I have found that a lot of the problems with injectors can be traced to boiler checks that are not large enough or else the ball does not lift off of its seat high enough to give un-restricted flow. Of course everyone already knows to use gently radiused curves in the delivery pipe, errrrr- - ahhh don't they??? I noticed a photo of a Vulcan Iron Works duplex that the delivery pipe passed completely over the boiler check and then made a gradual loop around the check to get to the bottom. This in essence was a 270° turn! I don't know why they did not come in under the valve so only one 90° radiused turn would have served. It might have had something to do with the large steam pipe from the dome down the side of the boiler to a "Y" that went to the front and rear engines. Another problem is sometimes not having the throttling valve between the injector and water supply adjusted correctly. Too little opening and the injector will starve and too much will result in waste through the overflow. Sometimes too, steam will leak back through the injector and get it too hot to start unless cooled by pouring water on it. Full size injectors could give problems so we must sort of expect them with our small stuff. A properly operating injector is a joy to listen to as it sings merrily along filling the boiler in what appears to be complete defiance of all known laws of physics. (I have heard that they actually have little men or gremlins inside them that pump like fury when called upon or else it is truly magic!) I did a boiler hydro test (200psig), set the two safeties, adjusted my automatic gas/steam regulator and fired up the boiler to make sure all was ready for the Mid-South meet in a few weeks. Jesse in Tennesse > Not so sure I agree with you there. On one of my engines I could not get the > injector to work reliably. It wasn't until I modified the clack valve, at > someone else's suggestion, to increase the ball lift that it worked > correctly, regardless of what make injector I fitted. Until then I was > sceptical of the idea as well. > > Andy MODEL ENGINEERING DISCUSSION LIST. To UNSUBSCRIBE from this list, send a blank email to, modeleng-request@xxxxxxxxxxxxx with the word "unsubscribe" in the subject line.