Capapalooza
Adam Munich, Fri Sept 17 2010, 02:05AMI'm going to use this thread to document my adventures with homemade capacitors, and the strive to possibly build a 1MegaVolt marx with them.
I went to office max, and they wanted $40 (forty!) for a pack of 100 OHP sheets. So that wasn't going to happen. I ended up getting a pile of them from a teacher, and they are assorted. There are HP ones, Avery ones, and staples ones.
The HP and Avery are cellulose acetate, while the staples ones are made out of the same stuff as sandwich bags. Unfortunately the HP ones are covered in the same kind of glue that is used on envelopes, so that will be a pain to get off. I'm speeding up the process by soaking them in my hot tub with the jets on.
I'm going to make 15kV caps for many reasons.
1. I don't want to deal with side flashover.
2. I don't want corona on my spark gaps
3. I want this to be LOUD. Lower voltage = Higher capacitance
4. I don't have a lot of sheets
5. My highest voltage DC source ATM is (2x) full wave rectified 15kV 25mA NST.
I plan on using 6 sheets per cap. I'm going to test both the acetate ones and the lunchbag ones, and see what's better.
Here's a picture of the sheets. This thread will be pic heavy.
Re: Capapalooza
Avi, Fri Sept 17 2010, 03:02PM
is this going to be in or out of oil?
those hole punched things so seem to work quite good (in air) because they are sealed at the sides but have their limitations eventually
Avi, Fri Sept 17 2010, 03:02PM
is this going to be in or out of oil?
those hole punched things so seem to work quite good (in air) because they are sealed at the sides but have their limitations eventually
Re: Capapalooza
803, Fri Sept 17 2010, 07:56PM
Good luck!
may I sugest you cut them into 4ths and use them, that would save space.
803, Fri Sept 17 2010, 07:56PM
Good luck!
may I sugest you cut them into 4ths and use them, that would save space.
Re: Capapalooza
Patrick, Fri Sept 17 2010, 08:07PM
some advice Grenadier, build a kind of cubic rectangle shaped cap that will fit inside the largest pipe diameter, you can get. this will give you the ability to vacuum pull oil, after all an 8.5 " dia pipe 11" long might be hard/expensive to find. but a steel water pipe 4 " x 10" is easier/cheaper, and your capacitence is assumed to be the same in each case, as an exapmle.
Patrick, Fri Sept 17 2010, 08:07PM
some advice Grenadier, build a kind of cubic rectangle shaped cap that will fit inside the largest pipe diameter, you can get. this will give you the ability to vacuum pull oil, after all an 8.5 " dia pipe 11" long might be hard/expensive to find. but a steel water pipe 4 " x 10" is easier/cheaper, and your capacitence is assumed to be the same in each case, as an exapmle.
Re: Capapalooza
Adam Munich, Fri Sept 17 2010, 10:41PM
3 hours and one box of M. Clean magic erasers and all the HP ones are all clean.
I measured the thickness of both the HP and (Avery was the other brand name) sheets and they are both 3 mil. I bought a box of Reynold's wrap and I'm going to put some together now and I'm going to use 8 sheets per cap. I'm not doing any vacuum or oil, as that will be a big pain in the *ss to do for every cap; these will be used in a marx generator after all. Flash over won't be a problem since these will be run at 15kV.
Adam Munich, Fri Sept 17 2010, 10:41PM
3 hours and one box of M. Clean magic erasers and all the HP ones are all clean.
I measured the thickness of both the HP and (Avery was the other brand name) sheets and they are both 3 mil. I bought a box of Reynold's wrap and I'm going to put some together now and I'm going to use 8 sheets per cap. I'm not doing any vacuum or oil, as that will be a big pain in the *ss to do for every cap; these will be used in a marx generator after all. Flash over won't be a problem since these will be run at 15kV.
Re: Capapalooza
Muttyfutty!, Sat Sept 18 2010, 01:40AM
Ahhhhh.... Oh dear, the same caps I made last week failed
on my single un-rectified 15 KV NST. 1 neat, burning hole through
7 layers of OHP Transparency. Ouch!
Be VERY careful with 6 layers, Dont Operate them for too long. Good luck!
regards,
mike
Muttyfutty!, Sat Sept 18 2010, 01:40AM
Ahhhhh.... Oh dear, the same caps I made last week failed
on my single un-rectified 15 KV NST. 1 neat, burning hole through
7 layers of OHP Transparency. Ouch!
Be VERY careful with 6 layers, Dont Operate them for too long. Good luck!
regards,
mike
Re: Capapalooza
Adam Munich, Sat Sept 18 2010, 02:37AM
Well mine (2 sets of 4 layers) seem to be holding up just fine!
I made 5, then ran out of OHP sheet. Need to find a place to get more. I think I may have bitten off more than I can chew with a 1 MV marx, because I'll need 70 capacitors. That's 560 OHP sheets! I intend to try my best though, but I'll still need to find a lot more sheets. No more HP ones though, I'm done scrubbing glue.
So I measured them, and I must say they are all pretty close. 3 were exactly the same! However I killed my multimeter on the last one. Apparently it built up enough charge by simply putting it together to kill my DMM.
Now I need to go meter shopping tomorrow and burn a hole in my wallet.
Values are as follows.
3x 2.80nF
1x 2.86nF
1x 2.97nF
These are very stable caps, and pressing them or changing the temperature doesn't seem to affect the value at all. Definitely suitable for TC use.
I hooked them all in // to my 15/25 NST (half wave rectified by 32 1n4005's) and they make one hell of a crack. It left my eyes burning and my ears ringing after 5 seconds of that racket. There is no problem with flashover and they don't seem to mind not having a spark gap either. They don't hold a 15kV charge for very long due to corona loss, but hold 5kV quite well. So If I build a 70 stage(15kV) marx I'll need relatively low value resistors (~250K). They also make an interesting sound when the power is cut, kind of like TSSSSSHHHHHHEEEWW
I'll record it sometime.
I also hooked them up to my supposedly GFCI NST's. Guess what didn't happen? :p
Sadly no video, my video camera's dead and the charger is nowhere to be found. Got pics though.
I had to put the NST on a box in the last one because electricity found it's way through the floor adn gave me a tingle.
Adam Munich, Sat Sept 18 2010, 02:37AM
Well mine (2 sets of 4 layers) seem to be holding up just fine!
I made 5, then ran out of OHP sheet. Need to find a place to get more. I think I may have bitten off more than I can chew with a 1 MV marx, because I'll need 70 capacitors. That's 560 OHP sheets! I intend to try my best though, but I'll still need to find a lot more sheets. No more HP ones though, I'm done scrubbing glue.
So I measured them, and I must say they are all pretty close. 3 were exactly the same! However I killed my multimeter on the last one. Apparently it built up enough charge by simply putting it together to kill my DMM.
Now I need to go meter shopping tomorrow and burn a hole in my wallet.Values are as follows.
3x 2.80nF
1x 2.86nF
1x 2.97nF
These are very stable caps, and pressing them or changing the temperature doesn't seem to affect the value at all. Definitely suitable for TC use.
I hooked them all in // to my 15/25 NST (half wave rectified by 32 1n4005's) and they make one hell of a crack. It left my eyes burning and my ears ringing after 5 seconds of that racket. There is no problem with flashover and they don't seem to mind not having a spark gap either. They don't hold a 15kV charge for very long due to corona loss, but hold 5kV quite well. So If I build a 70 stage(15kV) marx I'll need relatively low value resistors (~250K). They also make an interesting sound when the power is cut, kind of like TSSSSSHHHHHHEEEWW
I'll record it sometime. I also hooked them up to my supposedly GFCI NST's. Guess what didn't happen? :p
Sadly no video, my video camera's dead and the charger is nowhere to be found. Got pics though.
I had to put the NST on a box in the last one because electricity found it's way through the floor adn gave me a tingle.
Re: Capapalooza
Adam Munich, Wed Sept 22 2010, 11:41PM
More plastic! Here's what I got.
+ About 25 more acetate sheets, (HP). More scrubbing tonight.
+ 2x 75 foot rolls of acetate. One is covered in drawings and words, but vis-a-vis markers are easy to clean. The other one is old (yellowed), but nothing else seems to be wrong with it.
+ 1x 75 foot roll of sandwich bag stuff. About 1/3 used, 2/3 unused and clean.
So what should I do with the rolls? A giant 5kV cap, or cut them up to use in multiple small 15kV caps?
Adam Munich, Wed Sept 22 2010, 11:41PM
More plastic! Here's what I got.
+ About 25 more acetate sheets, (HP
). More scrubbing tonight.+ 2x 75 foot rolls of acetate. One is covered in drawings and words, but vis-a-vis markers are easy to clean. The other one is old (yellowed), but nothing else seems to be wrong with it.
+ 1x 75 foot roll of sandwich bag stuff. About 1/3 used, 2/3 unused and clean.
So what should I do with the rolls? A giant 5kV cap, or cut them up to use in multiple small 15kV caps?
Re: Capapalooza
Adam Munich, Fri Sept 24 2010, 12:36AM
2 hours of scrubbing that dang glue and all I got was 2 made. I ran out of sheets before I could finish another. It's anyone guess as to what their capacitance is, but it's likely to be around the 2.85nf mark.
I tried making a cap with the rolls of plastic and it was fail to say the least. Off the roll the plastic was curved, so it didn't want to stay flat. The plastic sure wasn't acetate, it looked more like a PVC type material. It was also 1mil thinner. I attempted to make the thing with the rolled stuff by adding 2 more sheets to compensate for the thinner-ness. I couldn't get anything to stay flat and when it was rolled uo a whole bunch of air got inside. I thew the thing out before even hooking it up to an X-former. I still haven't tried the sandwich bag stuff.
I was searching online and I found 100pks of sheet for $15 shipped. I don't think it's worth it to buy the stuff. $60 will get me about 40 caps, and even though they will be cheaper than ready made 2.85nF 15kV caps, I'm not sure it's worth it. I'll still attempt to get some more free sheets though.
edit: Dang I just realized I triple posted. Not a real big deal though I suppose. It's not like they're one line posts.
Adam Munich, Fri Sept 24 2010, 12:36AM
2 hours of scrubbing that dang glue and all I got was 2 made. I ran out of sheets before I could finish another. It's anyone guess as to what their capacitance is, but it's likely to be around the 2.85nf mark.
I tried making a cap with the rolls of plastic and it was fail to say the least. Off the roll the plastic was curved, so it didn't want to stay flat. The plastic sure wasn't acetate, it looked more like a PVC type material. It was also 1mil thinner. I attempted to make the thing with the rolled stuff by adding 2 more sheets to compensate for the thinner-ness. I couldn't get anything to stay flat and when it was rolled uo a whole bunch of air got inside. I thew the thing out before even hooking it up to an X-former. I still haven't tried the sandwich bag stuff.
I was searching online and I found 100pks of sheet for $15 shipped. I don't think it's worth it to buy the stuff. $60 will get me about 40 caps, and even though they will be cheaper than ready made 2.85nF 15kV caps, I'm not sure it's worth it. I'll still attempt to get some more free sheets though.
edit: Dang I just realized I triple posted. Not a real big deal though I suppose. It's not like they're one line posts.
Re: Capapalooza
Muttyfutty!, Fri Sept 24 2010, 01:05AM
WOW, That is certainly impressive!
The largest Value caps I got was about 2 nF. What thickness are your transparencies?
mine are .1mm. (I think)
Have you tried them unrectified?
Nice Diode chain by the Way!
For some strange reasons, My caps are failing on both the Voltage side and the Capacity.
Mains kinda fluctuates over here, We can get up to 270V, Is it a possibility that I’m
getting considerably more than 21KV peak, 15KV RMS? That would explain why My 7 layered caps keep failing....
Muttyfutty!, Fri Sept 24 2010, 01:05AM
WOW, That is certainly impressive!
The largest Value caps I got was about 2 nF. What thickness are your transparencies?
mine are .1mm. (I think)
Have you tried them unrectified?
Nice Diode chain by the Way!
For some strange reasons, My caps are failing on both the Voltage side and the Capacity.
Mains kinda fluctuates over here, We can get up to 270V, Is it a possibility that I’m
getting considerably more than 21KV peak, 15KV RMS? That would explain why My 7 layered caps keep failing....
Re: Capapalooza
Adam Munich, Fri Sept 24 2010, 01:32AM
Mine are 3.5 mil thick, and I used 4 in between each plate. (8 in total). I haven't tried them unrectified yet. A key tip is to pull your zip ties VERY tight. You want as little movement and air as possible.
As for your voltage issue, that fluctuation could very well be the problem. Have you ever thought about getting a gasoline generator? You could also use one of those 12VDC to 120V inverters. They're really cheap now and the NST's don't seem to mid the nonsinusoidal output. It doesn't get more regulated than that.
I'm undecided as to if I may splurge and buy some sheets. But the sparks are certainly impressive. And with a million volts at 2.8nf, that's 1.4kJ! Definitely not too shabby for $60
Edit: I was just playing with these a little more. They hold a charge for quite a long time with minimal hiss (corona). After a 3 second charge they seem to contain the peak voltage, not RMS judging by spark length. I dare not go longer because at full charge the HV from my 15kV NST find it's way into places it shouldn't.
They definitely seem suitably for marx use provided I use 250k resistors. :D
Edit2: I dared to go longer. The HV seems to like to spark to neutral. (small <20uA sparks) This doesn't happen without the caps. It doesn't seem to affect anything in my house lines though, so it's not too big of a problem.
Edit3: Found the problem. I had the NST on a .5inch thick piece of acrylic for insulation, but one HV wire was touching the wood floor. Seems that the charged wood acted a plate, and the NST case was capacitively coupled and a charge formed on it. This charge then sparked to neutral. (If that doesn't trigger the supposed GFCI that I have in my 15kV x-formers as well, absolutely nothing will.)
Edit4: I'm going to buy them. Apparently they're 4 mils thick, so they aren't crap. A 1kJ million volt marx made from "garbage" would be epic on many proportions as well. I have 1.4k in the bank from my summer job anyway, so if the marx is fail, I could use the caps in a tesla coil. They seem reliable enough, and if I use them in a tesla coil putting them under oil won't be a problem.
Adam Munich, Fri Sept 24 2010, 01:32AM
Mine are 3.5 mil thick, and I used 4 in between each plate. (8 in total). I haven't tried them unrectified yet. A key tip is to pull your zip ties VERY tight. You want as little movement and air as possible.
As for your voltage issue, that fluctuation could very well be the problem. Have you ever thought about getting a gasoline generator? You could also use one of those 12VDC to 120V inverters. They're really cheap now and the NST's don't seem to mid the nonsinusoidal output. It doesn't get more regulated than that.
I'm undecided as to if I may splurge and buy some sheets. But the sparks are certainly impressive. And with a million volts at 2.8nf, that's 1.4kJ! Definitely not too shabby for $60
Edit: I was just playing with these a little more. They hold a charge for quite a long time with minimal hiss (corona). After a 3 second charge they seem to contain the peak voltage, not RMS judging by spark length. I dare not go longer because at full charge the HV from my 15kV NST find it's way into places it shouldn't.
They definitely seem suitably for marx use provided I use 250k resistors. :D
Edit2: I dared to go longer. The HV seems to like to spark to neutral. (small <20uA sparks) This doesn't happen without the caps. It doesn't seem to affect anything in my house lines though, so it's not too big of a problem.
Edit3: Found the problem. I had the NST on a .5inch thick piece of acrylic for insulation, but one HV wire was touching the wood floor. Seems that the charged wood acted a plate, and the NST case was capacitively coupled and a charge formed on it. This charge then sparked to neutral. (If that doesn't trigger the supposed GFCI that I have in my 15kV x-formers as well, absolutely nothing will.)
Edit4: I'm going to buy them. Apparently they're 4 mils thick, so they aren't crap. A 1kJ million volt marx made from "garbage" would be epic on many proportions as well. I have 1.4k in the bank from my summer job anyway, so if the marx is fail, I could use the caps in a tesla coil. They seem reliable enough, and if I use them in a tesla coil putting them under oil won't be a problem.
Re: Capapalooza
mnr, Fri Sept 24 2010, 03:45PM
Keep in mind the total capacitance of the Marx when erected will be C/n (i.e. a series connection of n capacitors).
mnr, Fri Sept 24 2010, 03:45PM
Grenadier wrote ...
I'm undecided as to if I may splurge and buy some sheets. But the sparks are certainly impressive. And with a million volts at 2.8nf, that's 1.4kJ! Definitely not too shabby for $60
I'm undecided as to if I may splurge and buy some sheets. But the sparks are certainly impressive. And with a million volts at 2.8nf, that's 1.4kJ! Definitely not too shabby for $60
Keep in mind the total capacitance of the Marx when erected will be C/n (i.e. a series connection of n capacitors).
Re: Capapalooza
Adam Munich, Fri Sept 24 2010, 07:33PM
Are you sure about that?
Adam Munich, Fri Sept 24 2010, 07:33PM
Are you sure about that?
Re: Capapalooza
mnr, Fri Sept 24 2010, 07:48PM
Yes. From that site: "A common specification is the erected capacitance of the bank, equal to the stage capacitance divided by the number of stages."
Should still be an interesting project though, good luck!
mnr, Fri Sept 24 2010, 07:48PM
Grenadier wrote ...
Are you sure about that?
Are you sure about that?
Yes. From that site: "A common specification is the erected capacitance of the bank, equal to the stage capacitance divided by the number of stages."
Should still be an interesting project though, good luck!
Re: Capapalooza
ScotchTapeLord, Fri Sept 24 2010, 07:57PM
Take two capacitors, to make numbers easy, 2 farads at 1 volt would be:
.5*(2)(1)^2 = 1 Joule
for each.
Discharged in series would mean sum of the voltages and the total energy of all capacitors.
1 Joule x 2 = 2 Joules
1 Volt x 2 = 2 volts
Solving for capacitance:
.5*(C)(2)^2 = 2 Joules
C = 2/(.5*4) = 1 Farad
So yes, the total C is C1/#.
Another way to visualize it is to realize in its firing stage, a marx gen is just a bunch of capacitors and resistors in series. A string of # series capacitors of equal C charged to Vtotal. If you take one capacitor away and measure it, it is charged to Vtotal/# with its capacitance being #*Ctotal.
ScotchTapeLord, Fri Sept 24 2010, 07:57PM
Take two capacitors, to make numbers easy, 2 farads at 1 volt would be:
.5*(2)(1)^2 = 1 Joule
for each.
Discharged in series would mean sum of the voltages and the total energy of all capacitors.
1 Joule x 2 = 2 Joules
1 Volt x 2 = 2 volts
Solving for capacitance:
.5*(C)(2)^2 = 2 Joules
C = 2/(.5*4) = 1 Farad
So yes, the total C is C1/#.
Another way to visualize it is to realize in its firing stage, a marx gen is just a bunch of capacitors and resistors in series. A string of # series capacitors of equal C charged to Vtotal. If you take one capacitor away and measure it, it is charged to Vtotal/# with its capacitance being #*Ctotal.
Re: Capapalooza
Adam Munich, Fri Sept 24 2010, 09:14PM
Wow how'd I miss that, it's only 16 joules. I had a feeling 1.4kJ would be too good to be true. Man do I feel like an idiot.
.
Irregardless, I'll have a whole bunch more caps done tonight. I'm going to hae to start using earmuffs because these things are LOUD! No video though... I'm going to need to buy a new camera battery charger.
Edit: A whole bunch turned out to be eight. So far I have 15 total, or an estimated 37.5nF of 15kV capacitors. I'll need 55 more (440sheets) to reach 1MV. Pics at 11.
Adam Munich, Fri Sept 24 2010, 09:14PM
mnr wrote ...
Yes. From that site: "A common specification is the erected capacitance of the bank, equal to the stage capacitance divided by the number of stages."
Yes. From that site: "A common specification is the erected capacitance of the bank, equal to the stage capacitance divided by the number of stages."
Wow how'd I miss that, it's only 16 joules. I had a feeling 1.4kJ would be too good to be true. Man do I feel like an idiot.
.Irregardless, I'll have a whole bunch more caps done tonight. I'm going to hae to start using earmuffs because these things are LOUD! No video though... I'm going to need to buy a new camera battery charger.
Edit: A whole bunch turned out to be eight. So far I have 15 total, or an estimated 37.5nF of 15kV capacitors. I'll need 55 more (440sheets) to reach 1MV. Pics at 11.
Re: Capapalooza
Muttyfutty!, Fri Sept 24 2010, 11:52PM
Whoops, never mind, everyone makes mistakes
Sorry I didn't make myself clearer, I live in 240V land.
But technically...
15,000V/240v = 62.5v*270 = 16,875 Volts
So, because of mains fluctuation, I'm getting close to 17KV RMS.
That still doesn't really explain why Im getting arc overs, significant corona,
and less capacitance...
Im Very intrigued how you managed to get a higher capacitance than
me with thicker sheets...
I'm getting close to breaking down and buying 'Proper' Capacitors
regards,
mike
Muttyfutty!, Fri Sept 24 2010, 11:52PM
Whoops, never mind, everyone makes mistakes
Sorry I didn't make myself clearer, I live in 240V land.
But technically...
15,000V/240v = 62.5v*270 = 16,875 Volts
So, because of mains fluctuation, I'm getting close to 17KV RMS.
That still doesn't really explain why Im getting arc overs, significant corona,
and less capacitance...
Im Very intrigued how you managed to get a higher capacitance than
me with thicker sheets...
I'm getting close to breaking down and buying 'Proper' Capacitors
regards,
mike
Re: Capapalooza
Adam Munich, Sat Sept 25 2010, 03:43AM
Crap quality sheets? The HP ones, Avery ones, and now the 3M ones seem to hold up just fine.
And sorry, no pics yet. I fell asleep and I'm too tired now. I'll get them in the morning.
Adam Munich, Sat Sept 25 2010, 03:43AM
Crap quality sheets? The HP ones, Avery ones, and now the 3M ones seem to hold up just fine.
And sorry, no pics yet. I fell asleep and I'm too tired now. I'll get them in the morning.
Print this page