smoke machines

Trying to step into the 21st century and get away from diagnosising intake vacuum leaks with propane etc., I bought this really cool smoke machine and man does it make a giant cloud. I made a couple of intentional holes in an inlet tract and the smoke was clearly visible. Awesome! So my question to those who use smoke machines regularly how far up will smoke travel? Can you pump so much smoke that you might see it at the tailpipe? Any rules of thumb for how much smoke pressure in PSI is the most diagnostic?

 

I've had a Matco machine for about 4 years. I can't say I've ever seen smoke coming out of an exhaust pipe but I expect it you ran it long enough you would. I did do 5 cyl. Volvo turbo last week and it filled every thing in the intake/turbo system and the crankcase/breather system. It turned up .010" hole in the intercooler. I think they're great. I've also found it wonderful for tracking down tiny leaks in the floatation tanks on the small sailboat I race. They have to pass a yearly float test.
It operates at about 1/2 psi.

TomB

 

I have one of these. A great tool for diagnosing a variety of leaks.

I don't have any tips for you. Mostly I've used it to identify blown headers. It only takes a few minutes.
Image Unavailable, Please Login

 

Carl, I tried making one of these smoke machines myself awhile back (that little demon that all DIY'ers know well was whispering in my ear "how hard can it be"?)

It was based on one of those halloween type smoke machines, a fancy "eductor" and a bottle of compressed nitrogen with a very low pressure regulator. The concept was good, but the execution failed miserably - primarily because the halloween smoke machine wasn't built to have its smoke drawn into another system under partial vacuum.

When I get bored one of these days, I'll have another go at V2.0.

I think these machines (the ones that work) are great for finding leaks in intake manifolds, EVAP systems, gas tanks, etc (all known problem areas for us 550 owners ) For best effect, the pressure has to be adjusted for the system you're testing, ie EVAP systems are in ounces, while gas tanks are higher (I wouldn't go much over a couple PSI with a gas tank, and definitely wouldn't use anything but nitrogen).

I've often thought a smoke machine could be a valuable tool for leakdown testing if it could be adapted for the higher pressure, since the first thing you want to know is "how much" leakdown there is, but then the 2nd thing is "where is it coming from"? So if you have smoke coming out of the crankcase breathers, you know you have ring problems, whereas if it's coming out the intake manifold you have leaky intake valves - same if it's coming out the tailpipe - leaky exhaust valves. We're accustomed to listening at all these points to narrow the problem down, but sound carries well inside an engine, whereas the presence of the smoke would be simpler and more objective. But the very low pressure that the "normal" smoke machines operate at isn't going to find much in a leakdown test. To be useful, the smoke would have to be introduced at the intake of the compressor, and whether it would stay as smoke, or would condense and/or particulate out at high pressure would be the question.

Why don't you give it a try, and see if you can produce some smoky compressed air at 50-75 PSI

 

I'm having fun with this toy. Trying to find things to do with it. Variable pressure would be nice if it is needed hence my post to those who use them on a regular bassis. Me I'm just figuring it out. Honestly I deal with so few leakdown failures that I personally see little use. 99% of the time I'm just confirming everything is OK. The 1% is a catastrophic failure of an engine and just wanted more data before teardown. I have never played with that nebulous Ferrari 355 type tricky valve problem where the controversial hi pressure vs low pressure leakdown tests make a difference. Vacuum leaks well you see those all the time and smoke seems to be perfect for that and fuel system EVAP issues.

 

You will find that to be a very valuable tool, the uses are limitless!

Yes, it will go from the intake tract to the exhaust tips. Mine has a time out function and sometimes requires the button to be pushed twice to do that distance.

Make sure to use an inert gas when playing in dangerous situations.

 

TIM ----

Where did you get that unit -- looks nice -- how much $$$ ??? What is the "smoke" comprised of, and how is it generated ?

 

That's looking really good.Share some pics of smoke. It genertes

 

In a "jamb" one time, I made a smoke tester out of galvanized water pipe and fittings... it was a big "U" about a foot wide at the bottom and the sides of the "U" were about 2 feet long. I poured a little oil in the bottom and hooked one end to a well regulated compressed air source and the other end to a rubber hose. The bottom (horizontal) pipe was heated with a propane torch until smoke came-out.
It worked great, but I did not end-up needing to use it because as I started to hook-it-up the leak source became abundantly clear.
My set-up used 1/2" water pipe, but to do it again I would probably adapt-up to 1 1/2 or 2" pipe for the bottom (tank) and the initial segment of the exhaust end because in one test the bottom loop was too full of oil and too hot so some oil squirted-out. Also, I think it would be better to use an inert gas. But I don't think a fire is a big risk because there is a lot of smoke created right after boiling.
Another improvement might be to use and IR thermometer to keep the temperature within a safe range?
A set-up like this would facilitate using higher pressures for a leak down test.

 

It's made by OTC (they make a couple different models). The smoke comes from a proprietary oil that is supplied with the unit. It has a UV dye in it that leaves a residue that can be seen with a UV flashlight (included) where the leak is too small to see the smoke itself. Not a cheap toy (look up "Leak Tamer" on Google and you'll see a few for sale) but in my opinion worth the price for someone "in the business".

As Dave mentioned, it's recommended to use inert gas if you're testing fuel-related systems. For most things, just hook it up to shop air. An assortment of fittings and caps are included to close things up. Push the button and smoke comes out. If you're testing a closed system, the flow meter will tell you the approximate size of the leak. If the leak is very big, it will be hard to miss.
Image Unavailable, Please Login

 

Great stuff Mark! Good to see the homemade solutions. Would like to do so myself. Any pictures? Don't worry if its looks "amateur" - all the better!

 

Yes, a picture would be worth a 1,000 words.
I'll try to post something and it will look amaturish.

 

Attatched is a diagram and a picture of my "smoke machine" at work. When I tried it before it made lots of smoke this time there I only put a little oil in there.
My thought was that perhaps that air-conditioning oil with the UV leak detector in it may work (like teh other unit) with a UV light.
Also, this set-up may work well for a cylinder leakdown test because you could put 100 or more PSI into the system? I'm sure you would be able to tell if the intake or exhaust valves were leaking.
View attachment Foleymatic Smoke Mach Diagram.pdf
Image Unavailable, Please Login

 

Our Subaru Service centers use smoke machines almost daily. The biggest problem we find is with the pressurized fuel systems. There are nearly impossible locations to visually inspect without the smoke machine. Once we made a custom adapter to attatch the smoke machine to the fuel filler we cut the dio time down from several hours to a few minutes. We can follow the fuel system from tank to fuel rail and then follow the the return system back to the tank. Our machines are from Matco with variable pressure to really push the smoke from one end of the car to the other. I strongly suggest any professional service facility should have one.

Howard Musolf
1981 308gtsi
1982 400i Cabriolet
1994 348 spider
1988 Lotus Esprit Turbo

 

Have also used it to find interior water leaks.

 

Thanks for the pics Mark - your son looks as pleased as you must have been. If I ever try to build a version I will certainly post up some pictures (even if it doesn't work). I would only be looking for a leaky header.

 

Thanks !

Very cool, indeed, but yeah, the price is a bit hard to justify for the average DIY'er such as myself ----- I think I'd better stick with my old-school, cheap-o method......
Image Unavailable, Please Login

 

Sorry but when I saw the title of this thread I immediately thought of this, its an oldie but a goodie:

http://www3.telus.net/bc_triumph_registry/smoke.htm