Looking for a design to eliminate flaking material being left on the parting line caused by sub-gates into the cavity side of the mold. We have tried conventional, “D” and Chisel shaped designs.

What material, no matter the cold gate configuration it has to be sharp, gates do wear out try a valve gate hot runner.

Try adding a radius at the parting line where it intersects with the sub gate. A sharp edge there will shave flakes off the sube gate as it comes out of the mold.

Not enough information:Material type,angle of the axis of subgate,surface finish of gate,geometry of the surface gated upon.
This is quite a common and annoying problem.
1.For a subgate to work properly,the runner needs to be reasonably long as it too needs to flex a bit to release gate.The only shear point you want is at the part/gate interface.Stiffer materials (PC,ABS,PS) need a longer runner than flexible materials (PP,PE,PVC). Don't have the snatch pin too close to gate for the same reason.
2.Full round runner are best to enable it to 'rotate' at the gate end.
3.If the runner is only in the MH,the flex I mention will jam the runner in harder to the tool.
4.Good toolmaking practice is to polish the runner/gate but this is often overlooked as runners are dismissed as 'scrap' People forgot that its job it to deliver the polymer to the mould impression in the best possible condition.
5.A good fillet radius where gate meets runner (not parting line) helps.
6.The best(IMHO) shape is the spherical end on the subgate. It presents as a true circle on the part,giving the greatest flow for the smallest vestige.
7.Don't go too small on the runner-it will cool too quickly. If the mould can be opened with the runner still warm,flaking will be less,and it keep the gate live a little longer.

Sounds like a bit of a rant,but runners are often a neglected but vital part of a successful mould.
Pete

Peter Brown did an excellent job providing various key considerations-
What I would add is to try to determine the actual root cause so that you have another "tool for your toolbox of Lessons-Learned". This will save making the same mistake all over again.

What I would do is operate the press "by-hand" and try:
1. To warm the runner/gate as much as possible (short cycle or thermolate if feasible

1. Visually watch as you operate ejection "by-hand" -- what is happening to runner and gate - where is this "shearing/flaking action happening? Bring a lot of extra lighting to the press so you can plainly see the mold, etc.

1. Keep operating "by-hand" until you see what the source of the flaking is.

My guess is one of the following is causing the problem:
> The sucker (snatch) pin is not properly located thus preventing correct amount of flex
> The runner and gate aren't properly polished
> The angle of gate is not acute enough
> The included angle of the sub gate itself is too great - preventing proper flex
> Not enough radii at intersection of gate and runner
> The corner at end of runner and parting line is too "sharp" -- this needs to be broken to about a .010" radii so there's not a knife edge cutting the end of the sub gate as it passes by during ejection or mold open if you are gating into cavity side.

One thing for sure - you can correct it! It's not acceptable to mold this way -- it's very bad for the mold parting line and will cause future maintenance problems.

Good Luck!

I've also seen this when the gate lands are too long. Often times you see this with ABS/PC and other stiff resins. Dull gates mentioned above are a big cause for this, but incorrect angles and ejector pins too far away from the gate can cause this.
We run several bezel molds, and they tend to have this issue. We clean them once a shift, and I have recently added a blow off to the bottom of the EOAT to help clear the mold face.
Rick.

if the sharp edges are taken care then increasing the cavity temperature to keep the runner softer at the time of opening will help to reduce the flaking.
Many times the flaking is coming from the moving plate side,due to agresive under cut or not polished sucker pin area.
The shape of the flakes may be a good indication of the side from where they come from.
is the material 100% virgin?
some peeling /flakes may happen with excessive use of regrind or contamination of material

Not sure if this would be applicable in your situation without seeing a section view of the cavity, but I have seen this kind of problem resolved with a gate vestige.