Page 3 of 3

Re: Radiator swirlpot / header question

PostPosted: Fri Jan 29, 2010 12:29 am
by garyeanderson
gordont wrote:Thanks for the feedback, have to laugh though at some of the answers and dialogue......

I will post a pic when the header is in, I never got the answer as to the size of the TTR ones but looks like from Christians pics it will fit fine.


PS - as a though does anyone in the UK ever have a problem keeping an elan cool? surely it never gets hot enough!

I guess I read the question and I don't have an answer. You don't say what the problem is in your first post that started this topic that you looking to fix. Do you have the stock S2 radiator? Did it boilover? What was the temp guage reading? Do you know if the guage is accurate? I guess your question was "will the swirl pot fit" but you left us to guess why. My friend Rob in Southern California S2 with the stock S2 radiator that will barely get to temp in normal weather and at 100+ F (38 C) won't bust 195, in traffic 200F (93 C) so there is some info to compare your S2 with. My old S4 with the wide S3 triumph radiator would scare me in hot weather (100F) but never boiled over on me, 100 to 105C on the guage. In traffic sometimes it was hotter on a 90F day, if it was any hotter I left it home as I didn't like to have the heater on. I guess like most ever other post that I read, folks want answers that we can't give because they don't give the parameters to the problems. Everyone knows what there problems are but they don't explain them so everyone shoots at a problem and needs to make assumptions so the data is faulty. If you asked I have (13 in) 325mm between the gooseneck and the radiator someone could answer that in a post or two. I don't know how long "quite close" is.

My friend bill Gavin wrote this up a while ago and its over on Elanmods with photos too

here is the link ... 0Radiator/

join and I'll approve the membership.


"When I bought my S2 Elan it came as a semi-basket case, so I was able to inspect most of the parts before installing them on the car. Cooling issues are always a concern for a Lotus owner, so I checked the radiator and water jackets very carefully. The engine's water passages looked clean, and the radiator, an original S2 unit, looked like it had just come back from the radiator shop, so I installed them, expecting to have an efficient cooling system. As I put some miles on the car, however, I found that the cooling system was not adequate to the job. Given the consequences of a severe overheat in an Elan, I decided to replace the radiator with a new unit. Influencing my decision was the size and weight of the S2 radiator, and the fact that it probably couldn't be replaced in the event of an accident. I looked into several radiators, but decided on Tony Thompson's replica 26R radiator and header tank. Griffin, Summit, and others make fine units, but I wanted something made for the Elan, not something I would not have to adapt and debug.

After I had the new system, but before I installed it, I noticed an all-too-familiar oily haze on the windshield - anti freeze being blown around through the defroster vents. I removed the heater core and took it to my local radiator shop, who confirmed that I had a leak; it was what they call a "champagne leak", a fine crack in a soldered seam, too fine to allow a steady flow of coolant, but enough to coat the windshield, and to allow air to be sucked into the system when the engine cools. Fixing this leak solved most of the problems; the engine temperature stayed in bounds, but there was a small but steady loss of coolant. Since I had a new system on the shelf, it didn't make sense to debug the old system, so I decided it was time to install the new radiator.

I was expecting this to be a simple bolt-in job, but Lotus jobs are never that simple! The TTR radiator was intended to mount to the fiberglass 'tunnel' over the steering rack, similarly to the original radiator, but had mounts in somewhat different locations, and had threaded bosses rather than the studs of the S2 radiator. The fiberglass in that area of my car had been completely removed, as shown in picture "02_MissingMount", and the previous owner had built a clever but elaborate mounting to the steering rack blocks. The TTR mounting bosses did not line up with this mount, and the holes could not be relocated since they sat directly on top of the steering rack mounting blocks.

After some thought, I noticed that there were alternate mounting holes in the chassis for the steering rack, and these holes were not being used. I decided to build a pair of brackets that would bolt to these holes and support the radiator. The problem of the spacing of the mounting bosses on the new radiator would be solved by bolting a length of 1/2" square tubing to the radiator, then mounting this tube to the new bracket; short spacers set vertical alignment and add some room for standard length bolts. The brackets were a bit trickier than I first thought; there is very little clearance under the rack, so I designed a bracket consisting of a piece of 1/2" steel tubing mounted vertically on the rearmost mounting hole in the chassis, with a shallow horizontal extension that would pass under the rack, and pick up the front mounting hole. A short length of angle on top of the tubing provided a convenient place to attach the tubing on the radiator. Some careful mock-up and measurement determined the dimensions of the parts. For each bracket, I made a threaded bung on the lathe, which I brazed into the bottom end of the round tubing, and a short length of angle, which I brazed to the top. I cut a piece of square tubing on a lengthwise diagonal to provide the extensions to the forward mounting holes. I brazed these to the round tubing, being careful to place the extension at a right angle to the angle on the top of the tubing. A nut brazed to the forward end of the extension finished the bracket, as shown in picture "03_NewMountDetail". When all parts were assembled and aligned, the position of the pieces of angle were marked on the square tubing, holes were drilled, and nutserts installed in the square tubing; the length of the allen bolts that attach the square tubing to the radiator must be carefully matched to the dimensions of the tubing, spacers, and threaded bosses to provide solid support without puncturing the radiator; the heads are contained within the square tubing, so they do not interfere with the mounting brackets. The holes and nutserts for the #10-32 bolts that attach the square tubing to the brackets must be arranged to clear these bolt heads; it takes a bit of work to get dimensions that work for both sides, so the brackets don't need to be handed. A coat of paint and a pair of plastic plugs in the ends of the square tubing finish things off. The #10-32 bolts are inserted from the rear to provide easiest access; I use long ball-end allen wrenches for them. The photos should provide the details. I'm deliberately leaving off any measurements, since there is so much variation in the cars (my car has a Spyder chassis), and since I can't find the sketches I drew before making the brackets. Before committing to cutting and welding, it is necessary to be very sure that the bonnet, fan, and all other parts will clear the radiator and tank.

Once the radiator was in, the expansion tank and hoses need to be fitted. With the radiator positioned over the rear edge of the steering rack, there is just about enough room to fit the tank without any mods. Two short pieces of hose connect the head to the tank, and the tank to the radiator - see photo "TTR Header Tank".

On the road, the new system has proven to be quite effective, to the point that I have had to install a hotter thermostat so the engine will run properly on cool mornings. The system did lose a bit of water after a few cycles of heating and cooling, so I installed a coolant recovery tank, which has solved this problem. I used a one-quart polyethylene milk bottle as a temporary tank, but this has been satisfactory, so it will stay until I find or make something I like better. Two other changes will eventually be made; flaps will be installed to prevent air entering the grille from flowing around the radiator, and an electric cooling fan will be installed. The radiator has no provision for a fan switch, so that issue will have to addressed as well; if I have a bung welded into the radiator for the switch, I'll also have a drain plug installed.

If I were to do this again, I would do it pretty much the same, with one exception. In the tight confines of the Elan engine bay, parts often need to be moved around or removed to allow access to other parts; due to the rigid mounting, the radiator can't be moved without unbolting it from its brackets. Instead of mounting the square tube directly to the top of the brackets, I would use a pair of transverse pivot bolts, one at each bracket. Once the top hose is disconnected the radiator could then pivot forward out of the way without undoing the lower bolts. When the top hose and header tank are installed, the top of the radiator would have adequate support and location. The more flexible mounting should also reducing the chances of failure of the mounting brackets. "

I don't want to reply to every problem folks post, Everyone thinks my attitude sucks and it does but so what, that just the way I am...


Re: Radiator swirlpot / header question

PostPosted: Fri Jan 29, 2010 12:58 am
by msd1107
I had to smile when people maintained you couldn't have air in the system.

My car didn't have a recovery bottle. And natural evaporation would gradually lower the water level until I could see the fins in the radiator. Then it was time to top up.

The way I look at it is that as long as there is water at the inlet to the water pump (from the bottom of the radiator), the water pump will pick up the water and circulate it through the engine, with it returning to the top of the radiator. The water then drains through the radiator, even if there is air at the top.

The only thing air in the system does is gradually decrease the cooling capability of the radiator as the water level is progressively lowered and there is less water flowing through the radiator to get cooled.

If you had a temperature controlled electric water pump and temperature controlled radiator fan, you could get by in low load situations with very little water, just enough to fill the block and head, hoses, and a little for the radiator. Since water is relatively heavy, just think of the weight you could save! Very Chapmanesque!

Every aspect of the Elan's cooling system is very over sized to compensate for the fact that each part is used at a very small percentage of its capability. Assuming the use of a temperature controlled electric water pump, the car would need only a small radiator if it had a temperature controlled fan and proper ducting to an air inlet and ducting to a low pressure area for exhaust.

1968 36/7988

Re: Radiator swirlpot / header question

PostPosted: Fri Jan 29, 2010 1:01 am
by gordont
no worries and thanks Gary


PS - your attitude is fine :)

Re: Radiator swirlpot / header question

PostPosted: Mon Aug 19, 2019 8:50 pm
by LotusEngineeringBuff
Quite an indirect way of confirming what I suspected: This haze I observe forming on the inside of my Plus 2 windscreen is glycol seeping off my heater radiator. Let's hope it's not a nightmare to take the heater box out of the car. I am afraid it is :-(
Thanks anyway

Re: Radiator swirlpot / header question

PostPosted: Tue Aug 20, 2019 4:15 am
by The Veg
LotusEngineeringBuff wrote:Quite an indirect way of confirming what I suspected: This haze I observe forming on the inside of my Plus 2 windscreen is glycol seeping off my heater radiator. Let's hope it's not a nightmare to take the heater box out of the car. I am afraid it is :-(
Thanks anyway

It is as easy -or as difficult- as removing the dashboard, as that must come out to access the heater. Once the dashboard is out of the way, removing the heater is simple.

Re: Radiator swirlpot / header question

PostPosted: Tue Aug 20, 2019 5:12 pm
by mbell
Also need to remove heater hose from engine bay that game the best access.