Giant Wimshurst Machine - could use some guidance / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Giant Wimshurst Machine - could use some guidance

first previous 9 10 11 12 next last

Move Thread

LAN_403

AwesomeMatt

Thu Jun 05 2014, 09:41AM

Registered Member #4454 Joined: Sun Feb 26 2012, 12:47AM
Location: Western Canada
Posts: 74

wrote ...
Another thought on neutralizer brush material.

I think I have neutralizer brushes covered. Brass shim stock. Not as good as phosphor bronze, but, they'll do I hope.

2028

The little tabletop version uses solder wick for both neutralizers and combs. It's kinda finicky and crappy.

Solder wick is designed to be as soft as possible. It's not too difficult to work-harden copper, but, in braid form it'd probably just crumble.

I kinda gave up on finding carbon fiber combs and started making copper combs out of fine wire. If that fails, I'll just make contact with more brass shim stock and to hell with it.

Found out they already sent off the request for bids on the finished version, with a note saying they might cancel if it doesn't work out. So, I think that means people are more or less confident I can make this work, they just need to know whether it has the OOMPH when they actually see it work in the gallery. No mention of the deadline, other than not to slack off.

Visually, kind of an unproductive day. A lot of trial and error and figuring and not a lot of doing.

- Completed sector placement on second half of Disc #1.
- Measured and marked sector layout for Disc #2.
- Designed Leyden jars out of 4" PVC.
- Cut and test assembled Leyden jars. 42" tall, 18" gap at top required, leaves 24" of foil space.

2040

(Combs are pivoted out of the way, you'll see why in a second).
(Note the 4" PVC tube with square endcap, that's one of the to-be Leyden bazookas).

- Designed output post stabalizers.
- Drilled 2/4 combs with microbits, tested various soldering options. More stubborn than expected, but got them working.
- Designed comb framework, works but wheel frame is crooked, will integrate into post stabalizers. Need more copper pipe.
- Cut one comb and trimmed tines, ugly but should work. Backup plan is contact brushes or carbon fiber again.

2041

(Yep, it's missing a few tines on the left. They didn't survive the blowtorch).

I dunno if that's stupid or not, I like the idea of combs but, they're a pain in the ass to work with and now that they've been torched they're dead soft. Definitely not contact brushes.

- Cut free and threaded second pair of gazing globes.
- Designed new neutralizers, narrow them away from the electrodes.
- Designed how everything will work together, how to mount and assemble all the assemblies, what to upgrade/change if any of it causes problems.
- Wrote shopping list for tomorrow: More 1/2" copper pipe, 4" pulley, more 2" PVC, 3x 2" PVC "T" pieces, proper O-ring cord (almost tore a hose today), 2" PVC end cap to mount to the frame.

Oh, and Leyden jar questions:

1, When making the Leyden jars.. I need a sparkgap's length at the top (18" sparkgap = 18" of deadspace at the top of the jar), I think?

2. Is that both inside and outside? Or could I run the inside right to the lip, and the outside only part way up? Any point in doing that?

3. I'm just PVC gluing a bottom cap on the jar. Will that work for insulation? Or should I also not wrap foil for a few inches?

On the home stretch!

Ash Small

Thu Jun 05 2014, 10:48AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

The Leyden jars will only have capacitance where there is foil both inside and outside. Assuming you have a conductor going inside at the top, you'll need AT LEAST 18" gap on the outside, assuming the conductors aren't insulated.. Leave some space at the bottom. If the PVC glue isn't sufficient, you could always put some oil at the bottom inside for additional insulation if required.

Eleccentric

Thu Jun 05 2014, 07:37PM

Registered Member #33460 Joined: Tue Aug 27 2013, 06:23PM
Location: Seattle
Posts: 46

I would be wary of trusting the PVC glue to be as electrically solid as the pipe wall itself. You've got so much surface area to work with, though, that leaving foil off the bottom should not be an issue.

The problem with using brass shim stock for the neutralizers is that they may be harder than the aluminum. The hardness of brass and aluminum alloys are largely overlapping, so it's hard to say whether they will wear scratches into your sectors or not.

Carbon fiber tow is cheap and readily available. You ought not to have much trouble making brushes from it.

If you want the tines on your combs to be hard, temper them. Don't let the assembly cool slowly - quench it with cold water.

You could likely get away with securing them in place with epoxy - just make sure that the wire makes contact on the inside of the pipe. You could mix in some metal powder to help make sure there is good electrical contact, but again, at these voltages, it's not so critical.

AwesomeMatt

Thu Jun 05 2014, 08:19PM

Registered Member #4454 Joined: Sun Feb 26 2012, 12:47AM
Location: Western Canada
Posts: 74

wrote ...
The problem with using brass shim stock for the neutralizers is that they may be harder than the aluminum.

Oh, it's certainly harder than the dead soft aluminum tape.

I don't really care much. I want it working well enough to demo, and if they want more time put into it I'll swap it out for carbon fiber.

wrote ...
Carbon fiber tow is cheap and readily available.

I looked around for a place locally that had it, big city, couldn't find anything. Not sure what kind of place I'd be shopping from though.

wrote ...
You could likely get away with securing them in place with epoxy

JBweld, a metal epoxy even, is an insulator. Mixing metal shavings in doesn't help.

But, yeah, I may end up doing that.

Lots of mediocre solutions, all of which are finicky.

Ash Small

Thu Jun 05 2014, 08:41PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

The vast majority of ignition cables for cars, etc. have a carbon fibre core. You might even get some for free.

Another source is a rope called 'Dyneema'. There are several different sorts, but the really high tech stuff has a carbpn fibre core (I'd go for this over the ignition cable, but both should be ok), it's sold in 'yacht chandlers' and marinas and boatyards, if you have any near you.

It's pretty straightforward stripping the insulation off the ign. cable. I've not tried stripping the outer layer off the Dyneema, though.

Eleccentric

Thu Jun 05 2014, 10:08PM

Registered Member #33460 Joined: Tue Aug 27 2013, 06:23PM
Location: Seattle
Posts: 46

wrote ...

wrote ...
You could likely get away with securing them in place with epoxy

JBweld, a metal epoxy even, is an insulator. Mixing metal shavings in doesn't help.

But, yeah, I may end up doing that.

Lots of mediocre solutions, all of which are finicky.

Bear in mind that we are working with a voltage that will leap the gap between the sectors and the tines of your combs - the "insulating" properties of epoxy will do very little to impede the charge from transferring onto the copper pipes, especially if you let the wires extend into the pipe far enough that they touch the inner surface.

Ash Small's suggestion of using the core material from spark plug wires is excellent - I wish I had thought of that. Free or inexpensive, and should be much easier to find locally than raw carbon fiber tow. You could attach the fiber to the copper pipe with some of your aluminum tape.

AwesomeMatt

Fri Jun 06 2014, 06:44AM

Registered Member #4454 Joined: Sun Feb 26 2012, 12:47AM
Location: Western Canada
Posts: 74

Well... last Thursday I said I was going to try to pull an all-nighter to get it done by Friday. It was discouraged as unsafe.

They asked when I would get it done. I said I didn't know because nothing was being used for its intended purpose, everything part was an experiment and it might be the first idea that worked, or it might be the 5th idea. They wanted more info, so I talked them through it, working my way back:

- It would certainly be done in a month.
- It would almost certainly be done in 2 weeks.
- It would probably be done by next Friday.
- It would maybe get done by Monday/Tuesday.
- It would probably not get done by tomorrow night and if it did, all corners would be cut to do it.

And they were like "So, it'll be done Monday?" ... Uhhh... Let me just repeat this again. :p

That was a week ago. The machine that should be done tomorrow is a lot better than the one that would've been rushed for Monday,

But... I'm going to pull an all-nighter just to be sure. I get locked into the building for the night if I choose to stay, so, bunkering down time.

Progress as of 10pm:

- Swapped out 3" sheave for 4" pulley on left side, for flat pulleys. Outside bands are 1" wider than inside bands.
- Squared up the last bit of frame. Measured the horizontal and the vertical, did some trigonometry, cut a cross-brace, fits perfectly. And perfectly level! Do your homework kids, some day you may use it!
- Trimmed output supports, inserted T-pieces.
- Added beam between output supports, with T-piece to connect to main frame.
- Placed the pair of combs I have completed. Well look at that, they're dead-centered on the wheel, no adjusting necessary.
- Placed 50 sectors on Disc #2. Disc #2 is complete!

2042

(Extra wooden brace & PVC framework to keep everything properly spaced).

2043

(Note where horizontal PVC bolts to vertical wood... the end cap is bolted but rotatable. This way I can twist or untwist the end cap and vary the lineup of the combs. Planned on being crooked, turns out when everything is square, the measurements were spot on).

2044

Combs. Maybe 1/4" to 1/2" space between disc and comb tips. Hope that's close enough, it wobbles. If it doesn't work, brushes will quickly replace it.

I spin it and it looks like a Wimshurst machine. Hurray!

To Do:

- 2 more combs & arm
- Figure out how to mount neutralizer brushes
- Make and mount neutralizer brushes.
- Mount output electrodes.
- Add pair of globes to electrode #2.
- Turn PVC assembly into Leyden jars.

... wow, I think that's it. I was expecting a longer list.

The better way to mount the neutralizer brushes is inside the frame, next to the discs, where the PVC spacer is right now. But the discs weight 100+ pounds each, the axle is 7' (213cm) off the ground, and jammed on. I'm alone for the night, so, they're staying unless I have help holding the axle.

Instead, I will mount them on the extra axle that is sticking outside of the frame, secure them with masking tape or rubber bands, and not care because I'll be distracted by how awesome the sparks are.

First up: Globes and Leyden jars

Updates as I do stuff.

2am (last one was actually around 1am, took a lunchbreak first and wandered around trying to find something open).

- Globes are done, mounted to electrode #2.

2045

Did a bunch of research on Leyden jars, or tried to. No one really makes them this big. Or if they do, they don't give a lot of details. Should I be wrapping axially or lengthwise, how critical is smoothness, how insulatative is silicone or PVC glue, how thick does the foil have to be to not vaporize under the sudden current spike during the discharge, etc.

I read 40MV/meter for a dielectric constant. This PVC is only 4mm thick. That means it's only good for up to 160,000V. I'm maybe below that, maybe not.

Is the voltage potential of the machine that I have to worry about, split between the two Leyden jars? That is... if I have 300,000V being generated in the machine, am I okay? (160kv + 160kv = 320,000v > 300,000v).

I know I could build 4 leyden jars and put them in series, but, I don't really have anywhere else to put them except in the way of the visitor and the electrode pole handles.

Moot anyways, I covered both insides but only 1 outside before I ran out of aluminum tape. Added it to my shopping list for tomorrow... err... when stores open in a few hours.

Shattered an end cap anyways when one fell over. so, was off to the hardware store again anyway before they were done.

Next up... neutralizer brushes and bars I guess. Something way overdue for verifying the other parts of the machine.

7am.

Neutralizer bars... ehn. Sort of done. Wobbley-loose (will shim with tape for now) and missing the brushes. Problem is I know they're not staying this way, so I'm not sure I should finalize them any more. Right now they're just press-fit copper, not soldered.

Tried to rip protective backing off the discs... I of course can't reach that deep between the wheels. Which means I have to take them off. Which means I need 1 or 2 more people to help. Which means, I might as well mount the neutralizer bars where they belong and stop half-assing it.

But I really really really want to see the thing power up before I go calling for help.

Dawn has arrived.

2046

Compromise. Soldered some of the neutralizer's first joints that won't change, gave them a longer "bushing" (chunk of pipe).

Really don't feel like making a second set of combs. So I won't. Rather than drill the pipe, insert wires, solder it up... I'll just wrap aluminum tape around the pipe and sandwich a dozen wires in the process.

Brush material is getting assembled the same way if I can get away with it. I'll soon find out.

Ugh. Was feeling confident, then a pile of related problems.

- Brushes are too hard, scraping the sectors and the disc.

- Brushes need to be springy to keep the ghetto neutralizer bars balanced. The neutralizer bar assembly is awful and wobbly and there's no way to anchor it easily. They have nothing for a collar, and aren't pinned to the axle (right now rubberbanded to the frame so they don't spin like a clock).

2047

- Bars are actually enough leverage to prevent the wheel from turning and start the belts slipping.

- Pulleys need to be revamped. Maybe can't be, because they're using a 1" pulley up top and might not have enough friction even if I drop the lower one to, say, 6:1 instead of 12:1. Ugh.

Kind of natural, to see all the problems show up on the very last stage of test assembly. Just discouraging and... I had the physical fortitude to persevere until it was done because I thought there were no more design obstacles, only just work to do.

Morning of an all nighter is not the time to tinker or troubleshoot or come up with ideas.

Argg.

Odds of having something demo-able today went from like 85% to like 5%.

Dr. Slack

Fri Jun 06 2014, 04:25PM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

A few tips, the neutraliser bars are at ground potential, which may offer you more options for mounting them.

The neutraliser brushes don't need to carry amps, if you replace that meaty strip with a wisp of copper wire, they will still work electrically, and won't offer the disc as much friction. Copper takes a set, so maybe use a sliver of plastic to support it. A piece of thick copper wire could be hinged from the neutraliser bar and sprung lightly against the disc with a small spring or elastic band, there are endless low-friction alternatives.

The main problem with your gear ratio is not the big pulley at the bottom, but the small one at the top. What options do you have to increase this? If the discs are coming off at some time, then a replacement could be scheduled in.

If not, can you buy a couple of 4" or 6" pulleys, cut them in half, remove an appropriate sized semicircle from each half, and assemble them round the existing tiny pulley, maybe with glue, maybe with fish plates or even stitching between the two halves? With stretchy belts, these pulleys do not have to be exactly concentric. It will look ugly, but may fix your immediate demo, and schedule in the proper replacement before any public see it.

AwesomeMatt

Fri Jun 06 2014, 05:04PM

Registered Member #4454 Joined: Sun Feb 26 2012, 12:47AM
Location: Western Canada
Posts: 74

wrote ...
The main problem with your gear ratio is not the big pulley at the bottom, but the small one at the top. What options do you have to increase this?

Ugh, I know. It was the very first thing I bought (based on Mr. Teslonian's claim of 14.3:1 being feasible) and everything else is based around it. Pulleys that size are only 1/2", so the axle (that is now bowing under 300 pounds of weight) is 1/2". The 48" discs have all of 1.5" of boss and torque arm. The belt has only the 1" pulley circumference to grip onto, etc etc.

Almost every issue can be traced back to that first step.

And, I've almost no opportunity to change it. I suggested it an hour ago as a "What I really should do is..." and was given cringefaces and "How about we don't do that yet..."s. I kinda agree. It would be a lot of jig work to get it re-cut to a different size, and the guy who runs the Shopbot isn't in until next week.

Perfect storm.

I may try your pulley dicing trick, but what I'm afraid of is overtorquing the discs.. it won't change the fact that they're not bolted to the discs except at that 1.5" region which may shatter under more torque.

Copper/plastic combo is another idea I'll add to the list of things to try.

Threw in the towel around noon.

Found more aluminum tape though, so, finished the taping of the Leyden jars inside and out:

2048

Went shopping:

- New endcaps for the Leyden jar to replace the ones that shattered
- Found a nice gentleman that repairs copiers who gave me some carbon fiber brushes (they're only 1/4" long sadly, my wobble is worse than that so... probably won't work :( ),
- Some proper Oring cord ($4.37/foot... $100 later...) and I'm not even sure it will work because it's pretty stretchy itself.
- Two more 6" pulleys (previous 2 I only ever used as a collar for the black 12" pulleys), so now I have 4 total and can use them as my lower drive pulleys.

...

I don't think any first idea has worked. I was making a mental list of the parts and assemblies today, and everything is already on v2 or v3. Naturally, the parts that are still original are the places things are going wrong.

I guess I spend another weekend at it and try for Monday.

Found out that they'll probably go with a flexible quote from the fab shop, so, it's not just "It works? Okay, throw it away, we'll build our own now." Any details I figure out in terms of size or material or construction, all save them money on the bill. So, that's less demoralizing at least.

Wastrel

Sat Jun 07 2014, 05:24PM

Registered Member #4095 Joined: Thu Sept 15 2011, 03:19PM
Location: England.
Posts: 122

Do you have an estimation of how fast the discs can rotate before they are an explosion risk? Do you have an estimation of how fast a thirteen year old boy with attention deficit disorder can turn a hand crank?

How close to this sort of spark can a digital watch get and survive? Will the device need a faraday cage?

Awesome and cool btw.

first previous 9 10 11 12 next last

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob