Page 1 of 2 12 LastLast
Results 1 to 10 of 12

Thread: Please Critique Yet Another Universal Vac Hold Down Design

  1. #1
    Join Date
    Oct 2006
    Location
    Five Quarter Studio, Inc., Saugerties NY
    Posts
    33

    Default Please Critique Yet Another Universal Vac Hold Down Design

    I've got my new-to-me Desktop up and running. While I wait for my electrician's visit to add more circuits for Bot and vac(s), I've been working on universal vacuum hold down because none of the designs posted here meet all my reqts:

    1. Unobstructed front access
    2. Unobstructed feed-through for tiling longer parts
    3. Preserve max Z travel
    4. Vac hookup from rear
    5. No irreversible alterations
    6. Easy to cut, assemble, disassemble, and modify
    7. Easy to swap out with original T-slot rig

    The key to what I've sketched is to make use of the no-fly area to the left of the cutting volume. Please take a look and tell me if you spot anything unworkable or questionable.

    Thanks,
    Joel




    Attached Images Attached Images

  2. #2
    Join Date
    Mar 2004
    Location
    Delray Beach, FL
    Posts
    3,708

    Default

    Could be on the right track but sooner or later that raised bump for the vac hose will get bit unless it is totally off the move pattern of your spindle.

  3. #3
    Join Date
    Oct 2006
    Location
    Five Quarter Studio, Inc., Saugerties NY
    Posts
    33

    Default

    Quote Originally Posted by bleeth View Post
    Could be on the right track but sooner or later that raised bump for the vac hose will get bit unless it is totally off the move pattern of your spindle.
    For sure. I need to measure where the collet nut and spindle are relative to the cut volume boundary with my lowest likely cut scenario. Then adjust the outlet block to be clear.

  4. #4
    Join Date
    Jan 2004
    Location
    iBILD Solutions - Southern NJ
    Posts
    7,911

    Default

    You don't want to feed it from the top...Feed it from underneath, even if you have to use a few smaller tubes (1") to feed it. Emulate the bigger systems as much as you can.

    -B
    High Definition 3D Laser Scanning Services - Advanced ShopBot CNC Training and Consultation - Vectric Custom Video Training IBILD.com

  5. #5
    Join Date
    Oct 2006
    Location
    Five Quarter Studio, Inc., Saugerties NY
    Posts
    33

    Default

    Brady,

    Would you please explain why a top routing is bad? I'm not knowledgeable about fluid dynamics, but I have read that fewer sharp bends is better. This design has no bends at all from the edge of the plenum grid to the vac hose attachment. I've moved the outlet block further back as Dave suggested. Is there something else I'm missing?

    Can you point me toward an image or drawing of the under table routing you recommend for the D2418?

    Thanks,
    Joel

  6. #6
    Join Date
    Aug 2004
    Location
    Northern , new jersey
    Posts
    220

    Default

    search forum Desktop vacuum or this http://www.shopbotblog.com/2006/11/a...own-potential/ for some ideas
    ArtCam Pro 9
    VCarve Pro 8

  7. #7
    Join Date
    Jan 2004
    Location
    iBILD Solutions - Southern NJ
    Posts
    7,911

    Default

    Quote Originally Posted by joelschuman View Post
    Brady,

    Would you please explain why a top routing is bad? I'm not knowledgeable about fluid dynamics, but I have read that fewer sharp bends is better. This design has no bends at all from the edge of the plenum grid to the vac hose attachment. I've moved the outlet block further back as Dave suggested. Is there something else I'm missing?

    Can you point me toward an image or drawing of the under table routing you recommend for the D2418?

    Thanks,
    Joel
    Joel,
    When you feed it from the top, there's always the issue of sealing the intake from where it feeds the plenum and bleeder. By feeding it from below, you avoid all of that & make things simpler and easier to seal. This is of course talking about a high CFM 'universal' vacuum system, like those run on large machines. It is way easier to do a pex fed high HG/low CFM setup from the top (or even the bottom) - depending on what you are cutting.

    For a machine this small, with a shopvac-type vac source, you really only need one zone. You can feed the vacuum down from the top, between the last table extrusion (which is only 1.5 x 1.5) and pipe it over to the left or right side of the Y lead screw. From there you can pipe it up to the table by removing one of the 1.5x6" table extrusions and replace one with an 8020 1530 extrusion. You'll have a 3" gap in the middle of the table for piping up the vacuum.

    For the top, to prevent too much loss of Z, I would use 1/2" material for the plenum & 1/2" ultralight for the bleeder. Set it up so you can still bolt down to the underlying extrusions from the top. You'll need to make islands around where the bolts go to act as a sealant to bleeding off vacuum. The vac grid itself doesn't have to be deeper than 3/16" or so...in fact, I doubt there's much to gain over the grid being deeper than 1/8" (even less). Seal all the edges and bottom of plenum for max performance.

    1x3" AL rect tube is a great way to feed from front to back just like I did the BT48 system

    Attached crude pic shows what I am talking about...




    -B
    Attached Images Attached Images
    High Definition 3D Laser Scanning Services - Advanced ShopBot CNC Training and Consultation - Vectric Custom Video Training IBILD.com

  8. #8
    Join Date
    Jan 2004
    Location
    iBILD Solutions - Southern NJ
    Posts
    7,911

    Default

    Quote Originally Posted by joelschuman View Post
    Brady,

    Would you please explain why a top routing is bad? I'm not knowledgeable about fluid dynamics, but I have read that fewer sharp bends is better. This design has no bends at all from the edge of the plenum grid to the vac hose attachment. I've moved the outlet block further back as Dave suggested. Is there something else I'm missing?

    Can you point me toward an image or drawing of the under table routing you recommend for the D2418?

    Thanks,
    Joel
    Joel,
    When you feed it from the top, there's always the issue of sealing the intake from where it feeds the plenum and bleeder. By feeding it from below, you avoid all of that & make things simpler and easier to seal. This is of course talking about a high CFM 'universal' vacuum system, like those run on large machines. It is way easier to do a pex fed high HG/low CFM setup from the top (or even the bottom) - depending on what you are cutting.

    For a machine this small, with a shopvac-type vac source, you really only need one zone. You can feed the vacuum down from the top, between the last table extrusion (which is only 1.5 x 1.5) and pipe it over to the left or right side of the Y lead screw. From there you can pipe it up to the table by removing one of the 1.5x6" table extrusions and replace one with an 8020 1530 extrusion. You'll have a 3" gap in the middle of the table for piping up the vacuum.

    For the top, to prevent too much loss of Z, I would use 1/2" material for the plenum & 1/2" ultralight for the bleeder. Set it up so you can still bolt down to the underlying extrusions from the top. You'll need to make islands around where the bolts go to act as a sealant to bleeding off vacuum. The vac grid itself doesn't have to be deeper than 3/16" or so...in fact, I doubt there's much to gain over the grid being deeper than 1/8" (even less). Seal all the edges and bottom of plenum for max performance.

    1x3" AL rect tube is a great way to feed from front to back just like I did the BT48 system

    Attached crude pic shows what I am talking about...

    SBDTVac.jpg


    -B
    High Definition 3D Laser Scanning Services - Advanced ShopBot CNC Training and Consultation - Vectric Custom Video Training IBILD.com

  9. #9
    Join Date
    Oct 2006
    Location
    Five Quarter Studio, Inc., Saugerties NY
    Posts
    33

    Default

    Brady,
    Big thanks, O Vacuum Sensei! I'll revise my plan. "Easier to seal" is good - my design certainly wasn't that. I was hoping a tube of removable sealant would handle it, but that would be inelegant at best.
    I'm grateful for your take on plenum grid depth. I had been planning to do a 1" grid 1/2" deep in 3/4" MDF with a 1/4" ball nose.
    You also touch on preserving Z, which is a big concern. In addition to cutting parts out of thin (1/6" - 1/4") plywood and other sheet goods for products, I want to resume the 3D sculpture I used to do with my PRT120x60 in my spare time. For that, Z travel limits the max thickness of the blocks I can carve (without having to layer-cake). I was planning to remove the 1-1/2" thick aluminum T-slot extrusions entirely and replace with 3/4" MDF plenum laminated to a 3/4" BahamaPly base, which saves at least 1-1/2" of Z and eliminates the bolt seal issues. Based on your feedback, I'll change the plenum to 1/2" MDF with 1/4" round grid on 3/4" plywood. In past, with my big plenum, I found that cutting the grid with a ball nose pretty much eliminated the chance of having a grid island break off and head for the pump.
    From what you're saying, sounds like 3/4" rigid PVC pipe would be big enough when using a Fein Turbo I for suck. Is that correct or can I go smaller? There'd be 3 90º elbows (or maybe 2 90º and 1 45º) - do you think that's okay? I'm not experienced with PEX - what size PEX with the Fein? Elbows or heat bending or bend supports?
    Cheers,
    Joel

  10. #10
    Join Date
    Jan 2004
    Location
    iBILD Solutions - Southern NJ
    Posts
    7,911

    Default

    Hey Joel,
    I am by no means an 'expert' on anything...LOL! I guess it takes being an 'expert' to realize how much a load that really is...

    I'm still learning...Just like anyone else who machines things. Just last week, I was machining a large 3D panel with a lot of intricate detail and had to toss out about 20hrs of machining because the vacuum pump I used both sucked out moisture and compressed the ultralight bleeder board by as much as 0.100" (!!!!). Yeah...a kick square in the sweetbreads. I machined off my 'getting thin' 5/16" bleeder and put a new one down (5x12') - only to have the same issue happen a 2nd time, albeit not quite as bad @ 0.04" dip. I did a 3D rough out with .03" allowance left on for the finishing pass, and I witnessed the roughing 'meat' barely being taken off at some point in the Y, doing an X raster. I had to do some black magic using Nudge, pausing the machine every .05" and Nudging it down .002" (about 15 times) to get my .030" back. In the end it worked out...but it's something that I'll need to address soon for long cycle time parts, such as this one where the finishing pass alone took 13hrs+.

    The pump I used for that job maxes out about 9.5 Hg". It would have been worse if I used the Becker which maxes out at 25.5Hg". I never had an issue with ultralight MDF compression using a Fein Turbo III or Lighthouse motor simply because they just don't generate enough suction to compress things down. I pulled about 6 Hg" on the gauge (MDF machined part was bleeding some) so I imagine that compression could be an issue with the Fein if left long enough. Vacuum does suck moisture out of the board too...and this causes it to change dimension. My good friend in the Keys told me that he'd go through an ultralight bleeder over the course of 3 days processing 100 or so sheets for cabinets using Lighthouse motors. He would get the compression issue if for no other reason than it sucked out all the moisture - his dados and rabbets would go up and down prompting him to resurface.

    So what about the $500k+ CNC machining centers that use vacuum day in & day out? What do they use? They use regular MDF and multiple Becker pumps that pull 25Hg"+, with a usable vacuum (delta) in the 12-14Hg" range (gauge reads somewhere around 11-13 Hg" with all zones open and NOTHING on top). The regular MDF is a lot denser and doesn't crush down as bad as the ULMDF or LDF. So...if you have a Fein or low Hg" pump, you can't really use regular MDF because it just doesn't bleed enough and your vacuum source can't suck hard enough for it to bleed and hold parts down.

    On a small machine like the DT, there isn't a lot of real estate to keep flat - Plus, the extruded table really helps to keep the area consistently flat. I'm still rockin my original spoilboard on my DT because I only have to take off like .005-.01" to surface it perfectly flat. There's at least 5/8 of the original 3/4" there. I like the stability of the DT so much I am considering making a similar table for the big PRT (5x12) out of phenolic, 1/2" aluminum sheet or a bunch of extrusion. More on that as it develops.

    I have a few repeat jobs where I need a 24x32" area with as much vacuum as I can get to hold down small parts. I no longer use this jig, but did successfully in the past with the Fein. It's just a piece of 1/4" PVC (Sintra) with screw holes countersunk around the perimeter and vac channels created with an 1/8" bit. This left squares around 1/4 to 5/16" or so and the depth was somewhere around .090. It worked great with a piece of Trupan utralight or a vacuum mask on top. I don't know what the bottom end is for minimum grid depth, but you really don't need that much. It all boils down to CFM...

    The grid depth and plumbing tubing diameter (and elbows etc) directly affect airflow or CFM. If you just had a 1/4" tube plugged into the Fein, it wouldn't flow much air, which wouldn't let you create a vacuum seal, and there'd be little air to buffer leakage from the cutting kerf. Furthermore, the Fein would be sitting there cavitating, generating a lot of heat and under load without much work actually being done. IIRC a 2" pipe will flow about 200 CFM or so freely. A 1.5" dia pipe will flow about 100 CFM freely - that is to say, the vacuum will not generate negative pressure though the pipe if it's CFM is less than the free flow of a given diameter pipe. CFM is less of a concern for vacuum hold down because once negative pressure starts to build (a split second) - airflow falls off dramatically as it approaches 0 CFM. There still needs to be the 'possibility' for airflow as you're cutting, meaning the grid needs to be capable of moving air if it needs to, in order to make up for any leakage. 100 CFM across the DT machining area is a LOT of airflow - so no worries about not having enough.

    If Z height is a concern, I would consider a few things. First, It is completely possible to make a good performing low vacuum table using the BradyVac system. You could machine a grid into 1/2" or 3/4" ultralight MDF. Then glue (around perimeter AND on top of the squares) a piece of 1/4" MDF or any other glueable material, to the grid. Machine a hole or holes into the 1/4" material as your vacuum intake. If you want to bolt it down like the spoilboard, just use CAD to lay out the bolt locations, and leave that area alone when you machine in the grid. Make a boss around the countersunk bolt holes so the grid never goes through it. Vacuum will creep around those areas so seal the bolt hole walls and put a little rubber washer under the bolt head. I would paint or seal off the 1/4" bottom material so I didn't lose any vacuum through the bottom. You can avoid losing the corners of the grid by machining with a downcut spiral. A bit exotic, but you could replace the table extrusions with 1" or 20mm ones available from various sources. As long as you aren't machining anything too heavy (block of iron wood?) you could gain 1/2 to 3/4" extra Z height by lowering the table...

    I wouldn't worry too much about elbows and turns decreasing performance. It's only a hit on airflow & it doesn't really matter...if it was a race car manifold, that'd be different. If you can't get a 2" pipe in there, use a few 1.5 or 1" tubes to make up for the diametrical surface area in order to achieve the same flow. You can have multiple ports and they don't have to be on center (Y leadscrew location prevents this anyway).

    For those thin parts, I use the Vacuum Film Technique where I cover the back of the substrate with masking tape, then carpet/transfer tape, then mount it all on a piece of cheap 4mm coroplast. You can cut all the way into the first skin of coroplast and never break vacuum. Then simply peel off your parts or give to the customer mounted on the board for them to peel off.

    Sorry so long winded...

    -B
    High Definition 3D Laser Scanning Services - Advanced ShopBot CNC Training and Consultation - Vectric Custom Video Training IBILD.com

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •