Sciencemadness Discussion Board
Not logged in [Login ]
Go To Bottom

Printable Version  
 Pages:  1  2    4    6  ..  11
Author: Subject: Has anyone made a EBW setup.
Hennig Brand
International Hazard
*****




Posts: 1284
Registered: 7-6-2009
Member Is Offline

Mood: No Mood

[*] posted on 13-5-2015 at 10:43


Don't worry about it, it is an entertaining video anyway.



"A risk-free world is a very dull world, one from which we are apt to learn little of consequence." -Geerat Vermeij
View user's profile View All Posts By User
ecos
National Hazard
****




Posts: 464
Registered: 6-3-2014
Member Is Offline

Mood: Learning !

[*] posted on 15-5-2015 at 00:38


Quote: Originally posted by Spartan  
Has anyone try to connect a mobile phone to the EBW detonator something like this video:
https://www.youtube.com/watch?v=LrSdQjlUND8

Do you think it's possible? If it is safe i think it's a good idea, you avoid to use to long wires.


I have a friend who lost his hand because of mobile phone !
he made similar setup but he received an advertisement sms while the wires were connected to detonator !

another one made similar setup and when he turned on the phone it played the startup tone and he was severely injured.

avoid such mobile phones setup ! really risky.
View user's profile View All Posts By User
jock88
International Hazard
*****




Posts: 505
Registered: 13-12-2012
Member Is Offline

Mood: No Mood

[*] posted on 15-5-2015 at 01:38



Also the mother in law may send you a text at the appropriate time if she knows what you are up to.

You would need to put a tone decoder (dual tone would be better) on the output of the phone set it up so that pressing the (say) 4 will give an output at the other end.
View user's profile View All Posts By User
Hennig Brand
International Hazard
*****




Posts: 1284
Registered: 7-6-2009
Member Is Offline

Mood: No Mood

[*] posted on 15-5-2015 at 03:28


Tone encoding works well, I made a radio controlled switch from a pair of GMRS radios a few years back. The receiver had a tone decoder and also a timer which when activated disconnected the output to the igniter for a minute or so allowing time to safely make connections and retire to a safe distance before the system became active (I suppose the timer could have also disconnected the input to the encoder from the radio receiver). The tone generator at the transmitter was an mp3 player. The tones were made with a computer tone generator program and then saved as an mp3 file and put on a free mp3 player (someone got it for subscribing to something IIRC, very cheap mp3 player). The system could be used with any audio transmitter receiver pair. I built it just for fun, and other than a few tests I have always used wires or fuse when doing any serious blasting (wires are best).

Regarding controlling an EBW fireset, the system could be charged and fired from a single control. One control could charge the capacitor bank and fire it. A comparator could be used to automatically fire the EBW when the voltage of the capacitor bank rose to a set value. A suitably low proportional voltage could be made available for the comparator using a voltage divider. When the voltage rose above the set point and the comparator changed state it could send the trigger pulse to the trigatron and fire the EBW.


[Edited on 15-5-2015 by Hennig Brand]




"A risk-free world is a very dull world, one from which we are apt to learn little of consequence." -Geerat Vermeij
View user's profile View All Posts By User
Varmint
National Hazard
****




Posts: 264
Registered: 30-5-2013
Location: Near Atlanta, GA
Member Is Offline

Mood: No Mood

[*] posted on 15-5-2015 at 06:55


The wise engineer would make it a pass-key, where a specific sequence is sent, even wiser would be to time encode the key string.

i.e. lets say the pass string is DTMF 27639. The decoded DTMF is sent to a microcontroller and the first 2 chars are expected in 1 second, the 3rd char 3 to 5 seconds later, the last two characters keyed within one second, sent greater than 3, but less than 5 seconds after the 3rd character. Bad sequences cause a lockout of a programmed duration, or in fact failsafe the system requiring a physical reset.

Seem complex? Well, you want the largest margin of safety you can engineer, or at least I would if I were working in that area. Bad se
View user's profile View All Posts By User
Hennig Brand
International Hazard
*****




Posts: 1284
Registered: 7-6-2009
Member Is Offline

Mood: No Mood

[*] posted on 15-5-2015 at 07:06
EBW Capacitor Bank Charging Circuit (555 Inverter Based)


Thanks for the radio control encoding tips. I was aware that a lot more sophistication and safety could be added, but to tell the truth I never made it past using two tones at once. Always meant to go back and make it more sophisticated though. Electricity/electronics is just one of my hobbies, I went through Chemical Engineering.

First off I will say that Markx was right when he said I needed a beefier power supply than those bug zappers and his suggestion for a circuit was a good one. The post by Markx earlier in this thread can be found here:

http://www.sciencemadness.org/talk/viewthread.php?tid=23466#...

I have ordered some of the IR2153 and 21531, but they will take a couple of weeks to get to me (paid a lot more for the 21531, oh well live and learn). In the mean time I have been working on a 555 inverter circuit. The IR2153 is superior as a mosfet driver, but the 555 can be made to work and it is a much more common IC.

I used the circuit found here:
http://www.eleccircuit.com/220-volts-power-inverter-using-ne...

However, I was not interested in driving a big, heavy, slow 50 or 60Hz transformer or using a voltage multiplier with big capacitors as is required with such low frequencies. In order to bring the frequency of the astable circuit from 50Hz up to about 15kHz C1 was changed to 1nF, R1 was changed to 1kohm and R2 was changed to 47kohm. This also kept the duty cycle just slightly over 50%. The only part I used that was exactly the same as specified was the 555 timer IC. I was hesitant to go higher in frequency because I was only using a regular 555, not the faster cmos variety and I was only using slow 1N4007 diodes not one of the faster types such as UF4007.

The transformer was wound on a plastic bobbin which came with the salvaged pot core used. The ferrite pot core is 3019P, 3C8 material (presently called 3C81 from what I have read). The primary is 18 turns 26AWG bifular wound in two layers to give a center tap. The secondary was wound in 10 layers, with each layer having about 40 turns of double insulated 40AWG magnet wire. A layer/strip of printer paper just slightly wider than the bobbin winding surface was wrapped on between each layer which added insulation between layers and also gave a nice flat, smooth, surface to wind on. There are better commercial insulating papers available for winding. Apparently I had the primary hooked up out of phase yesterday because it took about 15seconds to even charge the capacitor bank up to 2000V and I was having some trouble with the upper transistor heating, but the heat sinks I was using yesterday were also tiny and the run times were longer so I am not sure that it is related. Today I changed the transformer primary winding wiring around and now the circuit charges up the 0.5uF capacitor bank to 4000V in about 5 seconds. I should also mention that there is 40Mohms of resistance across the capacitor bank as a bleeder and bypass network and that there is 1Mohm (10X 100kohm resistors) between the capacitor bank and the power supply. So the supply is putting out much more than what would be needed to charge the capacitor bank without these extra loads.

The heat sinks are just what I could find lying around and are not even matched. More than big enough for the 5 second run time though. The pot core was found on a piece of industrial electronics in a scrap yard. It looks to have been sitting out in the snow and rain for several years. The old windings were removed before the new windings were put on.

The battery pack I am using is made up of a mix of old AA batteries and so is not ideal. The circuit draws about 0.7-0.8A, from the little battery pack, which pulls it down to about 7-7.5Volts. The output from the transformer is about 500-550V which is fed into two six stage half wave voltage multipliers made up of 3kV 10nF ceramic caps and 1N4007 (1A, 1000V) diodes. The diodes for the two multipliers are run in opposite directions from one another. The voltage across the two multiplier outputs is greater than 6000V with no load.



Inverter, Voltage Multiplier & Capacitors.jpg - 259kB Inverter.jpg - 286kB Voltage Multiplier.jpg - 285kB
220-volts-power-inverter-using-NE555-and-MOSFET.jpg - 61kB Stacked Villard Cascade.jpg - 25kB



Found some mounting hardware. It looks a little less amateurish without the pink elastic band. :D

555 Inverter Circuit (500V Output).jpg - 250kB


[Edited on 15-5-2015 by Hennig Brand]




"A risk-free world is a very dull world, one from which we are apt to learn little of consequence." -Geerat Vermeij
View user's profile View All Posts By User
Varmint
National Hazard
****




Posts: 264
Registered: 30-5-2013
Location: Near Atlanta, GA
Member Is Offline

Mood: No Mood

[*] posted on 15-5-2015 at 12:10


Henning:

The rectifiers aren't at odds with the chopping frequency, but the conduction to off recovery time.

That's not to say the 1N4007s are inadequate, just that it might require further analysis if your efficiency is below expectation, especially if they are getting warmer than might be considered normal considering the forward drop vs average current.
View user's profile View All Posts By User
Hennig Brand
International Hazard
*****




Posts: 1284
Registered: 7-6-2009
Member Is Offline

Mood: No Mood

[*] posted on 15-5-2015 at 13:51


So when it takes them longer to change state, than a faster diode, they dissipate heat and even if it is not enough heat to destroy them it lowers overall circuit efficiency. I am not yet sure exactly how high a frequency designers would use a regular rectifier diode such as the 1N4007 in a voltage multiplier circuit. They seem to be working lovely at 15kHz, but 15kHz isn't really all that high as switching supplies go.

Quote: Originally posted by Varmint  
Henning:

The rectifiers aren't at odds with the chopping frequency, but the conduction to off recovery time.


I see what you mean, but the amount of times the diode has to change state per unit time is the frequency.


[Edited on 16-5-2015 by Hennig Brand]




"A risk-free world is a very dull world, one from which we are apt to learn little of consequence." -Geerat Vermeij
View user's profile View All Posts By User
Hennig Brand
International Hazard
*****




Posts: 1284
Registered: 7-6-2009
Member Is Offline

Mood: No Mood

[*] posted on 19-5-2015 at 06:59
Ferrite Transformer & Voltage Multiplier Design Guides


Ferrite Transformer:

Here is a good straightforward method for ferrite transformer design. Also attached is the Magnetics 2013 catalogue with specifications for most common core sizes and types.


Attachment: Ferrite Transformer & Inductor Design - MagneticsFerritePowerDesign2013.pdf (160kB)
This file has been downloaded 282 times

Attachment: Ferrite Core Specifications - Magnetics2013FerriteCatalog.pdf (1.9MB)
This file has been downloaded 301 times


Ferrite Transformer Design Sample Calculation:

WaAc = (Po * Dcma) / (Kt * Bmax * f)
WaAc = (5W * 500cir.mils/amp) / (0.001 * 2000gauss * 15000Hz) [used aggressive value for current density since power supply is only run intermittently]
WaAc = 0.0833

From table: 1814 Pot Core is large enough, however, since I had a 3019P core it was used.

Ac = 1.36 cm^2 [3019 effective area value from table]

Np = Vp * 10^8 / (4 * B * Ac * f)
Np = 24V * 10^8 / (4 * 2000 gauss * 1.36 cm^2 * 15000Hz) [12V DC supply voltage, which gives 24V peak to peak on transformer primary]
Np = 14.71 ---> round up to 15 turns
[Primary is double 15 (30 turns center tapped)]

Ns = Vs / Vp * Np
Ns = 500V / 24V * 15 turns
Ns = 313 turns
Secondary 313 turns

WaAc = 0.0833, Ac = 1.36cm^2
Wa = 0.0833 / 1.36cm^2
Wa = 0.06125

Kwa = NpAwp + NsAws
Wa = 0.06125, K = 0.6 for pot cores, let NpAwp = 1.1NsAws
Aws = (0.6 * 0.06125) / (2.1 * 313)
Aws = 0.000056 cm^2 or 0.0056 mm^2
Secondary: 40 AWG magnet wire

NpAwp = 1.1NsAws
Awp = (1.1 * 313 * 0.000056cm^2) / 15
Awp = 0.00129 cm^2 or 0.129 mm^2
Primary: 26 AWG magnet wire

I used a pot core, but an E core would have been fine too, maybe better in some respects. Also notice that I used a bit more turns than was really necessary when I wound the transformer above, but it works fine and the extra turns will keep the flux density down even lower. I hadn't done this calculation when I wound the transformer.


Voltage Multiplier:

Here is a voltage multiplier design guide. I made a small excel sheet, based on a couple of the equations from the guide, which allows quick half wave multiplier design comparisons to be made. Notice at low frequencies that the capacitors must be much larger in capacitance than at high frequencies. Notice how above I used 3kV capacitors for the two half wave multipliers made. Even going as low as 1kV caps probably would have been enough, since the components only need to be rated for the multiplier stage voltage which is about double the supply voltage (supply voltage is ca. 500V in the above case). Using components rated at least a little higher in value than absolutely needed is good practice though.


Attachment: Voltage Multiplier Design Guide.pdf (289kB)
This file has been downloaded 390 times

Attachment: Voltage Multiplier Excel Sheet.xlsx (10kB)
This file has been downloaded 326 times


[Edited on 20-5-2015 by Hennig Brand]




"A risk-free world is a very dull world, one from which we are apt to learn little of consequence." -Geerat Vermeij
View user's profile View All Posts By User
Hennig Brand
International Hazard
*****




Posts: 1284
Registered: 7-6-2009
Member Is Offline

Mood: No Mood

[*] posted on 21-5-2015 at 13:42
105ft of Blasting Line, 10g of ETN Putty Explosive Initiated With Custom Made ETN EBW Detonator


This test was performed nearly exactly the same as the last test, except that a new homemade power supply was used. The 0.5uF capacitor bank was charged to ca. 3800V before the homemade brass trigatron was triggered dumping the electrical charge through the blasting cable and bridgewire. The detonator was made using a 7.6mm id Al casing and 0.7g of ETN; 0.5g was pressed at 100lbs of force, 0.1g was hand pressed and 0.1g was put in nearly loose. The bridgewire used was 1.5mil diameter copper, about 100mil long. The witness target was 1/8 inch wall thickness square steel tubing. The blasting cable was made up of 100ft of RG-6 (standard cable, satellite, TV coaxial) and 5-6ft of 18 gauge speaker wire. The main charge was once again 10g of 80% ETN putty explosive.

The test was a complete success. Both 1/8" sides of the steel tubing were perforated in a very brisant detonation.

Right now this system is a bit inconvenient because it takes about 15 minutes to set up the modular fireset and run the line to the charge. Once all fireset components are assembled together permanently in a suitable small enclosure it should only take a couple of minutes to set up.

Being 105ft away is normally far enough for my purposes, but the distance could be doubled to 210ft with the same setup by simply using two 210ft lengths of the same cable type. Two 210ft pieces in parallel basically have half the inductance and resistance compared to a single cable of the same length and should perform similarly as a single piece 105ft long. There are also much better cables available, which would allow much longer cable runs to be made. In my opinion, however, 100ft is enough for most normal small scale blasting. If more distance is needed the blasting box could be sandbagged 100ft away from the blast and the operator could control the blasting box through long thin cable at a much greater distance away.

I used the old trigatron, in the wooden block, because I wanted to keep everything as much the same as earlier tests as possible.


100ft RG-6.jpg - 984kB Fireset 1.jpg - 425kB Fireset 2.jpg - 413kB Fireset 3.jpg - 443kB Setup 1.jpg - 1.1MB Setup 2.jpg - 1MB Setup 3.jpg - 970kB Post Blast 1.jpg - 505kB Post Blast 2.jpg - 503kB Post Blast 3.jpg - 530kB



Here is a current versus time graph generated using the "Distributed Parameter Method" Excel sheet attached earlier in this thread.

Current Versus Time Graph.jpg - 32kB

The theoretical numbers for the earlier failed test look better than this test, however the theoretical model assumed fast efficient switching. A trigatron was used in this successful test, and simple touching of 18 gauge wires was used as the method of switching for the earlier failed test. Fast efficient switching appears to be very important, but I am unable to quantify it exactly at this time.

For 3800V, 0.5uF, 105ft RG-6, Cu BW 1.5mil D, 100mil L
Current at or above Burst Action = 853 A
Time at or above Burst Action = 1.85 usec
Average rate of Current Rise = 462 A/usec
Trigatron Switching
Detonation Confirmed

For 2400V, 0.67uF, 51ft RG-6, Cu BW 1.5mil D, 100mil L
Current at or above Burst Action = 876 A
Time at or above Burst Action = 1.54 usec
Average rate of Current Rise = 876A/1.54usec = 569 A/usec
Switching accomplished by touching two 18gauge wires together
No detonation from test


[Edited on 22-5-2015 by Hennig Brand]




"A risk-free world is a very dull world, one from which we are apt to learn little of consequence." -Geerat Vermeij
View user's profile View All Posts By User
jock88
International Hazard
*****




Posts: 505
Registered: 13-12-2012
Member Is Offline

Mood: No Mood

[*] posted on 24-5-2015 at 14:19



How many ebw's could be put in series for a given set up or would you be better off using a bigger cap (same Voltage) and going for parallel?

Would NG be around the same sensitivity as ETN for ebw' detonators?
View user's profile View All Posts By User
Hennig Brand
International Hazard
*****




Posts: 1284
Registered: 7-6-2009
Member Is Offline

Mood: No Mood

[*] posted on 24-5-2015 at 15:28


These documents give a feel for what is possible with commercial gold EBW detonators. Results would likely be somewhat different with my homemade copper EBW detonators and fireset. The specs of the capacitors used in the RISI firesets are normally 1uF and 4000V from what I have seen. It looks as though quite a few commercial detonators can be fired simultaneously in various series parallel configurations depending on cable length, etc.

I have more reading material on series parallel connecting EBWs, but I can't locate it right at the moment.


Attachment: Series Parallel Firing of EBW Detonators.pdf (29kB)
This file has been downloaded 300 times

Attachment: FS-43 Firing System.pdf (35kB)
This file has been downloaded 285 times


Nitroglycerine is a homogenous explosive not well suited to this application, at least not in its pure form. It has very low stable detonation velocities, while PETN or ETN or MHN, etc, can be initiated low order by an EBW and then quickly accelerate to high order. Progressive increase in explosive density aids the explosive in accelerating from low order to high order detonation as well, something that would be impossible to do with a homogenous explosive like NG.


[Edited on 25-5-2015 by Hennig Brand]




"A risk-free world is a very dull world, one from which we are apt to learn little of consequence." -Geerat Vermeij
View user's profile View All Posts By User
Hennig Brand
International Hazard
*****




Posts: 1284
Registered: 7-6-2009
Member Is Offline

Mood: No Mood

[*] posted on 24-5-2015 at 18:19
Much Better Cables Are Available


Just did a theoretical comparison between RG-6 coax and the "C" cable sold by RISI using the distributed parameter method. Using "C" cable could easily double the distance that my system would function effectively, compared to when using RG-6, everything else being equal. I would like to get some of this cable or something similar.

RG-6 Coaxial
For 3800V, 0.5uF, 105ft RG-6, Cu BW 1.5mil D, 100mil L
Current at or above Burst Action = 853 A
Time at or above Burst Action = 1.85 usec
Average rate of Current Rise = 462 A/usec
Trigatron Switching
Detonation Confirmed

"C" Cable Coaxial
For 3800V, 0.5uF, 210ft RISI "C" Type Coax, Cu BW 1.5mil D, 100mil L
Current at or above Burst Action = 763 A
Time at or above Burst Action = 1.44 usec
Average rate of Current Rise = 531 A/usec


RG-6 Coaxial Versus RISI C Type Coaxial.jpg - 68kB


[Edited on 25-5-2015 by Hennig Brand]




"A risk-free world is a very dull world, one from which we are apt to learn little of consequence." -Geerat Vermeij
View user's profile View All Posts By User
Hennig Brand
International Hazard
*****




Posts: 1284
Registered: 7-6-2009
Member Is Offline

Mood: No Mood

[*] posted on 29-5-2015 at 03:39


Quote: Originally posted by Hennig Brand  
These documents give a feel for what is possible with commercial gold EBW detonators. Results would likely be somewhat different with my homemade copper EBW detonators and fireset. The specs of the capacitors used in the RISI firesets are normally 1uF and 4000V from what I have seen. It looks as though quite a few commercial detonators can be fired simultaneously in various series parallel configurations depending on cable length, etc.


What I said about the capacitors in the commercial EBW firesets was kind of misleading I think. More accurately, normally the commercial units I have seen described have 1uF of capacitance (in total) and are designed to be charged to 4000V. If the capacitor or capacitor bank is designed to be charged to 4000V it is likely rated for 5000V or more.




"A risk-free world is a very dull world, one from which we are apt to learn little of consequence." -Geerat Vermeij
View user's profile View All Posts By User
Hennig Brand
International Hazard
*****




Posts: 1284
Registered: 7-6-2009
Member Is Offline

Mood: No Mood

[*] posted on 3-6-2015 at 11:35


I was going from my memory of earlier tests when I wrote down the amount of force used to press the ETN with the lever press. I just checked today and I am easily pressing with at least 300lbs of force the way I have been doing it. I incorrectly stated that it was only 100lbs of force in several places in the last couple of pages in this thread. Here are four examples that I found. Any more that state 100lbs of force should also say 300lbs as well.

First post second line here:
http://www.sciencemadness.org/talk/viewthread.php?tid=23466&...

" A half gram of ETN was pressed in a 7.6mm aluminum casing with over 100lbs of force with a lever press and then another 0.2g was added loose on top."

It should say "with about 300lbs of force".

Same thing here:
http://www.sciencemadness.org/talk/viewthread.php?tid=23466&...

"About 0.2g was pressed with a lever press at about 100lbs of force, about 0.2g was pressed firmly by hand and about 0.2g was poured in loose and then only very gently compressed just to settle and remove large voids."

It should say "at about 300lbs of force".

And here:
http://www.sciencemadness.org/talk/viewthread.php?tid=23466&...

"The 7.6mm casing contained about 0.7g of ETN, 0.5g pressed at 100lbs, 0.1g hand pressed, 0.1g left almost loose."

And here:
http://www.sciencemadness.org/talk/viewthread.php?tid=23466&...

"The detonator was made using a 7.6mm id Al casing and 0.7g of ETN; 0.5g was pressed at 100lbs of force, 0.1g was hand pressed and 0.1g was put in nearly loose."

Maybe I am being a little obsessive, but there is a big difference between 300lbs and 100lbs.




"A risk-free world is a very dull world, one from which we are apt to learn little of consequence." -Geerat Vermeij
View user's profile View All Posts By User
nux vomica
National Hazard
****




Posts: 265
Registered: 18-7-2013
Member Is Offline

Mood: No Mood

[*] posted on 9-6-2015 at 05:32


I'm getting the closer to being able to do some proper testing now ive got new capacitors, there Epcos 5uf 1300v radial metalized polypropylene film with 0.006 ohm ESR I bought 4 so that gives me 5200v and 1.25 UF.

I had issues with mach3 on my cnc router that took time to work out so I just finished routing the board on the weekend but every thing seems to fit on okay , I built the spark switch into the board to tidy things up a bit and keep it close to the capacitors.

I've taken a short video of the piezo fireing the spark switch but thats about it so far , cheers nuxy.


20150609_181751.jpg - 1MB

20150609_185121.jpg - 954kB

[Edited on 9-6-2015 by nux vomica]
View user's profile View All Posts By User
nux vomica
National Hazard
****




Posts: 265
Registered: 18-7-2013
Member Is Offline

Mood: No Mood

[*] posted on 9-6-2015 at 05:45


Spark switch video

Attachment: 20150609_181820_001_001.3gp (9.3MB)
This file has been downloaded 492 times

View user's profile View All Posts By User
Hennig Brand
International Hazard
*****




Posts: 1284
Registered: 7-6-2009
Member Is Offline

Mood: No Mood

[*] posted on 9-6-2015 at 12:15


Looks very professional, well done. Those capacitors are far superior to what you had before for this purpose. You did a nice job of making a small tidy spark gap switch. Your setup will be quite compact that is for sure, which is a good thing.


I have a bunch of 2153 chips now so I should probably make at least one inverter using one. The inverter based on the ubiquitous 555 is working well enough for my purposes already though, at least in this application. The 2153 inverter circuit diagram below was taken from the following webpage:

http://danyk.cz/menic230_4_en.html

The 2153 datasheet has a graph allowing component choices to me made for a desired operating frequency.


IR2153 Inverter Circuit.jpg - 31kB


Attachment: IR2153 Data Sheet.pdf (156kB)
This file has been downloaded 274 times


[Edited on 9-6-2015 by Hennig Brand]




"A risk-free world is a very dull world, one from which we are apt to learn little of consequence." -Geerat Vermeij
View user's profile View All Posts By User
nux vomica
National Hazard
****




Posts: 265
Registered: 18-7-2013
Member Is Offline

Mood: No Mood

[*] posted on 10-6-2015 at 04:48


Just had a successful test used 0.030 mm dia x 1.6mm long copper bridgewire with .8 grm etn, dont know compression figures as in a rush but compressed .7 then used .1 grm loose alloy tube 8mm id x20mm long at the end of 7 meters of rg6 coax and 200mm speaker twin core.

Charged capacitors for 11seconds volt measurement would be over 4500 volts didn't have gauge just used times tested in shed then hit piezo button nice bang and a dent nearly through the 2mm s/s plate nice:D


20150610_210719.jpg - 1000kB 20150610_210709.jpg - 1.1MB 20150610_210835.jpg - 927kB

[Edited on 10-6-2015 by nux vomica]

[Edited on 10-6-2015 by nux vomica]

[Edited on 10-6-2015 by nux vomica]

[Edited on 10-6-2015 by nux vomica]
View user's profile View All Posts By User
Hennig Brand
International Hazard
*****




Posts: 1284
Registered: 7-6-2009
Member Is Offline

Mood: No Mood

[*] posted on 10-6-2015 at 10:42


A positive result is nice. So the bridgewire was around 0.63 mils in diameter or 54 AWG right? Seems extremely thin, but if it works it works. Must be very difficult not to break when handling. Was the bridgwire copper? It would have been nice to know more about the specifics of the test, such as capacitor bank voltage prior to firing. Your assumption about the voltage could be right, but maybe not.



"A risk-free world is a very dull world, one from which we are apt to learn little of consequence." -Geerat Vermeij
View user's profile View All Posts By User
nux vomica
National Hazard
****




Posts: 265
Registered: 18-7-2013
Member Is Offline

Mood: No Mood

[*] posted on 10-6-2015 at 14:46


I made a error on the wire dia have changed it in post to 0.030 mm or 1.225 mill (1 mill is 0.0245 mm so 0.030mm / 0.0245 equals 1.225 mill ) its copper from flash transformer I cant measure voltage over 2800 volts at the moment as I am useing a 5 volt analog meter with 5 m ohm resistance and power supply wont go over that level if connected, so I have run power for set times then touched wires on cap bank to measure the voltage.
Cheers nuxy.

[Edited on 11-6-2015 by nux vomica]
View user's profile View All Posts By User
Hennig Brand
International Hazard
*****




Posts: 1284
Registered: 7-6-2009
Member Is Offline

Mood: No Mood

[*] posted on 10-6-2015 at 15:05


edit: Oh I see, the power supply is low current as suspected, which could also mean that it won't take too much abuse before it becomes damaged. Good news it that it is fairly easy to build a decent power supply. It makes a nice little project in itself too.

Use a voltage divider which gives a lower proportional voltage for measuring. Even regular 1/4watt resistors can be used if enough of them are connected in series. Often times they are only rated for a few hundred volts each IIRC, but if you add ten in series the series string can easily handle a few thousand volts. High voltage resistors are also available, such as thick film resistors, if you want to go that route. I made a voltage divider that has about 40Mohms resistance and my multimeter is attached to it at the appropriate place so that 4000V on the capacitor bank reads as 25V, but it could be any voltage you choose. The meter I am using is only rated for 1000V DC so there was no way I could use it to measure the 4000V directly.

Voltage Divider.gif - 13kB


[Edited on 11-6-2015 by Hennig Brand]




"A risk-free world is a very dull world, one from which we are apt to learn little of consequence." -Geerat Vermeij
View user's profile View All Posts By User
nux vomica
National Hazard
****




Posts: 265
Registered: 18-7-2013
Member Is Offline

Mood: No Mood

[*] posted on 10-6-2015 at 15:45


Quote: Originally posted by Hennig Brand  
A positive result is nice. So the bridgewire was around 0.63 mils in diameter or 54 AWG right? Seems extremely thin, but if it works it works. Must be very difficult not to break when handling. Was the bridgwire copper? It would have been nice to know more about the specifics of the test, such as capacitor bank voltage prior to firing. Your assumption about the voltage could be right, but maybe not.

I should listen to the quote about assumption being the mother of all fuxkups. ;)
View user's profile View All Posts By User
Hennig Brand
International Hazard
*****




Posts: 1284
Registered: 7-6-2009
Member Is Offline

Mood: No Mood

[*] posted on 11-6-2015 at 13:16


I have made a few assumptions in my time too and reaped the rewards. ;)

From the above picture it looks as though your power supply is being supplied by only two AA batteries. That in itself tells a lot about the limited current supplying capability of the power supply. Power supplies have losses, usually quite significant especially with small power supplies, but lets assume for a minute that yours has no losses (Power In = Power Out). Lets assume you can pull 1A from the two AA batteries at 3V. Using conservation of energy, power in is 3V * 1A = 3W, so at 4500V the current drawn could be a maximum of 3W / 4500V = ca. 0.67mA. At 4500V the 5 Mohm Volt meter would draw, 4500V / 5 000 000 ohm = 0.9mA (0.9mA > 0.67mA). I went through these same problems when I was using the bug zapper supplies.




"A risk-free world is a very dull world, one from which we are apt to learn little of consequence." -Geerat Vermeij
View user's profile View All Posts By User
nux vomica
National Hazard
****




Posts: 265
Registered: 18-7-2013
Member Is Offline

Mood: No Mood

[*] posted on 11-6-2015 at 15:21


Had another test also successful l used same setup with copper bridgewire 0.030 x 1.6mm long .7 grams etn compressed to 1.49 grams cc and .1 gram loose on top, 7 meters coax and 200mm twin speaker wire.
Haven't been able to get to electronic supply shop so don't have accurate voltage levels just charged for 11 seconds as before and pushed pezio button detonation was instantaneous and plate was more damaged than last shot :D nuxy

[Edited on 11-6-2015 by nux vomica]

20150612_082522.jpg - 2MB20150612_083537.jpg - 1009kB20150612_083515.jpg - 977kB
View user's profile View All Posts By User
 Pages:  1  2    4    6  ..  11

  Go To Top