice came over and remove the rear bumper from another Karmann Ghia and installed it on our LSR Ghia. I don't recall all the days events but it was a full day.

The next day, Larry Edson came over, though he had only one hand functional due to surgery to relieve a symptom related to carpal tunnel. Over coffee, we designed the oil filter / cooler system and came up with a parts list and he and another enthusiast named Beate went off to get me a helmet and source the oil system's fittings. Meanwhile, I installed the headliner and then turned my attention to the pesky engine. STILL hadn't started it!

I realized the fuel pump issue was really a dead-battery issue, so I quickly remedied that problem, and I immediately found out that the float level on the left carburetor was _perfect._ However, I also immediately found out that that carb's banjo leaked from the metal line to banjo fitting. Damn. I tried to fix it about four times, but as it had been filled with fuel, it was a problem to solder! After each unsuccessful attempt, the line had to be drained and dried - and some fuel fumes remain for a LONG time.

Meanwhile, I got the oil system full of oil, primed the oil pump, and the gauges told me we had lots of oil pressure - that, and the puddle beneath the engine! I think the Denzel's design for the oil system's return of oil to the crankcase is just plain dumb: an M12 X 1.0 thread into a magnesium crankcase by a steel part! This puts the threads at an incredible risk every time the oil system's banjo is tightened - and that's how you're supposed to cure leaks! Instead, after asking some friends for engineering advice, I ended up stumbling across a great solution by changing the seal from a copper crush washer (which in fact does just about nothing as the magnesium is of similar hardness as the copper!) to an aluminum crush washer of larger internal diameter with a very small diameter nitrile O-ring fitted inside that ID.

[EDITORIAL NOTE: The fixes cited above didn't fix the oil leak as the banjo fitting continued to leak oil and nearly caused the loss of three runs at Bonneville! Later, a cause and solution were found, but only after this year's 36hp Challenge at Bonneville was over.]

...Somewhere along the way, the starter failed and I ended up spending something like four hours trying to solve it. I was using a special 6v self-supporting starter ("auto-stickshift") - a VERY rare beast. I had selected it because both the Dyno and the target transaxle case take 12v starter bushings, and because the autostick starter has more starting torque... A very long story short, very late in the day I discovered another problem: The 6v and 12v autostickshift starters have the same pinion gear diameter and tooth-count, thus possibly explaining the otherwise unexplained metal shavings found in the starter hole! This led to the discovery that all Ghia starters have nine teeth, and the only difference is tooth-depth and total diameter, with the standard NON-self-supporting starter being the odd man out and the ONLY one that actually works with a 109 toothed ring gear! Oy. Learning that HURT in terms of time. Hurt bad.

I did manage to get some more things done, in particular work on the transaxle, but it was not really enough - and so it became yesterday, the 11th - the day I had planned to leave!

Early in the AM, I put a tap through the 12v starter bushing and, taking Ron Jone's (of Ron's Transaxles) advice, I installed a standard 6v starter in a 12V bell housing WITHOUT ANY bushing! This worked! I ignored the fuel leak and set the right hand carburetor's float level. Beate took the volute and went off in search of four balls that would just fit snugly inside the volutes to keep foreign material out during transport to Bonneville. Aaron installed the front bumper and found a working horn - after testing something like 15 of them! Chris installed the door guts on the driver's door and much of the passenger door...

And I, meanwhile, got the engine running! It only took a little ignition timing and it started right up and purred like a kitten! Push the throttle, though, and it roared like the Lion King, head of the Pride! Oh My Gawd is it loud! We're talking an estimated 120 dB - maybe more! - when the throttles are opened!

Unfortunately, _both_ my tachometers failed - they show a reasonable idle speed but ridiculous numbers once the throttles open, swinging up to 7 grand and down to 1 grand in a matter of moments when the engine itself was decidedly stable at some intermediate speed, like, say 4 grand...

Oh well. I swapped out the breathing valve cover for a standard one so that my breather's valve will help actually provide a slightly _negative_ crankcase pressure! (I'll only use the breather cover if there's any sign of excessive pressure.) And I completely redid the left hand carb's banjo fitting and this time, no leak!

Somewhere in there I also spent two hours or so on the transaxle. I probably only have another two hours to go on it.

So, the rain has stopped - what the hell am I doing writing this damned update for! I've got work to do!

Oh, and not forgetting again! Here's the URL:

https://ghiacoachworks.com/share/denzel_lsr.html

Regards,
Richard

P.S. There's a video of the engine running - with audio - on the site...
RT

Contact Us