AR Sponsor
[ARCHIVED THREAD] - Blueprint Basics... (Page 1 of 2)
Posted: 11/30/2005 5:02:47 PM EDT
|
This is kinda cool, I really don't start many threads! Anyway, it seems I get a slew of messages at the end of each day with questions about stuff, today from a guy that says he was told that the picrail on top of the flat top upper could cant between .164 and 0.020 and wanted to know how much error is allowed as per the drawing? Well... none. Below is a snapshop of what I believe he is referencing, but I think the problem comes from not understanding mechanical drawings. The shape and surface of a design like the rail is based on a "box" drawn in the center, with each major surface touching a corner of that box at the midpoint.  It seems a little weird, but this is a great way to make the design. As you can see, the person is confused by the dimension listed 0.164 -0.020 which really means the relation to the top surface -- but this variation does in no way indicate any allowance for being out of true. But, the big deal here is that this drawing has nothing to do with the specs for the upper, this is a generic spec for the rail itself and gives no positional data other than how much clearance the rail must have. The relation of this rail on the upper receiver is specified in the Tech Data Package (I can't reprint naturally), but trust me, the rail must be true and square by the TDP. 0.164 is more than an eigth of an inch, that is sloopy by many wood working standards and is beyond imagination for machined parts of any kind. |
|
You've got it backwards. The .164-.020 (.164 +0 - .020) dimension does not specify the location of the top of the rail, instead it references a feature below the top of the rail (shown off drawing posted) The origin of that dimension is the GD&T call out of Profile of Surface (Top surface of rail). Without knowing the exact canting question posed, I can't provide much more insight other to say that a Perpendicularity call out would cover that, not a general feature of size dimension. (i.e. .164-.020) Further, based on the information shown in the drawing, I am led to believe that the .164 -.020 dimension is a theoretical reference used to establish Datum -C-; although, I am unable to confirm this without seeing the entire drawing. Datum -C- is then used to specify the True Position of the rail profile. (This would not control canting either) If you can post more complete drawings I would be happy to decipher the GD&T for you. |
You are right, I had rewritten this three times trying to make it sort of simple -- it is pretty bonehead to say that a measurement is made against an undefined point "in space"  If I recall, without looking, the perpendicularity is 0.005 in the call out. Thanks for the catch, I tried to fix it so it is accurate and still easy to understand? The only thing I have to go by is the question that seems to be asking if the top of the rail can cant off to either side? eta: Just looked again, the 0.005 is the callout for flatness (you can catch part of that in the pic), no detail given for perpendicularity... full drawing up now. |
|
The top of the rail is the datum, marked with a C. The perpendicularity of the sides of the rail in relation to the datum is covered by the width dimesion (.617 -0.01), and the location tolerance given for it. The location tolerance (marked with the "crosshair") dictates, that the sides of the rail must be between two straight lines, that are perpendicular to the datum and the distance of which must not exceed 0.010 inches. The "M" indicates the maximum material tolerance. For example, in case the width dimension is manufactured to the lower edge of its tolerance range (0.617"-0.01"=0.616"), the perpendicularity constraint lines can be 0.020" apart. If the width was manufactured to 0.6165", the perpendicularity constraint lines can be 0.015" apart, and so on. Looking at those values it could appear that the perpendicularity is not very strictly tolerated, but we have to remember that the rail must adhere to these tolerances thorough it's length, no matter how long it is. Keeping the perpendicularity withing the tolerance range in a 12" rail is more difficult than in a 7" rail. In answer to the question, yes, the sides of the rail can be canted in relation to the top of the rail. However, this drawing is for the rail only, it doesn't relate to the upper receiver in any way. The upper receiver drawing will most likely give the position of datum C in relation to one or several controlled surfaces in the upper receiver. If that position is tolerated strict enough, the perpendicularity of the sides of the rail in relation to the datum matter little, since the datum is the decisive surface when mounting optics for example. |
Man, I didn't even dwell into what that other "M" marking means in the location tolerance.  I only explained the "M" after the "0.01" marking, but didn't bother to explain the "M" next to the "-C-" marking..... And already I'm accused of getting too technical. 
|
I appriciate your feedback... The original question I guess was that someone was asking about how far off thigs that are mounted to the top rail are allowed to "lean over to the left or right", I guess he was told that the rail specs *allow* for the rail to cant up to 0.164 -- of course this is defined in the actual tdp for the upper receiver... I can't post a scan of it, I would smoke a turd in hell for it I am sure, but I sort of figured out where that number came from looking at the rail specs... |
| I really enjoy this info. It's the forum version of the TV show "Numb3rs". Math war!hippie.gif |
Well, not sure "war" is the right word? I think everyone is saying the same thing, just in varying degrees of *tech*  I hope people notice that these forums can be informational and helpful, I recently got a message from a guy that talked about the way it used to be on here, when arfcom was a great technical help forum -- but as he says, it is little more than an extension of the EE nowadays... Anyway, I hope that people can get that old charm back, it is easy... just speak the truth and leave all attitudes and hidden agendas at the door! |
Please understand, I am in total admiration of the amount of the technical info that abounds on this forum. In no way was my comment intended as criticism in any way and my apologies if it was taken that way. What scares me is that I even understood some of this discussion. |
 Not according to ARMS! At least that's what I hear when people say that the ARMS rails are out of spec due to lack of clearance. 
|
3rdtk did always say that. C4 |
No Problem! I understood what you were saying... I just hope that we (the people that use the forum) can both gain and provide valuable information from the forum! I am with you on the fear of understanding some topics... At times it makes you think "I need to get out and hit the o-course..." Anyway, I hope that we can share good info, learn from each other and police our own ranks here. I have been sitting in a briefing for c4isr all morning and just sitting down here during lunch is a good way to relax -- If I can keep from trying to cut my own wrist with a KFC plastic spork, while thinking about having to go back for an afternoon of more... peace |
C4ISR (Command, Control, Communications Computer Systems and Intelligence, Surveillance and Reconnaissance)!!! Now were talking my language. C4 |
I'd need to see the drawing to say for sure, but all in all I think 0.164" sounds like a very unrealistic value. Seeing that the nominal rail width is only 0.835", the 0.164" cant would mean over 10 degrees off from the theoretical vertical axis.. Thats outrageous, wouldn't you agree? Even if all the dimensions in the drawing were left without tolerance, the nominal tolerances of any CNC machine would produce far better results than that, unless the fixture is total crap. I'm inclined to believe that the rail datum also serves as a datum for other features in the upper receiver, like the pivot pin and take down pin holes, which determine how the upper receiver is positioned in relation to the lower receiver. If the pivot pin and take down pin holes are tolerated to be parallel with the rail datum, then the natural result is that the rail will also be level for all intents and purposes. This parallelity tolerance need not be tighter than 0.005" to get it right. Naturally the cylinder cutout for the barrel extension, bolt carrier and receiver extension also need to be tolerated in relation to the rail datum to get the rail level with the barrel too. The cutout for the barrel index pin would also need to be tolerated in relation to the rail datum. I'd love to see the drawing, but since that isn't happening, I need to rely on my gut feeling that the 0.164" value is utterly wrong. |
I am sure, from the way it was put to me, that the person that made the assertaion simply does not know what they are talking about... the original statement was that someone had said that "things akimbo"(tm) on the rail are a result of a sloppy tolerance in the rail, he was told that the drawing allowed for 0.164 to 0.020 cant -- I figured from looking at the drawing that whoever pulled that out of their butt got the numbers from the picatinny drawing as shown above? The tdp shows that relational position of the rail, with a call out to the picrail specs and certainly does not allow for over an eigth of an inch in variation. Anyway, just wanted to clear up the question I got. |
I edited my post above to clarify why I think the value is unrealistic. As I stated, that 0.164" would mean over 10 degrees off from the theoretical vertical axis. I don't buy that. I tend to agree with you that the person who got this value didn't know jack about interpreting technical drawings. To me the rail appears to be tolerated nice and tight, and if a manufacturer can make a rail adhere to all the tolerances thorough the rail length, any visibly detectable cant in optics or what ever would be a result of a bad mount. |
The question people are talking about is not whether the rail can cant, to that variance, but to the height of the rail is the variance. Not, if one side of the rail is .164 and the other side is .144. It’s whether the rail is completely straight at .164, or completely straight at .144. I think this thread had something to do with the Troy Canting sight thread if I am not mistaken, and I emailed them. They emailed me back what I am sure they emailed everyone else (someone posted the same email I received this am) saying they built there sight to the middle of the tolerance I am guessing around .154ish ??? and if there base's as they say are made to fit that tolerance and someone has a upper that has it built at .164, then it throws the 45 angles off and moves one of the sides of their base's off. That’s as much sense as I can make of it... ?? |
A bit off topic but my question to Troy would be "Why the hell did they change how the sights were being made?". There were never any canting issues prior to this and my Troy Gen 3 BUIS, that wasn't made with this batch, is straight so why in the world would they change the specs? Regardless of their story it's obvious some changes were recently made or else we would have been seeing this issue from the begining. What's even worse is that the "mod" Troy is doing to fix them is a shim... 
|
The original question was asking if something mounted on the rail is allowed by spec to cant (or lean) to the left or right -- Your tip to the thread you mention does clear this up, it seems that it was posted that:
As for the rail being manufactured to a middle of the road spec, that should not matter if whatever is being mounted to it uses the bearing surfaces correctly. |
I suspect I know who may have sent you that, and he's spot on with that comment 
|
|
I'm glad we have smart machinists here in America because if I had to depend on myself I wouldn't have toys half as nice as I have. That mechanical BS is crazy to me and I took a drafting class. For some reason the rail drawing seems more complicated than most of the stuff I have drawn or worked with. |
The interpretation of the rail drawing in the Troy sight thread is wrong, the 0.164-0.020 measure has jack to do how things level up with the rail, if the stuff being mounted is correctly manufactured. Knowing the internet, this misinterpretation will soon have a life of its own and pretty soon it will be "common knowledge" that the rail can be off that much. Oh well...
|
It is probably already law over on The M.O.S.S. Forum
|
All I can say after a full day of sitting in a theater listening to that *world* is, wow... I am sorry Got a break today from the drone though... it is an interesting world, but a little well, non-operational for my blood -- at least for now
|
Gunzilla wrote:
Interesting how a member from a largely unregulated forum, while using a screen name, accuses another forum, where people with Real Names must take ownership of what they say, of spreading misinformation.... |
Come on Tim... check to see if there is an updated version of humor available for you current operating system? I will edit that post if it hurts that much, was simply trying to make a point that these things do take a life of their own and travel around as gospel literally overnight... sorry I picked on your forum, but I dont really spend time on any others but your and here. You can do me a favor though, please note that what has been said in here *is* accurate and for the purpose of trying to set things straight -- Grant misread the print and made a post that could have quite possibly been one of those thngs that grows legs and starts running from forum to forum, even it is not true. Not a big deal, we get it fixed up quick, fast and in a hurry. Again, sorry to have offended, it was not at all my intent... |
|
Gunzilla, I have nothing against you, though your tactic appears to be very passive-aggressive: make a snide comment, and when someone calls you on it, tell them they over-reacted/lack sense of humor/etc. That may or may not be your intent, but it certainly appears that way. As to the print/TDP, I have no dog in that fight. I am an end user, not an engineer. In any event, apology accepted. Have a nice day, Tim |
|
I dont know if this was cleared up or not, frankly I got tired of reading all the post...shere There is alot missing there as well. |
Holy crap.. that is like the first time in my life I have ever been called that! You are supposed to be a LEO right? You should know that we can take less than 10% of a persons meaning from spoken word, and even less than that from written word -- remember your effective comms training. I know we argue Mehrabian's model somewhat today... this is one of the classes I teach, but much of the original content is still quite valid. stay safe.
The rail is allowed to be of of flatness by 0.005, there is no indication of allowed cant. |
|
Gunzilla, I am assuming this thread is in response to the Troy BUIS thread? If so, I think(?) they are referencing the tolerances that vary the height of the dovetail, and that the variance yield a cant in the buis? Not that the tolerances allow for a cant in the rail itself. Unless I read the thread wrong? |
That variance *can* result in something mounted to the rail canting, but not if whatever is mounting to the rail is made to spec -- because of the tolerance, the designer has to adhere to the intention of the design and make sure that whatever he or she is designing uses the top surface and the bottoms of both sides of the rail as a mounting surface. The problem is that it is easy to just buy a cutter that cuts the double 45 angle (on one side) and the clamp the other side. Depending on how the adjustment side is made, the part being mounted may either sit high, with a space underneith, or cant towards the adjustment side as a result of poor design. |
Ya that comment was out of line (about 10-8). There is more ACCURATE knowledge on there anywhere else on the internet. It must be nice to hide behind a fake screen name and throw stones.  For the record I was passed that the M1913 spec held true for receivers as well. If my good ole buddy 3rdtk (Dick S) was still on here he would have backed that up as well. If the TDP has a specific spec for the receiver rail then great, but I have never seen it. Receivers are all over the place (including Colt) in their dimensions which is where the problem is. C4 |
I know what you are saying. I was just trying to clairify which of the two issues the discussion was about: (the tolerances causing) 1) a cant in the rail or 2) a cant in the accessory mounted on a straight rail |
It was off on one of the less traveled topics, not sure where, but I guess the thread was about a BUIS that canted over pretty good on some rifles. I got a message asking about how far the picrail spec allows it to cant, because I guess it was posted that the spec allows for up to 0.164 cant in the rail itself -- I just wanted to clear up that was misread in the drawing, there is no real tolerance for canting in the drawing. The tdp for the upper specifies the positional data for the rail and calls out the picrail specs for the detailed construct... but still, no room for canting (at least not an eight inch plus). |
|
Like it or not, but the rail drawing is very specific about geometric tolerances. Anyone who says it allows the rail to be canted this way or that way by any meaningful amount is wrong and should not be making such comments in the first place. These drawings are for professional machinists and quality assurance people, they are not easy to interpret. Just looking at one dimension and its tolerance will lead to wrong assumptions, like in this case. One needs to be able to understand the location tolerances and the maximum material tolerances and know how they affect each other. In this drawing the tolerated surfaces have VERY complex relationships, and even trained engineers can make mistakes interpreting them. After a short crash-course practically anyone could make a CNC program to manufacture a rail that RESEMBLES the picatinny rail. However, it takes a professional to program the machine so that the actual end result will pass the quality assurance tests! Each and every dimension is measured and checked that it adheres to the tolerances, and the tester must also understand the maximum material principles and all that really complex stuff. Very few people are up to this stuff, it takes extreme care and professionalism. |
The operations guy (Brent) from Troy sent me this e-mail today:
C4 |
Yes... I had posted this earlier in this thread - the TDP callout for the rail is the picrail spec as shown here -- however, the positional and relational (where the rail islocated) data for the rail is part of the tdp. |
|
I've always wondered about the "extra numbers" on such drawings since my freshman drafting class lo these many years ago. I could draw the whole rail, but couldn't even begin to have any idea what most of the non-normative numbers meant. Now I have a bit more insight. Thanks all! Here's a machinist-technical question. The point was made that the basic linearity of the rail sides was not very strictly tolerated, but that may be due to the rail needing to fit that tolerance for its entire length, no matter how long it is. How does that relate to the typical "slotted" rail? Do more slots mean tighter or looser rails? Would longer spacing between the slots mean tighter or looser rails? Thanks! |
I never got that from your post. What I took away is that ONLY the TDP describes the spec for the rail on the receiver. C4 |
nope... the tdp calls out the actual specs for the rail as the standard picrail design. The tdp only shows where the rail goes -- sorry if that was not clearer in my previous posts.
|
[ARCHIVED THREAD] - Blueprint Basics... (Page 1 of 2)
AR Sponsor