Sensor for molds

I am trying to come up with a way to check if a casting mold has emptied after casting.
It is for a lead casting machine, that pours molten lead into multiple molds. The molds rotate and after the lead solidify they open and drop the bullets down, then they close and repeat the process. The thing is that sometime bullets will stay stuck in the mold and that will cause molten lead to spill down, creating all kinds of problems.
The molds have two holes, that vary in diameter depending on the caliber - 9mm, 10mm, .45. The molds are hot after the lead is poured. Although the molds are made of iron, they are sensitive to physical damage - any burs that are created will hang the bullets.
This is a picture of the machine and the molds.

Just a quick thought:

A light and optical sensor mounted just past the exit point detects the reflection of a still occupied space (because the new casting is shiny), pauses the machine, moves it back and wallops the mould with a hammer (OK, weighted linear actuator) to free and drop the slug.
It restarts.
If the mould is then still occupied, it stops the machine and sounds an alarm to call the operator.
----------
Update: After reading Tony Matteo's comment below, an alternative is to detect the exit of the bullet from the mould by an optical or mechanical sensor, then have the machine continue only if that sensor is activated.
If not, it will stop and activate the operator attention signal.

(For absolute safety, you could fit both sensors to the control system at minimal additional cost.)

Might seem like more trouble than it's worth but incorporating an automated sprayer that spits out a puff of graphite powder, or something similar, into each mold after each casting would probably eliminate the burring and seizing issues. Graphite might not be the exact right compound but I strongly believe adding such a component to the overall mechanism will get it to work without further issue. .

It is worth a try.
I've never used them, can you recommend something?

The thing with spraying or applying any type of coating too often is that eventually it will build up on the walls, decreasing the diameter of the bullets. Also any liquid left in the mold when the molten lead is poured will result in an explosion caused by the steam, the so called "tinsel fairy visit" as its known among lead casters.

The graphite will adhere to the bullets and make them shoot much much better, You must be unfamiliar with graphite, it's not a liquid though some companies blend it with oils and sell spray graphite.

An example of graphite is what everyone calls "lead" in pencils.
Trust me, it will work and improve the bullets performance as well.

Mr. Bushnell
Pretty much answered your request (checking casting mold contents )

Also as the product is going down the shoot after mold releases, a diverter / resistance (flap/panel/wheel) in synchronys timing of operation allowing to release by weight > when weight is not grater than X mold will pause - alarm waiting for attention with out urgency.

However A few questions may lead to eliminate this glitch, Is this random or specific mold,molds. How offend occurrence, conditions, always been a problem, new, more frequent? Your referring to Iron molds sensitive to physical damage, any Burs that are created? will hang the product ! is that what is happening or just a condition you know of or has been suggested?

As the release goes please ( do-not ) use any release product not approved for this specify problem.
With out first identifying the exact reason!
The mold s will benefit from being polished and harden with out changing dimensions. Professional harding & polishing as in racing part.

The manufacture may have a answer for you or want to sell you new molds

All The Best

Trust me, graphite dust puffed into each mold before casting will save you the trouble of building a sensor system unless you really want to. Should also make the bullets shoot more accurately and faster.

I have to find an electrician to try the optical sensor idea.
The manufacturer strongly warns against any burs on the mold, they recommend cleaning with toothbrush only.

I've used fiber optic sensors in similar applications in the past. I know high temperature solutions are available (Glass rather than plastic). Similar to the following:

http://www.automationdirect.com/adc/Overview/Catalog/Sensors_-z-_Encoders/Fiber_Optic_Sensors

I've used Keyence and Moeller Electric i in the past. One similar project we used a very small sensor to detect the existance and proper orientation of tiny plastic bushings.

*You could also look into capacitive sensors*
http://www.engineerlive.com/Electronics-Engin...sensors_feature_extended_sensing_range/21590/

It may be more resistant to lead contamination if you can find one with the right size and temperature rating.

There are many other possibilities.. Mold vibrator? Camera? Reed Swtich

It sounds from what I've read here preventing the bullets from getting stuck in the first place is the answer. Although after that you can have a sensor that weights the bullets as they come out or optical sensor that counts them as they come out.