I've used stop-leak sealers with mixed results in the old beater cars that I owned when I was much younger and poorer. Even with what little I knew at the time, something just didn't feel right about pouring thick brown sludge from a yellow plastic container into the radiator of the only car I had and depended upon.
If you visit today's online automotive forums you'll find many questions about whether these products actually work. Lots of 'experts' respond positively, and an equal number swear they'll cause immediate self-destruction of an engine. The truth is somewhere in between because there are different kinds of leaks and different kinds of stop-leak products.
The most difficult (practically impossible) leaks to fix for any significant amount of time are head gasket leaks, large radiator holes, burst hoses, and leaking water pump seals. Stop-leak products containing long fibers and/or sodium silicate (water glass) may help long enough in the first two cases so one can return home from a trip and make a proper mechanical repair. The only thing they can do for a leaking water pump seal is to clog up its weep hole so the problem is out of sight for a while. And nothing, except maybe a silicone tape wrap, is going to even temporarily repair a burst hose.
Candidates for long term repairs, though, are casting porosity problems and some freeze plug leaks. General Motors, in the nineties, routinely added their own stop-leak product to the coolant in a number of models as they came off the assembly line. GM had issues with casting porosity, and they hadn't yet learned to reliably seal aluminum heads to cast iron blocks. Their ACDelco coolant sealing tablets are still available and continue to be used today by some shops when performing flushes and coolant changes on these older cars. GM eventually terminated their use in new vehicles after developing their long-life DexCool coolant because their sealant tended to shorten the life of their new coolant's anti-corrosion package. Once dissolved in coolant, the tiny organic particles from these tablets continually circulate throughout the engine. The material (ground up walnut shells and ginger root) tends to find its way into small defects and seals them for pretty much the life of the coolant. When time comes for a coolant change, the engine is flushed, and new tablets are added to the new coolant.
These problems weren't limited to U.S. auto manufacturers. The 1990 Jaguar XJ-S Driver's Handbook (page 176) recommends adding two 135 ml bottles of Jaguar Radiator Leak Sealer to the vehicle's radiator after every coolant change, probably for the very same reasons.
This brings me to the Quarter Scale whose coolant leak I have good reason to believe is due to a porosity problem in one of the block or head castings. I researched the stop-leak products currently available so I could make an informed decision before irreversibly dumping someone's snake oil to the Merlin's coolant system.
I decided early on to avoid the common silicate-laden products because of my fear of the glass particles eroding the numerous tiny o-ring seals in the engine. As an aside, coolants themselves, unless labeled 'silicate free', may contain their own silicates whose purpose is to scrub the walls of the engine's coolant passages. The owner's manual for my Honda Valkyrie specifically warns that silicate-free coolant must be used or the water pump's factory warranty will be voided.
The second group of sealants that I eliminated from consideration were the products containing copper flakes. Although the manufacturers of these products claim they can be safely used in engines with aluminum heads and radiators, this doesn't make a lot of sense to me. I was careful to use only aluminum and stainless steel materials in the Quarter Scale's coolant system in order to avoid issues with galvanic corrosion. Using a stop-leak product that stuffs copper particles into a hole in an aluminum head in the presence of an eventual electrolyte would seem to guarantee that the hole will eventually grow larger. In addition, I didn't like the idea of continually circulating metal particles past the seals in the Quarter Scale, and especially around the one in its tiny water pump. So, I eliminated the aluminum particle sealers as well. I also crossed off my list the sludgy-looking fibrous sealers, just because. And so at this point I had eliminated all the stop-leak products available at my local auto parts stores.
The two products that I did consider had to be ordered online. The first was claimed by its marketing to be a polymer only product. It's a (very) expensive product sold under the name Titan Block Stop Leak on the Titan Sealer website and also under the name Dura-Seal Engine Block Sealer on the Dura-Seal website. Curiously, both websites look very similar and have exactly the same contact information. The second product was the inexpensive ACDelco sealant tablets described earlier. These are available through Amazon as well as several other online sources.
My own testing included the inspection of the sealers under a microscope after samples had been mixed in their proper proportions with coolant. I wanted to be sure that most of the particles were small enough to seal a .020" hole, and that there were an insignificant number that might be large enough to block the tiny coolant tubes in the Quarter Scale's radiators. Both sealants passed this test.
Next, I measured the time it took for a 100ml mixed sample to plug a leaking paper cup whose bottom had been perforated with five .060" holes. Both products plugged the leaks rather quickly - less than 10 seconds for the Titan Sealer and about 30 seconds for the ACDelco Sealer.
The last comparison was to allow the mixed samples of each sealant to sit undisturbed overnight in a clear container. The Titan product contains thousands of tiny polymer flakes in a thick carrier that isn't miscible in the coolant. This carrier hardens when exposed to air, and its likely purpose is to not only cement the flakes inside large defects but also to fill pinholes that are too small for the flakes. It's important that this particular product settles beneath the coolant in the bottom of an engine when it's shut off and away from any possible air pockets. After sitting for only several minutes, the Titan sealer separated out from the coolant and into a coagulant at the bottom of the container. The fine polymer particles appear to remain in suspension in the goo.
The coolant itself is the carrier for the particles from the dissolved ACDelco tablets. These particles remained in solution in the coolant, also near the bottom of the container, but quickly spread throughout the coolant when lightly agitated. There was no evidence of a goo or sludge that was reminiscent of some of the sealants I've used in the past.
In the end, I chose to use the tablets over the Titan sealer for a couple reasons. The goo bothered me because it would tend to settle at the bottom of the coolant system, and in my case that would be inside the water pump. I was concerned that during long storage periods the goo might thicken and cause the tiny pump to seize. In addition, adding the Titan sealer to the Merlin's coolant system felt like a somewhat irreversible decision. If I later changed my mind and wanted to remove it from the engine, I'd have to work quickly to remove it before the carrier had a chance to air harden.
Since there is about 1-1/4 liters of coolant in my particular Merlin, I dissolved 3/8 of a tablet in about 200 ml of coolant before adding the mixture to the coolant reservoir. This dosage followed the manufacturer's recommendations. I immediately ran the engine for a minute or so to circulate the material. After a cool down period, I made another one minute run after which there was no more sign of leakage. It's now been over two weeks with considerable run time on the engine and the seal is still intact. - Terry