(03-13-2012 11:15 AM)Fly So Hi Wrote:  Hehe, yes, we're planning too much in advance here.

Oh most definitely, although it's always good to give the next step some thought before making the steps at lead up to it so that the later steps are kept in mind in the earlier designs. When things that will be added later aren't kept in mind in initial designs, it tends to lead to the need to completely start over to add anything.

But agreed.

(03-13-2012 11:15 AM)Fly So Hi Wrote:  True, but you could implement a reset XY position for that in the code. Your system sounds great actually, use whichever is more simple. We can always upgrade the machine on future versions. Let's remember that the focus/goal here is to make a working machine. We can make it work with tridium later, but for now, let's make it work...

... Haha, no accelerometers or gyroscopes for now. Let's keep it simple for the average Joe to be able to replicate it...

I like the button system. Ought to make it cost less too, in addition to saving some time programming. It might need 2 stops per bottle though: one to pick up the pipette tip from some holder near the bottle, and one to actually interact with the bottle itself.

Tridium... NOW you're thinking! Hahaha

With slight seriousness though: if we do make it spill proof for car chases and such, it would be better if it could achieve that purely through a mechanical design. That applies to most everything here I think, until the mechanical design becomes more complicated and less robust than the programming solution, but that would be unusual for most of what we're doing I think.

(03-13-2012 11:15 AM)Fly So Hi Wrote:  I'll draw you my idea for this when i get home.

Can't wait! I'd draw mine, but I figure we should go with the rotary for the initial design anyways and worry about space saving stuff later. I'm thinking the main linear actuator design comes first in importance, then design for interacting with the bottle (raising/lowering said main linear actuator) and expelling the liquid into the mix bottle, then lastly transporting from concentrate to concentrate. After that it's just final touches like the user interface.

(03-13-2012 11:15 AM)Fly So Hi Wrote:  All ranges of prices. Abut US$ 6,00 for this:
http://www.amazon.com/Velleman-6A-PELTIE...grid_i_0_2

Cool! About how many would you think we would need? I still think this would be a very final touch, but no reason to rule it out if it isn't difficult or expensive.

(03-13-2012 11:52 AM)Fly So Hi Wrote:  Let's assume all is alcohol for now. We worry about DPG later.

Sounds good. Might still be a slight problem as I notice it still has a small bubble when I use my 28 guage syringes with alcohol, but that may just be from the dead space and just from using inferior equipment in general. Also might have something to do with the needle being steeply beveled, increasing the surface area for the bubble to cling to.

But I agree, should be good enough to just ignore it initially.

(03-13-2012 11:52 AM)Fly So Hi Wrote:  We could use a cone-shaped container like you said yourself. That way, the tip would almost touch the bottom of the cone without touching the walls.

I meant more just whether it would be physically able to reach a place to touch the drop of liquid to.

(03-13-2012 11:52 AM)Fly So Hi Wrote:  

(03-13-2012 10:28 AM)gentlmentlemen Wrote:  Maybe if the mix bottle wasn't rotated and raised, but translated sideways, then raised.

Sounds even better than the rotating thing!

If it were as easily achievable I would agree, but after thinking about it I think it would be slightly more complicated. Not tons, but enough to merit using the rotating thing if we can. Either one could work though.

I was thinking that in order to have both functions from a single motor, it would just run the lifting gear chain until it jammed, which would then activate the gear chain which moved things sideways. Or maybe vice versa would be better... not sure.

In both cases, it should be very easy in future designs to add more mix bottles at a time, so that if you're making 2 different mixes at once which both include the same molecule, it doesn't have to rotate the rotary (or go around the track) again for each extra mix bottle. That way it could make 2 mixes in one trip and only make 1 stop per bottle. Could be useful for seperating mixes, like into say social and sexual components, or self-effects and others' effects so that those affected by Androstadienone with depression could more easily apply that part of the mix on the back of their necks or the like.

(03-13-2012 11:52 AM)Fly So Hi Wrote:  Actually, that was shown on the video, but not emphasised. I honestly think that's not a must.

Well that's good, because I wouldn't think it would be, and that helps for some other stuff too if it isn't a must. I figure it can't be too much of a must other than to touch the liquid's surface for the drop that sticks to the end, and it doesn't need to be submerged for that to work. Obviously we don't want to squirt it from miles away, but as long as its relatively close to the surface such that there isn't splashing I figure it will be fine.

(03-13-2012 11:52 AM)Fly So Hi Wrote:  

(03-13-2012 10:28 AM)gentlmentlemen Wrote:  So basically fill it with however much PA and the like it needs, then fill it with the molecules?

Yes, that's how i do my mixes. 1st fixative, 2nd alcohol, 3rd molecules.

Wouldn't it work better to go from lightest and thinnest to heaviest and thickest? That way the heavier stuff has to sink through the lighter stuff to get to the bottom and would have a greater tendency to seperate and be dilluted and would do so more quickly than if it was sitting on the bottom below the lighter stuff. So wouldn't it work better to go: 1st alcohol, 2nd fixative, 3rd molecules? Probably isn't a huge deal, but just my thoughts.

Also, do you know of any cheap or free CAD type programs we could possibly use to share designs? It isn't essential or anything, but might speed things up especially as far as re-designs go. I used to have a program a long time ago from one of the competition years, but I forget what it's called and didn't use it much. I doubt I still have it.

I don't have a whole lot of experience using CAD programs, but I'd be willing to learn more and I figure they're generally speaking straightforward to use after an initial learning curve.

Would probably be superior to just drawing things and sending the pictures, but for now it isn't really a big deal. Just a thought I had.

(03-13-2012 1:40 PM)gentlmentlemen Wrote:  I like the button system. Ought to make it cost less too, in addition to saving some time programming. It might need 2 stops per bottle though: one to pick up the pipette tip from some holder near the bottle, and one to actually interact with the bottle itself.

Good point, a tip holder near each bottle on the grid.

(03-13-2012 1:40 PM)gentlmentlemen Wrote:  I'm thinking the main linear actuator design comes first in importance, then design for interacting with the bottle (raising/lowering said main linear actuator) and expelling the liquid into the mix bottle, then lastly transporting from concentrate to concentrate.

I was thinking the same thing!

(03-13-2012 1:40 PM)gentlmentlemen Wrote:  Cool! About how many would you think we would need? I still think this would be a very final touch, but no reason to rule it out if it isn't difficult or expensive.

Enough to cover about a quarter of the grid's surface, i think. Check this video, they can get to 2.5 ºF

(03-13-2012 1:40 PM)gentlmentlemen Wrote:  ...I notice it still has a small bubble when I use my 28 guage syringes with alcohol, but that may just be from the dead space and just from using inferior equipment in general.

Very likely the empty space. This problem goes away with pipettes/tips.

(03-13-2012 1:40 PM)gentlmentlemen Wrote:  I meant more just whether it would be physically able to reach a place to touch the drop of liquid to.

Now you got me thinking. Very well observerd. Maybe extra pressure for the small ammounts, so it squirts the concentrate. We'll have a problem if it sprays instead of squirts... I'll think of something.

(03-13-2012 1:40 PM)gentlmentlemen Wrote:  In both cases, it should be very easy in future designs to add more mix bottles at a time, so that if you're making 2 different mixes at once which both include the same molecule, it doesn't have to rotate the rotary (or go around the track) again for each extra mix bottle. That way it could make 2 mixes in one trip and only make 1 stop per bottle.

I know what you mean, and that's a good idea, i just think we gotta walk before we can run. Let's focus the basics first to keep it simple and practicable.

(03-13-2012 1:40 PM)gentlmentlemen Wrote:  Wouldn't it work better to go from lightest and thinnest to heaviest and thickest? That way the heavier stuff has to sink through the lighter stuff to get to the bottom and would have a greater tendency to seperate and be dilluted and would do so more quickly than if it was sitting on the bottom below the lighter stuff. So wouldn't it work better to go: 1st alcohol, 2nd fixative, 3rd molecules? Probably isn't a huge deal, but just my thoughts.

I usually mix the components in this particular order because a more viscous component like DPG evaporates much slower than Alcohol, for example. The moment i add alcohol to DPG, the alcohol is already "fixed", so it will evaporate less, not spoiling the measures.


Here's the drawing i wanted to show you:



Fig. 1 - The light grey part is the top layer.

Fig. 2 - The dark grey cover is the middle layer, between the top layer (grid) and the bottom layer (tube with liquid). It slids back revealling the liquid inside the tube.

---

(03-13-2012 1:48 PM)gentlmentlemen Wrote:  Also, do you know of any cheap or free CAD type programs we could possibly use to share designs?

I'll google something.

(03-13-2012 4:52 PM)Fly So Hi Wrote:  Good point, a tip holder near each bottle on the grid.

Yup. I was thinking the only other way would be to somehow integrate the pipette tip into the bottle. While that would be cool, it would be annoying to do, and even if there were pre-made bottles like that, I'd like to not have to use special bottles just for the machine eventually. That tends to be annoying and in general if this is something to be built by others in then future, then we should avoid the need for specific parts which might be unavailable at a later date.

(03-13-2012 4:52 PM)Fly So Hi Wrote:  

(03-13-2012 1:40 PM)gentlmentlemen Wrote:  I'm thinking the main linear actuator design comes first in importance, then design for interacting with the bottle (raising/lowering said main linear actuator) and expelling the liquid into the mix bottle, then lastly transporting from concentrate to concentrate.

I was thinking the same thing!

Cool! In that case, lets talk linear actuators!

We could probably buy one premade, but we might have to add onto it after that, idk. We would need accuracy relative to the distance which the single volume pipette must travel, and as such the first and most important thing in my mind is to look at how far the button/plunger on the back of pipettes travels for various pipettes, or whether it's all the same.

(03-13-2012 4:52 PM)Fly So Hi Wrote:  Enough to cover about a quarter of the grid's surface, i think. Check this video, they can get to 2.5 ºF

<a href="http://www.youtube.com/v/3x-wxLEfLEQ&amp;rel=0&amp;fs=1" target="_blank" style="background:url(images/midnight/jump.gif) no-repeat right bottom;padding-right:13px;">http://www.youtube.com/v/3x-wxLEfLEQ&amp;rel=0&amp;fs=1</a>

Woah. That's pretty crazy. Well, giving the bottles a 3" by 3" square each (I doubt we'd need more than that on average, maybe a couple crazy big ones just for fixatives and PA), we'd have a grid from somewhere around 270 to 450 square inches for the whole grid, but I'm shooting pretty large there. Probably wouldn't have to be near so big. I don't like the idea of a lot of different bottle-holder sizes though, I figure we may want some bigger ones for convenience of not changing them out, but with all the concentrates (which will constitute the cast majority of the bottles), we probably would only need something like 1.5" by 1.5" squares even for the largest bottles.

(03-13-2012 4:52 PM)Fly So Hi Wrote:  Very likely the empty space. This problem goes away with pipettes/tips.

That's what I figured. I noticed that in the video it was a pretty intensely thick liquid that was beading up there. In contrast, pretty much any normal liquid will have a small bubble you can push out to the end with my syringe.

(03-13-2012 4:52 PM)Fly So Hi Wrote:  

(03-13-2012 1:40 PM)gentlmentlemen Wrote:  I meant more just whether it would be physically able to reach a place to touch the drop of liquid to.

Now you got me thinking. Very well observerd. Maybe extra pressure for the small ammounts, so it squirts the concentrate. We'll have a problem if it sprays instead of squirts... I'll think of something.

Well let's see how much of a problem it is first. It may turn out that it only is a problem for fixatives, and then we can focus on a more specific solution for those. One solution (for those) might be to use pipette tips with larger openings for the thicker stuff, but maybe that would just make things worse, idk. But I think that as soon as you touch it to another material (including the surface of the liquid in the mix bottle) then the same strong cohesive and adhesive forces keeping it stuck to the tip of the pipette would be combined with gravity, and if you still have room left with the linear actuator, you could throw in it's force of air pressure by depressing it past the "second stop" to the "third stop" so to speak after you've touched the drop to the liquid. Ought to work. The only significant problem I see right now would be dealing with the variance of drop size, but I would think that would primarily be determined by the carrier, and in addition I'm willing to bet that it will either be fine, or it will be a really large drop compared to the accuracy of the machine.

Still, if we have a way to touch the side of the bottle that would be nice, but then it might have to be a different side for every touch needed.

Which brings me back to my earlier point: lets observe what happens first.

(03-13-2012 4:52 PM)Fly So Hi Wrote:  

(03-13-2012 1:40 PM)gentlmentlemen Wrote:  In both cases, it should be very easy in future designs to add more mix bottles at a time, so that if you're making 2 different mixes at once which both include the same molecule, it doesn't have to rotate the rotary (or go around the track) again for each extra mix bottle. That way it could make 2 mixes in one trip and only make 1 stop per bottle.

I know what you mean, and that's a good idea, i just think we gotta walk before we can run. Let's focus the basics first to keep it simple and practicable.

Agreed. All I meant by that was that if we wind up with a choice between walking on our hands vs walking on our feet, then lets be sure to walk on the ones without opposeable thumbs. If we have to go all frankenstein and piece a thumb onto our foot, so be it, but the ones already built in tend to work better and take less work.

But I'd rather it become a walking freak with weird sewn on thumbs than a cool drawing that can't walk at all of course.

(03-13-2012 4:52 PM)Fly So Hi Wrote:  I usually mix the components in this particular order because a more viscous component like DPG evaporates much slower than Alcohol, for example. The moment i add alcohol to DPG, the alcohol is already "fixed", so it will evaporate less, not spoiling the measures.

Ah... makes sense. I was going to mention something about some of the accuracy being ruined by evaporation... very smart indeed.

(03-13-2012 4:52 PM)Fly So Hi Wrote:  Here's the drawing i wanted to show you:



Fig. 1 - The light grey part is the top layer.

Fig. 2 - The dark grey cover is the middle layer, between the top layer (grid) and the bottom layer (tube with liquid). It slids back revealling the liquid inside the tube.

---

Very cool. So just to be clear on it, its basically the same sliding lid idea I had, but individualised for every bottle? That would actually make things a ton easier... no drip problems... and changing out the covers is as easy as changing out the bottles!

(03-13-2012 4:52 PM)Fly So Hi Wrote:  

(03-13-2012 1:48 PM)gentlmentlemen Wrote:  Also, do you know of any cheap or free CAD type programs we could possibly use to share designs?

I'll google something.

Can't wait! I'll try as well and see what I find. Any way we have of showing each other what we mean and of sharing designs can then later be attached to the posts showing where to get the materials and how to build it, etc.

(03-13-2012 6:21 PM)gentlmentlemen Wrote:  Yup. I was thinking the only other way would be to somehow integrate the pipette tip into the bottle. While that would be cool, it would be annoying to do...

Yes, agreed. That wouldn't be as nice.

(03-13-2012 6:21 PM)gentlmentlemen Wrote:  We would need accuracy relative to the distance which the single volume pipette must travel, and as such the first and most important thing in my mind is to look at how far the button/plunger on the back of pipettes travels for various pipettes, or whether it's all the same.

For fixed volume pipettes, i think the distance is the same. For variable volume ones, the distance changes. Anyway, we could work with a large pipette, like 1000uL. As i see it, the amount of liquid drawn will be based on the number of turns the linear actuator will do to pull up the plunger, not on the volume of the pipette itself, am i right?

(03-13-2012 6:21 PM)gentlmentlemen Wrote:  ...we'd have a grid from somewhere around 270 to 450 square inches for the whole grid, but I'm shooting pretty large there.

How many concentrate bottles are you thinking of?

(03-13-2012 6:21 PM)gentlmentlemen Wrote:  It may turn out that it only is a problem for fixatives, and then we can focus on a more specific solution for those. One solution (for those) might be to use pipette tips with larger openings for the thicker stuff,...

They'll be needed for sure.

(03-13-2012 6:21 PM)gentlmentlemen Wrote:  But I think that as soon as you touch it to another material (including the surface of the liquid in the mix bottle) then the same strong cohesive and adhesive forces keeping it stuck to the tip of the pipette would be combined with gravity

Good point. So the tip must dive into the final bottle. The way i can think of now is if we use a tip rack and after each liquid is dispensed, the tip is replaced.

(03-13-2012 6:21 PM)gentlmentlemen Wrote:  The only significant problem I see right now would be dealing with the variance of drop size, but I would think that would primarily be determined by the carrier, and in addition I'm willing to bet that it will either be fine, or it will be a really large drop compared to the accuracy of the machine.

Remember pipettes work with volumes, not drops. 200uL of crude oil is, to a pipette, the same as 200uL of alcohol. Besides, fixative is not something you need microscopic precision for. And if we implement the item above, this issue is history.

(03-13-2012 6:21 PM)gentlmentlemen Wrote:  Very cool. So just to be clear on it, its basically the same sliding lid idea I had, but individualised for every bottle? That would actually make things a ton easier... no drip problems... and changing out the covers is as easy as changing out the bottles!

Oh that's cool. I just illustrated it because i wasn't sure that's what you meant. Anyway, i drew only one bottle but my idea was that the dark grey part is just one big part serving all bottles.

Roughly like this:



This way we just use a single mechanism to uncover all of the bottles, be it a magnet or whatever.

(03-13-2012 7:32 PM)Fly So Hi Wrote:  

(03-13-2012 6:21 PM)gentlmentlemen Wrote:  We would need accuracy relative to the distance which the single volume pipette must travel, and as such the first and most important thing in my mind is to look at how far the button/plunger on the back of pipettes travels for various pipettes, or whether it's all the same.

For fixed volume pipettes, i think the distance is the same. For variable volume ones, the distance changes. Anyway, we could work with a large pipette, like 1000uL. As i see it, the amount of liquid drawn will be based on the number of turns the linear actuator will do to pull up the plunger, not on the volume of the pipette itself, am i right?

Just to be clear then, the fixed volume ones don't change button depression distance from brand to brand then?

And I was also thinking a 1000uL one. Anything else would have potential to take waaayyyy too long with the fixatives and carriers. Or at least I expect it would with how much we'll be slowing down the linear actuator's motor to achieve the accuracy we want. But time issues outstanding, there's also the issue of increased inaccuracy from having to repeatedly draw liquid and expell it from the same bottle.

(03-13-2012 7:32 PM)Fly So Hi Wrote:  

(03-13-2012 6:21 PM)gentlmentlemen Wrote:  ...we'd have a grid from somewhere around 270 to 450 square inches for the whole grid, but I'm shooting pretty large there.

How many concentrate bottles are you thinking of?

A lot. Total bottles I'm thinking 30 to 50. The main reason I think it should have a lot is because if you look at the DIY stuff and the huge number of molecules being uncovered by Chris and dbot (big list at HtH), we'd need actually quite a bit more than that I think. Even without that insane amount of additions I could probably justify 30 to 40, theres just so freakin many... and thats not even counting the various cover fragrances for adding, and at least 2 carriers.

Which reminds me: it should remember which bottle type you used for later. There's a lot of stuff like that it should remember though... gah. I've got to remember to leave that user interface for last.

Either way, I'm thinking 30 to 50. It sounds crazy, but if its going to become the main way I make mixes, having less than that many might require semi frequent to frequent changing of bottles. Maybe that isn't so bad though...

For the first design (that uses the grid anyways), I'm thinking like 2 rows of 5. First row for testing if it can make it down the track, 2nd row for testing if it can make it around the bend in the track.

(03-13-2012 7:32 PM)Fly So Hi Wrote:  They'll be needed for sure.

We'll see. I hope not but I expect you're right.

(03-13-2012 7:32 PM)Fly So Hi Wrote:  

(03-13-2012 6:21 PM)gentlmentlemen Wrote:  But I think that as soon as you touch it to another material (including the surface of the liquid in the mix bottle) then the same strong cohesive and adhesive forces keeping it stuck to the tip of the pipette would be combined with gravity

Good point. So the tip must dive into the final bottle. The way i can think of now is if we use a tip rack and after each liquid is dispensed, the tip is replaced.

Well, if the machine can accurately enough keep track of the height of the liquid in the final bottle, then it ought to be able to just touch the drop to the surface, but we've already accumulated more than enough reasons I think for replacing the tips, so I say plunge it in.

(03-13-2012 7:32 PM)Fly So Hi Wrote:  

(03-13-2012 6:21 PM)gentlmentlemen Wrote:  The only significant problem I see right now would be dealing with the variance of drop size, but I would think that would primarily be determined by the carrier, and in addition I'm willing to bet that it will either be fine, or it will be a really large drop compared to the accuracy of the machine.

Remember pipettes work with volumes, not drops. 200uL of crude oil is, to a pipette, the same as 200uL of alcohol. Besides, fixative is not something you need microscopic precision for. And if we implement the item above, this issue is history.

I was talking about the size of the drop which hangs off the pipette at the end with the thicker liquids, which depends on surface tension.

But agreed, lets just implement the method of plunging the tip in slightly and be done with it. If we're replacing the tips anyways, it should pretty much be a non-issue. The only thing I can think of would be residue on the sides of the tip of the pipette adding to the amount put into the final mix bottle, but if they don't worry about that with the .2uL accuracy ones, then I figure it probably isn't really an issue. If it is, then we'll just minimize the amount that goes below the surface --mainly on the expulsion that is. The stuff on the outside of the tip may be an issue, but the air bubbles from not keeping the tip submerged in the concentration bottle are a known issue.

(03-13-2012 7:32 PM)Fly So Hi Wrote:  Oh that's cool. I just illustrated it because i wasn't sure that's what you meant. Anyway, i drew only one bottle but my idea was that the dark grey part is just one big part serving all bottles.

Roughly like this:



This way we just use a single mechanism to uncover all of the bottles, be it a magnet or whatever.

Looks good, although I have to say I'm pretty sold on the single bottle cover idea now. I just see the changing of lid covers when you change what bottle is in what slot and switch one out being a lot easier if each bottle has its own. Could allow for a more custum seal on each bottle too which would probably work better. Gets around any potential drip issues from condensed liquid getting on the cover, or if the liquids get sloshed onto the top from the thing being bumped...

If we did single bottle covers, I would probably use the motor for raising and lowering the pipette tips into the bottles for opening the covers of the individual bottles instead of the track motor.

Also, this way only the bottle being worked with at any given time is open. Not only does that mean that the other bottles won't be open with their contents evaporating into the air during the mix making process, but it also means great things for if the whole rig gets bumped.

With the "one lid to rule them all" lid, if the machine gets bumped while mixing, every single bottle would get sloshed. Thus, one little accident could lead to a lot of wasteage of product and money, as well as contamination of a lot of the bottles.

If each bottle gets its own lid though and it gets bumped, then whatever bottle is open at the time will still get sloshed and there will still be some wasteage of product, but it will be far less and in addition there will be no risk of it getting into the other bottles.

The main reason I thought we should just have one lid at first was because I knew I didn't want to deal with the screw caps of each bottle, but I think we could come up with a cap that stayed with the bottle-holder and was attached to it wi either a sliding track or a hinge.

The lowering of the pipette tip towards the bottle could open the cap, and the raising could close it. I'm thinking a hinge would be easier and cheaper for a single bottle cap design though. What type of securing of the cap do you think would work best? I'm thinking some type of latch one, like a snap close cap. The motor ought to still be able to operate that if the cap extends past the hinge or past the latch to give the motor lowering the pipette some leverage. Then it would just be a matter of a little stationary arm which preceeds and pipette tip to open/close the cap. At first I was thinking of maybe using some rubber bands to have the cap be self-closing, but then I thought that might be less reliable.

That reminds me of another problem which has been bouncing around in my head for the multi-cap lid solution. If it is all one lid, I would expect that a decent amount of force would need to be applied to make sure it is sealing all the bottles. If that force is not applied evenly, it may take a lot of force to seal the bottles which don't recieve much of that force. Of course I suppose you could just add some lead weights above each bottle. That would be simple enough.

Still, I think it would also have a tendency to have worse seals unless we used only one bottle type, simply because it would be difficult to adjust the middle bottles' height correctly (which would be harder to view once more bottles are in there).

But the main reason is because that looming accident scares me. Having the entire phero collection all open at once is just asking for disaster to strike with her cruel hand imo.

The one other thing I forgot to mention is that in my seal's design I had pictured an uncut square of the sealing material being pressed down on the bottle's top. I think yours would work better (with the hole in the middle) for bottles of a specific top size and shape, but when dealing with a multitude of bottles I don't know how that would work.

(03-12-2012 2:31 PM)shadowknight Wrote:  I have several websites to get bottles not sure about the which the vendors use
http://www.sks-bottle.com/
http://www.lotioncrafter.com/
http://www.specialtybottle.com/
http://www.newdirectionsaromatics.ca/
http://www.sunburstbottle.com/
http://www.essentialsupplies.com/

I would like to add this website, good for some ultra small bottles that fit in the tightest pocket, as well as 3 ml glass atomizers:
http://www.pilotvials.com/

New Directions Aromatics has awful customer service.