entropy
having dreams about rocks
Member since June 2020
Posts: 71
|
Post by entropy on Jul 3, 2020 17:51:21 GMT -5
I chucked up that scrap piece of 8” DR-11 (the small leftover section from my 24 foot stick). I turned it in my small 13” lathe. Observations: The material is quite flexible. I used a 3-jaw self-centering chuck, and clamped it from the inside. The chuck bulged the pipe – not a lot – but visible to the eye. Note that the plastic immediately springs back to original size when removed from the chuck (aka – the plastic is not yielding). Note that if I had machined the OD of the pipe in the near-vicinity of the chuck, it probably would’ve left a witness-mark in the pattern - upon removal from the chuck. I used a small carbide insert intended for turning mild steel. I did not use coolant. When people say HDPE is easy to machine, it’s an understatement. HDPE is super free-machining. The lathe cuts it effortlessly. I faced the end with an aggressive feed. The finish looks great. And had I used a sharp HSS bit, I could probably get it even smoother – but I’m not sure that’s necessary. The big question being - is the lathe-finish acceptable for welding? I’m thinking it is. I turned the OD a bit – just to see what happens. Again, with an aggressive feed, and the finish looks great. The chips are one long continuous ribbon of plastic. If you let them get too long, they will rat-nest on the cutter - or simply wrap around the chuck. What's nice is you can grab the turnings with your bare hand and not get cut (unlike steel chips - which are one giant coil of razorblade) I didn’t attempt any small skim-cuts. I was just tickled it cut so easily.
|
|
EricD
Cave Dweller
High in the Mountains
Member since November 2019
Posts: 1,142
|
Post by EricD on Jul 3, 2020 18:45:29 GMT -5
Very cool to see your experimentation and progress! Having never welded HDPE I cannot comment on if the finish you achieved will work well, but it does look very smooth!
Since HDPE welding essentially melts the HDPE One would think the finish you got would be more than smooth enough
|
|
entropy
having dreams about rocks
Member since June 2020
Posts: 71
|
Post by entropy on Jul 3, 2020 19:00:05 GMT -5
|
|
|
Post by knave on Jul 3, 2020 19:12:06 GMT -5
It’s plenty smooth. I’ve done a ton of small diameter (1.25) socket fusion on HDPE pipe that looked fairly rough, ie it was drug through the ground with a boring rig and it it welded just fine. Awesome stuff man.
|
|
entropy
having dreams about rocks
Member since June 2020
Posts: 71
|
Post by entropy on Jul 4, 2020 0:00:04 GMT -5
I started the process of fabricating my heating plates (for welding the tumbler barrels together). Starting material is 1-3/8" steel plate. I'm machining the raised-face features required for heating the end-caps (aka the 8" outer perimeter, and the 4" inner perimeter only) - also for heating the pipe ends. The plate temperature must be maintained between 400 and 450 degrees F. I'm going to use electric range heating elements as the heat source. My thought for using such thick steel, is that the large mass of the plates will provide some thermal dampening (better temperature control). Unfortunately I must incorporate a non-stick feature.
I have two options for the non-stick, and I'm not sure which way to go.
Option 1 - use thin PTFE fabric cloth over the steel. This is actually how some commercial HDPE welding mirrors are fabricated. The cloth is good to 600 degrees F, and is cheap to replace. The only potential downside is that I'm not covering a flat plate. It has a raised-face feature. Will this be a problem? I don't know. The cloth is pretty thin, and I'll be able to see the impression of the raised-face.
Option 2 - get a PTFE coating applied to the plates. There are two sub-options here. The less-expensive is a single-stage PTFE coating, but it's only good to 400 degrees F max (no good). This means I need the more expensive two-stage coating (base & top). Another potential issue is that the PTFE coating will eventually fail, and I'll have to send the plates out again. I've read it normally takes thousands of welding cycles for the coating to fail, and I've only got 45 joints to make - thus life expectancy shouldn't be a big concern. However, I've also read that sometimes the PTFE coatings fail prematurely - without warning.
I'll probably try the fabric cloth first (a few dollars, vs a few hundred for the coating).
|
|
entropy
having dreams about rocks
Member since June 2020
Posts: 71
|
Post by entropy on Jul 4, 2020 0:10:20 GMT -5
I suppose option three would be to get a 900 degree oven, and apply the PTFE myself.
I don't think I want to mess with applying Teflon (health & environmental reasons).
|
|
entropy
having dreams about rocks
Member since June 2020
Posts: 71
|
Post by entropy on Jul 4, 2020 1:56:27 GMT -5
jamesp, question - You've recommended 60 RPM for 6 inch barrels, and 45 rpm for 8 inch barrels (for rough-cutting). The 8" DR-11 pipe is 7" ID. Linear interpolation between values has the 7" ID pipe turning about 53 rpm. Does this seem right?
|
|
EricD
Cave Dweller
High in the Mountains
Member since November 2019
Posts: 1,142
|
Post by EricD on Jul 4, 2020 3:43:56 GMT -5
"@ jamesp" without the space will get his attention much faster
|
|
entropy
having dreams about rocks
Member since June 2020
Posts: 71
|
Post by entropy on Jul 4, 2020 13:32:59 GMT -5
The ideal tumbler drive is staggered rollers, with the lower roller pushing the barrel towards the upper. It is unnecessary to power both rollers, and it would be a significant error to drive just the upper roller. The reasoning behind this can be explained by with a simplex brake-drum analogy. However I got some pushback on this. (Please don’t be mad – as the below picture is straight off Wikipedia) The best way to explain it (while completely avoiding the brake drum comparison) is to show a better picture of what’s happening. This morning I did an engineering analysis of staggered tumbler bars. Below is the free-body diagram, and below that are derived equations. Important note: the force (Ft1) required to rotate the idler (roller 1) is purely a function of friction within the bearings. Install some good bearings, keep them greased, and the shaft will rotate freely (minimizing Ft1, and reducing overall power-input). EDIT: The RPM in the horsepower equation is the RPM of roller 2, and specifically not the rpm of the tumbler barrel. (edit complete). EDIT: I made an error on my initial solution. Below is the corrected copy. Discussion: you can see that Ft2 increases as Ft1 increases (because roller 2 is driving the tumbler into roller 1 - aka a self-engaging wedge). If you were to drive roller one, the friction-force would decrease, because you'd actually be turning the tumbler away from roller 2, and not into it. (edit complete)
|
|
jamesp
Cave Dweller
Member since October 2012
Posts: 36,154
|
Post by jamesp on Jul 4, 2020 14:04:46 GMT -5
jamesp, question - You've recommended 60 RPM for 6 inch barrels, and 45 rpm for 8 inch barrels (for rough-cutting). The 8" DR-11 pipe is 7" ID. Linear interpolation between values has the 7" ID pipe turning about 53 rpm. Does this seem right? These speeds have a range. I have run 6 inch up to 80 rpm, 8 inch up to 60 rpm. keep in mind I run the barrels 80 to 85 percent full. Much different than a barrel filled to 60%. 60% fill creates a very violent tumble. I also run a clay slurry thickener that stays consistent for weeks. This thickener offers a lot of protection at high speeds.
|
|
entropy
having dreams about rocks
Member since June 2020
Posts: 71
|
Post by entropy on Jul 4, 2020 14:18:12 GMT -5
jamesp , I recall you saying (don't remember which thread) that when doing a fast-tumble rough-cut, that the grit will basically consume in a few days - even though you need to tumble for one, two, or even three weeks. Question - which stages is the clay slurry utilized? So after a few days of rapid tumbling in stage 1, and the grit is essentially pulverized - do you simply add more SIC grit into the barrel, and keep going? Or do you empty, rinse, and add both fresh clay & grit every few days? How do the variables change for fine tumbling in stage 3 and 4? (% fill, clay mixture (if applicable), and RPM)? Do you purchase the clay in bulk? Pottery supply?
|
|
entropy
having dreams about rocks
Member since June 2020
Posts: 71
|
Post by entropy on Jul 4, 2020 14:21:51 GMT -5
My little 15 pound tumbler turns at 21 rpm. Non-adjustable. I ran it for a week (stage one), and after that week 100% of the grit was gone (the water looked like mud). This makes me wonder how much time I was wasting - tumbling rocks with no grit in the mix. Perhaps I should've opened it up after day 3, and added more grit?
|
|
jamesp
Cave Dweller
Member since October 2012
Posts: 36,154
|
Post by jamesp on Jul 4, 2020 15:24:06 GMT -5
I started the process of fabricating my heating plates (for welding the tumbler barrels together). Starting material is 1-3/8" steel plate. I'm machining the raised-face features required for heating the end-caps (aka the 8" outer perimeter, and the 4" inner perimeter only) - also for heating the pipe ends. The plate temperature must be maintained between 400 and 450 degrees F. I'm going to use electric range heating elements as the heat source. My thought for using such thick steel, is that the large mass of the plates will provide some thermal dampening (better temperature control). Unfortunately I must incorporate a non-stick feature. I have two options for the non-stick, and I'm not sure which way to go. Option 1 - use thin PTFE fabric cloth over the steel. This is actually how some commercial HDPE welding mirrors are fabricated. The cloth is good to 600 degrees F, and is cheap to replace. The only potential downside is that I'm not covering a flat plate. It has a raised-face feature. Will this be a problem? I don't know. The cloth is pretty thin, and I'll be able to see the impression of the raised-face. Option 2 - get a PTFE coating applied to the plates. There are two sub-options here. The less-expensive is a single-stage PTFE coating, but it's only good to 400 degrees F max (no good). This means I need the more expensive two-stage coating (base & top). Another potential issue is that the PTFE coating will eventually fail, and I'll have to send the plates out again. I've read it normally takes thousands of welding cycles for the coating to fail, and I've only got 45 joints to make - thus life expectancy shouldn't be a big concern. However, I've also read that sometimes the PTFE coatings fail prematurely - without warning. I'll probably try the fabric cloth first (a few dollars, vs a few hundred for the coating). Stop by Home Goods (a discount department store) and look at the plethora of teflon coated frying pans and cook plates. most are less than $10. Sure looks like perfect candidates as high heat transfer/non-stick heat sinks. Most are thick machinable aluminum with the tough teflon coating. I had great success using a teflon frying pan on the electric stove. I found if I put the pan on the burner and the pipe on the pan and then turned it to high(burner cherry red). As soon as it turned cherry red I would crank the heat down till the burner was not cherry red and let them sit for 5 to 10 minutes to soak heat I got the best repeatability in welds. Or put the burner on low and set the pipes on the pan and let them soak for 15 minutes. Then crank heat up till they were soft. If they stick to the teflon pan or leave plastic on the pan you were too hot. A shiny smooth surface was the sweet spot. " The surface temperatures must be in the temperature range 400 - 450°F (204 - 232°C). Install the heater in the butt fusion machine and bring the pipe ends into full contact with the heater at fusion pressure to ensure that full and proper contact is made between the pipe ends and the heater. After holding the pressure very briefly, it should be released without breaking contact. On larger pipe sizes, fusion pressure must be maintained until a slight melt is observed around the circumference of the pipe before releasing pressure. Continue to hold the components in contact with each other, without force, while a bead of molten polyethylene develops between the heater and the pipe ends. A minimum of 4.5 minutes per inch of pipe wall is recommended as the minimum heat soak time. Then continue the heat soak cycle until the minimum bead size is formed against the heater on both sides (see table)."
|
|
jamesp
Cave Dweller
Member since October 2012
Posts: 36,154
|
Post by jamesp on Jul 4, 2020 15:40:04 GMT -5
jamesp , I recall you saying (don't remember which thread) that when doing a fast-tumble rough-cut, that the grit will basically consume in a few days - even though you need to tumble for one, two, or even three weeks. Question - which stages is the clay slurry utilized? So after a few days of rapid tumbling in stage 1, and the grit is essentially pulverized - do you simply add more SIC grit into the barrel, and keep going? Or do you empty, rinse, and add both fresh clay & grit every few days? How do the variables change for fine tumbling in stage 3 and 4? (% fill, clay mixture (if applicable), and RPM)? Do you purchase the clay in bulk? Pottery supply? At 60 rpm/6 inch barrel 30 grit was gone in 24 hours. raw SiC might go 4 days to a week depending on size of rocks in tumbler. But the raw is deceptive because most of the grit was broken down, just the big chunks were left after 4 days. so if using 30 I add a half dose every 1 to 2 days. if the slurry has gotten to thick I pour off some slurry and thin it with water. usually add a half dose of raw every 3 to 4 days. The clay is only used in the rotary when rolling SiC raw-30-46-220-500. Then I move rocks to vibe step 3-4 using alum ox with sugar thickener. Georgia red clay from most any source in a road bank or exposed clay spot. It is a kaolin, so colloidal. Some has a higher sand content. High in iron so high in conductivity: I like the clay with a bit of sand: www.flickr.com/photos/67205364@N06/sets/72157651621802089For sand free clay a dried silt pond: www.flickr.com/photos/67205364@N06/sets/72157676639073297
|
|
|
Post by HankRocks on Jul 4, 2020 15:46:23 GMT -5
My little 15 pound tumbler turns at 21 rpm. Non-adjustable. I ran it for a week (stage one), and after that week 100% of the grit was gone (the water looked like mud). This makes me wonder how much time I was wasting - tumbling rocks with no grit in the mix. Perhaps I should've opened it up after day 3, and added more grit? Several folks out here re-charge their grit, depends on how much grit you put in initially and the mix of material you are running. For a load of decently rounded agates I put in about 6 tablespoons in a 15 pound barrel. I will check it at 3, 4 or 5 days, if grit is gone I will add 4 or 5 tablespoons and let it run for 5 more days(or so). I usually run it over by a day or two after the 2nd re-charge to let the broken down grit get the rocks ready for my next stage. Opening it for a re-charge let's me check the slurry thickness. For a load of slabs and preforms, I do not do a re-charge, letting it run for 10 days or so as the grit breakdown is much slower. Please note, there is no perfect formula for tumbling. Everyone needs to learn their tumbler, the mix of the rocks being tumbled, the length of time, the grit size, etc, etc. I am repeating myself from a week ago, I bet there are 20 to 30 or more different processes used by folks out here and all of them can produce well polished rocks. It's almost a tumble by experience. Even I do not have one process I follow and will vary it for numerous reasons, sometimes just because I want to know how something will work or not work. Best trait one can have when tumbling rocks is patience. Remember, the rocks really don't mind rolling for extra days. Good luck Henry
|
|
entropy
having dreams about rocks
Member since June 2020
Posts: 71
|
Post by entropy on Jul 4, 2020 15:49:26 GMT -5
jamesp, how would one in Washington state get some Georgia red clay - without actually leaving Washington state? Would Bentonite work (the stuff used for sealing well casings)? I thought Bentonite clay was axle grease the first time I saw it. I'd imagine it would make for an excellent slurry additive. I believe it comes in dry pellets - until you get it wet. Then it acts like extremely thick grease.
|
|
jamesp
Cave Dweller
Member since October 2012
Posts: 36,154
|
Post by jamesp on Jul 4, 2020 15:51:44 GMT -5
|
|
jamesp
Cave Dweller
Member since October 2012
Posts: 36,154
|
Post by jamesp on Jul 4, 2020 15:56:58 GMT -5
jamesp, how would one in Washington state get some Georgia red clay - without actually leaving Washington state? Would Bentonite work (the stuff used for sealing well casings)? I thought Bentonite clay was axle grease the first time I saw it. I'd imagine it would make for an excellent slurry additive. I believe it comes in dry pellets - until you get it wet. Then it acts like extremely thick grease. Bentonite, Florida gumbo, pure white kaolin, any of those will work. Best not to re-use it. Add fresh clay after clean out. Look up colloidal slurry for abrasive operations. Interesting how people make big money selling clay in suspension mixed with a pretty dye.
|
|
entropy
having dreams about rocks
Member since June 2020
Posts: 71
|
Post by entropy on Jul 4, 2020 16:20:36 GMT -5
jamesp, Do you ever try and recover/reuse grit during cleanouts? Example - assume I'm running SIC-30 (with Bentonite) turning 8" DR-11 barrels 85% full at 55-ish rpm - for three days. Is it worth trying to salvage any of the remaining grit after 3 days? I'm assuming the gold-panning technique would work - aka washing away the clay, and recovering any loose grit for reuse. This seems cumbersome. Grit seems cheap (relatively - if purchased in bulk). Is it worth the effort to reuse grit?
|
|
EricD
Cave Dweller
High in the Mountains
Member since November 2019
Posts: 1,142
|
Post by EricD on Jul 4, 2020 17:01:11 GMT -5
Jamesp, I was thinking about your idea for 10 foot bar spacing. I did some calculations. Generic bars are shown in the sketch, at 10 foot spacing. A “very heavy” tumbler barrel of weight (w) is placed in the center (representing the weight required to bend the bars). An interesting design consideration is the angle in which the bars make with the center-axis of the barrel. Too narrow of angle, the barrel might tip off. Too wide, and you can imagine the barrel slipping between the bars. I did a calculation for 30 degrees, 45 degrees, and 90 degrees. You can see that the resultant force applied to the bar increases with a given barrel weight - as the angle increases. My assumption is that for a real-world application, you’d probably want something between 30 and 45 degrees. Once you pick your bar spacing, the in-service angle becomes a function of barrel diameter you're running. But I digress. A pair of 1.5” solid steel bars (10 foot each) will weigh 121 pounds total. A pair of 2” solid steel bars will weigh 216 pounds total. A pair of 2.5” NPS Schedule 40 pipe will weigh 117 pounds total. For the 1.5” solid steel bars, the tumbler weight required to bend the bars (assuming 36 ksi yield): 707 pounds at 30 degrees 674 pounds at 45 degrees 502 pounds at 90 degrees For the 2” solid steel bars, the tumbler weight required to bend the bars (assuming 36 ksi yield): 1713 pounds at 30 degrees 1634 pounds at 45 degrees 1225 pounds at 90 degrees For the 2.5 NPS Schedule 40 pipe, the tumbler weight required to bend the bars (assuming 36 ksi yield): 2408 pounds at 30 degrees 2301 pounds at 45 degrees 1747 pounds at 90 degrees This is the force required to actually bend the bars. You’d need to cut all these weight values in half - to avoid cyclical fatigue concerns. I would cut the values by 2/3 – for extra safe & reliable design. Observations: pipe is way cheaper than solid bar, and is actually appreciably stronger – weight for weight - provided you take advantage of larger roller diameters. I just had 1.5" DOM tubing pop into my head to use as shafts. I build suspension and roll cages out of it, it's some pretty tough stuff and can be had in any OD and wall thickness you want, pretty much. Pillow block bearings are readily available to fit... I did a quick search for yield specs but didn't spend a lot of time and didn't find what I was looking for. The ideal tumbler drive is staggered rollers, with the lower roller pushing the barrel towards the upper. It is unnecessary to power both rollers, and it would be a significant error to drive just the upper roller. The reasoning behind this can be explained by with a simplex brake-drum analogy. However I got some pushback on this. (Please don’t be mad – as the below picture is straight off Wikipedia) The best way to explain it (while completely avoiding the brake drum comparison) is to show a better picture of what’s happening. This morning I did an engineering analysis of staggered tumbler bars. Below is the free-body diagram, and below that are derived equations. Important note: the force (Ft1) required to rotate the idler (roller 1) is purely a function of friction within the bearings. Install some good bearings, keep them greased, and the shaft will rotate freely (minimizing Ft1, and reducing overall power-input). EDIT: The RPM in the horsepower equation is the RPM of roller 2, and specifically not the rpm of the tumbler barrel. (edit complete). EDIT: I made an error on my initial solution. Below is the corrected copy. Discussion: you can see that Ft2 increases as Ft1 increases (because roller 2 is driving the tumbler into roller 1 - aka a self-engaging wedge). If you were to drive roller one, the friction-force would decrease, because you'd actually be turning the tumbler away from roller 2, and not into it. (edit complete) Interesting work there.
Don't mind me, you can talk about drum brakes all you want.
I'm just not going to discuss drum brake theory and operation anymore in a thread about building a tumbler, since I think the thread almost got de-railed when I did. I don't want 10 pages of you and I getting into a pissing match over brakes when we need to be discussing building a tumbler
|
|