All of which can be returned to the mash
Right. All samples can be returned. I’m not sure Robert was hinting at an issue with sample size to begin with.
I think what he was saying was separate from that, i.e. it’s just a lot easier taking multiple refractometer samples than multiple hydrometer samples. Something along those lines.
Tracking conversion can be done by sampling the wort throughout the mash and recording the gravity values. Obviously if gravity is increasing toward the estimate value for first wort gravity you are on the right track. If for some reason it isn’t, then you can troubleshoot based on that information.
Another interesting way of doing it that not everyone is setup for is tracking conversion visually through a sight glass and using constant recirculation in the mash.
Yeah, I was just suggesting that if you’re tracking conversion during the mash as Derek describes, as some of us do, it’s a heck of a lot easier to take a couple drops, put them on the refractometer stage, and Bob’s your uncle, than to pull multiple hydrometer samples, separate the solids, and try to quickly cool them.* Now, should I be returning all those little drops to the mash mixer…? 
*For that, I remember a guy had a very clever solution in last year’s gear issue of Zymurgy. Like a little jockey box to pour a sample through.
Separate the solids? Not on yer life! ;) And why do that for a hydrometer and not a refractometer anyway?
Well, when I take a couple drops from a settled mash with a pipette all I get is clear liquid, but if it’s really sludgy having just been stirred, (I do this with fermenting beer too,) I’ll put a couple drops through a coffee filter. I find clear samples really give a more accurate reading even on a digital unit. With a hydrometer sample, you don’t want the hydrometer getting pushed up by the suspended particles and giving a false reading (remember the thread about swimming pools and poodles? I mean, suspended yeast?) So it just seems all in all more trouble to keep taking mash readings with a hydrometer. But the end result is, you get the same information. Just my preference.
I also filtered a sample and but I saw no difference in my reading. Yes, if you have a thick, solid layer of sludge it can hold the hydrometer too high, but I make sure not to do that.
i would agree that efficiency is not important but only to a certain point. if you have extremely low efficiency, you can easily excede the abilities of your mash tun with a higher gravity brew. I know an efficiency below 65% would have me bothered. If your kettle size isn’t an issue, consistency is important for nailing target gravities. You can always add water and dme to dial in your OG until you figure it out.
This is an interesting read…it seems a fine crush may be working against you…
But note they’re talking about brewers doing conventional lautering (like them and me.) There it makes sense. But the OP is doing BIAB. In that case, and especially full volume BIAB, you could grind finer because the flow concerns just don’t apply. I’d think finer would help in that case. Same should apply to batch sparge if you can adequately filter the wort. Anyway, good summary of the issue there.
I just got through listening to the Master Brewer’s podcast about this. He had me going for a while… but then lost me. I’m “Havig” a hard time believing that his findings are true for all small brewers. It’s definitely something to think about and play around with, but I’d take great care in assuming either a hard crush or lighter crush is necessarily better or worse or whatever based on just one study. More experiments are needed. And like Robert says, any correlation to homebrewing, especially BIAB or batch sparging, is very likely irrelevant. We as homebrewers need to take great care in determining what stuff from the commercial world we can apply to our stuff at home, and what doesn’t fit well with what we are doing. This might be the perfect example right here.
Which was the reason for my post above. (#11)
I think you should put “More experiments are needed.” in your signature.
Personally I like to condition the grain every so slightly to keep the husk intact, allow it to grab better in my simple 2 roller mill, and also so that I can crush a little tighter.
Heh… It certainly is a common phrase from me. However at least for the time being, and probably for the rest of the 21st century, I prefer the prevailing point pertaining to general idiocy. 
I believe it’s Socratic wisdom: The point on general idiocy implies the need for more experiments, so you’re good! 
But that will depend on each person’s system and crush. My efficiency definitely increased when I went to a finer crush. You just can’t generalize.
Let’s see, BIAB, Batch sparge with a hose braid, false bottom and fly sparging. One would expect different results for different equipment and process.
This is just one data point, but I found my efficiency dropped when my mill gap slipped a couple years ago. I thought it was the grain, initially, but checked the gap with feelers and there was the culprit. I grind around .025 gap and condition the grain, also, as I like the greater intact husk results I get with conditioning. BIAB with HERMS recirc and Hochkurz step mash as the process for most of my beers (lagers).
Undamaged husks, finer grind, and recirc is where’s it’s at. Clear wort, smooth lautering and great efficiency.
Something I’ve noticed lately trying a lot of different malts, local and not. With the same mill gap and the same rpm, some malts give nearly intact husks (recent examples for me, Simpson’s and a local product) and others noticeably more fragmented husks (e.g. Rahr and even Weyermann.) It would seem some combination of barley variety and processing contribute to the quality of milled malt, and I’d like to see more study on this. Note that each category in my above examples includes a high protein domestic and a low protein European representative.
Interesting observation, Rob. I use Maris Otter, Briess 2 Row Brewers Malt, Simpsons Golden Promise, Wyermann and Best Maltz Pilsner. I will check out the husks on these when I next use them and see if I have the same issue.
I know that different lot numbers will give different extractions having experienced that before both at home and professionally. I also find my efficiencies are usually 5% or so lower with my darker beers (Porters and Stouts) than they are with my lighter beers. I think that may have something to do with the darker grains in the mash as I mash everything together and not separate the dark grains out and add them back before lautering.