Incorrect geometry check in SYNCVMLC

[ojalaj]

According to Varian TB information for Millennium MLC, WBS < WG + WTIP. Similarly than in #442 for SYNCHDMLC, code check W_(bottom of support rail) >= W_groove + W_tip should be W_(bottom of support rail) < W_groove + W_tip.

I.e. in SYNCVMLC_cm.mortran should line 1298 be:
IF(WRAILBOT_$SYNCVMLC(2)>WGROOVE_$SYNCVMLC(2)+WTIP_$SYNCVMLC(2))[
and then change lines 1301-1302 accordingly?

Now I get:

 ***ERROR IN CM   5 (MLC) in TARGET leaf definition:
 WRAILBOT_MLC(2)<WGROOVE_MLC(2)+WTIP_MLC(2).
 Lower rail too narrow.

[blakewalters]

@ojalaj, this would also mean the geometry of target leaves in DYNVMLC is incorrect. I'm not all that familiar with the Millennium MLCs, but I think DYNVMLC and SYNCVMLC were intended to model the 80-leaf Millennium, while SYNCHDMLC, in which W_railbot < W_tip + W_groove, is intended to model both the 120-leaf Millennium and the HD MLCs. @marenaud, do you have some experience with the 80-leaf Millennia?

[ojalaj]

Hi @blakewalters & @marenaud. I have not modelled 80-leaf Millennium at all, so I don't have any experience on the possible problems with DYNVMLC/SYNCVMLC related to 80-leaf Millennium. What I do know, however, is that 80-leaf model has OUTBOARD leaves #1 and #40 and then 38 FULL leaves in between (correct me if I'm wrong). For years I've used SYNCVMLC for modelling 120-leaf Millennium and for couple of years I've used SYNCHDMLC for HD120 MLC. I have not even thought that SYNCHDMLC could/should be used for 120-leaf Millennium (maybe just because there is "HD" in the CM name and in Varian terminology "HD" is used only HD120 MLC). Do you have any thoughts on this @marenaud ?

[marenaud]

Hi all,

I don't have any experience modeling the 80 leaf MLC, though I know that the original DYNVMLC (and therefore SYNCVMLC) module was specifically for the 120-leaf millenium.

I have some memories of wrangling with geometry checks but I don't remember exactly what I did to make BEAMnrc accept our measured data, especially now that there's been a few changes to the checks. I now use SYNCHDMLC for both the millenium 120 and the HD (thanks to a bug fix by @blakewalters), but up until last year I had two different accelerators (one with SYNCVMLC, and one with SYNCHDMLC).

[blakewalters]

Thanks for the input, @marenaud, and the information (reminder?) that DYNVMLC and, thus, SYNCVMLC, were coded with the 120-leaf collimator in mind. In this case, then, it does, indeed seem like the restriction on W_railbot is wrong. If you happen to have any of your old SYNCVMLC/DYNVMLC inputs lying around, @marenaud, I'd certainly be interested in seeing them!

[ojalaj]

Thanks for the contribution @marenaud. I should have remembered the paper @marenaud referenced to and only 120-leaf Millennium contains HALF leaves. For DYNVMLC PIRS-509 also mentions Millennium MLC and that there are usually 120 leaves.

Anyhow, now that I see that there is no restriction to change from SYNCVMLC to SYNCHDMLC for 120-leaf Millennium, I may make the change as well. Another limitation both CMs still share is the accurate model for OUTBOARD leaves #1 and #60, but that is another story then.