AndyCam

One problem with flying low is that other drivers may not see your brake lights, or only be looking for a brake light in the center of the vehicle. Not wanting to get rear ended by a truck or SUV on my recent trip to Spa Francochamps I headed off to the workshop to have a think. There are plenty of high level brake lights available, but none matched my requirements: As high as possible on the vehicle Lightweight Very bright when lit, but unobtrusive when not Looks like it was designed to be there InexpensiveSo I designed my own. My starting point was looking for LED motorcycle tail lamps. These typically have 19 or more bright LEDs, are low weight, inexpensive and available on eBay. There are various elliptical shaped lights, some with black or red lens, "E" marked or not. I paid £11 for a 108mm wide red lens "E" marked unit, but cheaper versions are available. Next step was to design the casings. Fortunately I have access to a 3D printer in the US office and Onshape, a 3D CAD application, so I set to work creating a clamshell casing that fits on the underside of the rollbar. I wasn't sure if I wanted to use adhesive tape to attach the light or bolt through the petty-strut holes, so I designed variants for each. This is the 'bolt on' version which has 'ears' that curl around the rollbar and recessed holes for the nuts (bar not shown): The 'Tape on" variant just has the ears removed. Both variants have 4mm Dia hole in the top casing for the wiring and the bottom casing attaches using 3 self-tapping cross heads. Next up was the printing stage which was done on a Makerbot FDM printer by exporting an STL file out of Onshape. For those not experienced in 3D printing, its not like sending a document to your inkjet - it takes a many hours to get a decent quality 3D print. But as I needed to run to the airport for my return trip I used some coarse settings for a fast print within 2 hours. The result wasn't as smooth as I'd like, but good enough for a first prototype, so I wired it up using the Caterham high level brake loom and some 1mm twin core black cable. For more details how to make this all neat and tidy see Daniel French's excellent blog post, besides his car is faster and cleaner than mine. The result was light and bright! I did weigh it but I cannot remember exactly what it was now, something close to 100g IIRC. So for a little over £20 I have a neat little unit that meets most of my requirements.Best still you can have one too. You can take a look at the 3D model here, you just need Google Chrome or Firefox browser and an internet connection. No charge. Better still, in the spirit of the Open Source movement, grab a copy, refine it and share it with others. As this was a prototype I have since made some changes. I have modified the strengthening webs inside the casings to give more space for the wires, moved the wire holes for a smoother transition out of the casings and reduced the wall thickness a little (another few grams saved). The next prototype will be on the finest settings, though I might also print one on the Formabs too. I am also working on a variant for a 120mm wide LED unit which now appear to be more numerous and cheaper on eBay. If you have any other suggestions or comments let me know.

The "red" one is a cross head on mine not a rivet. It needs to be removed to get the knee panel out.

It is a can of worms, the variability of coolant temperature between different cars could be due to many factors. From experience mine only exhibited this problem after a drilled stat was installed and the rad was flushed out. So at a risk of post hoc ergo propter hoc, either my rad now works better or the flow through the thermostat has an undesirable side effect, or both. I had the coolant temp drop with 88 and 82 rated stats, though the latter is less marked. I found this article which may give a clue related to the K series in the Elise "A common attempt to improve things is to drill one or more small holes in the original thermostat outer ring to allow a small flow through even when closed. As it works as a small extra bypass it does reduce the dynamic issue, while the thermostat will adjust to maintain the static issue as is. It also needs to be said that the drilled thermostat does increase a little the flow with the stat closed (which is good) but it also cools the bulb, keeping the thermostat closed for much longer (which is very bad)." So maybe you can make it better but not solve the problem completely?

This all sounds familiar. As mentioned in multiple threads, airlocks cause symptoms of overheating and the bottom hose doesn't get warm. The methods above will work, but also I'd also suggest blowing down the "jiggle" ball-value tube on the inlet manifold (insert drinking straw) just to double check its not crudded up. Do the same with its hose into the header. As for the overcooling I'm running the stock rad on a 1.4k classic (plastic capped rad) and after flushing the rad and installing new stat & sender the temp would drop significantly when the car was driven hard. Appears that the rad is just too efficient now. Solution is blanking 1/3rd of the rad with cardboard. Hat tip to Chris at 7 Workshop for that. Happy all day at 80 now.