Welcome
Username:

Password:


Remember me

[ ]
[ ]
Online
Guests: 31, Members: 1 ...
Justin viewing error.php

most ever online: 333
(Members: 3, Guests: 330) on 06 Jun : 15:15

Members: 3343
Newest member: kzoo
Members Birthdays:
All today's birthdays', congrats!
101111 (26)
Danny (45)


Next birthdays
03/28 uzzors2k (28)
03/29 GrantX (27)
03/30 Adam Horden (32)
Contact
If you need assistance, please send an email to forum at 4hv dot org. To ensure your email is not marked as spam, please include the phrase "4hv help" in the subject line. You can also find assistance via IRC, at irc.shadowworld.net, room #hvcomm.
Support 4hv.org!
Donate:
4hv.org is hosted on a dedicated server. Unfortunately, this server costs and we rely on the help of site members to keep 4hv.org running. Please consider donating. We will place your name on the thanks list and you'll be helping to keep 4hv.org alive and free for everyone. Members whose names appear in red bold have donated recently. Green bold denotes those who have recently donated to keep the server carbon neutral.


Special Thanks To:
  • Aaron Holmes
  • Aaron Wheeler
  • Adam Horden
  • Alan Scrimgeour
  • Andre
  • Andrew Haynes
  • Anonymous000
  • asabase
  • Austin Weil
  • barney
  • Barry
  • Bert Hickman
  • Bill Kukowski
  • Blitzorn
  • Brandon Paradelas
  • Bruce Bowling
  • BubeeMike
  • Byong Park
  • Cesiumsponge
  • Chris F.
  • Chris Hooper
  • Corey Worthington
  • Derek Woodroffe
  • Dalus
  • Dan Strother
  • Daniel Davis
  • Daniel Uhrenholt
  • datasheetarchive
  • Dave Billington
  • Dave Marshall
  • David F.
  • Dennis Rogers
  • drelectrix
  • Dr. John Gudenas
  • Dr. Spark
  • E.TexasTesla
  • eastvoltresearch
  • Eirik Taylor
  • Erik Dyakov
  • Erlend^SE
  • Finn Hammer
  • Firebug24k
  • GalliumMan
  • Gary Peterson
  • George Slade
  • GhostNull
  • Gordon Mcknight
  • Graham Armitage
  • Grant
  • GreySoul
  • Henry H
  • IamSmooth
  • In memory of Leo Powning
  • Jacob Cash
  • James Howells
  • James Pawson
  • Jeff Greenfield
  • Jeff Thomas
  • Jesse Frost
  • Jim Mitchell
  • jlr134
  • Joe Mastroianni
  • John Forcina
  • John Oberg
  • John Willcutt
  • Jon Newcomb
  • klugesmith
  • Leslie Wright
  • Lutz Hoffman
  • Mads Barnkob
  • Martin King
  • Mats Karlsson
  • Matt Gibson
  • Matthew Guidry
  • mbd
  • Michael D'Angelo
  • Mikkel
  • mileswaldron
  • mister_rf
  • Neil Foster
  • Nick de Smith
  • Nick Soroka
  • nicklenorp
  • Nik
  • Norman Stanley
  • Patrick Coleman
  • Paul Brodie
  • Paul Jordan
  • Paul Montgomery
  • Ped
  • Peter Krogen
  • Peter Terren
  • PhilGood
  • Richard Feldman
  • Robert Bush
  • Royce Bailey
  • Scott Fusare
  • Scott Newman
  • smiffy
  • Stella
  • Steven Busic
  • Steve Conner
  • Steve Jones
  • Steve Ward
  • Sulaiman
  • Thomas Coyle
  • Thomas A. Wallace
  • Thomas W
  • Timo
  • Torch
  • Ulf Jonsson
  • vasil
  • Vaxian
  • vladi mazzilli
  • wastehl
  • Weston
  • William Kim
  • William N.
  • William Stehl
  • Wesley Venis
The aforementioned have contributed financially to the continuing triumph of 4hv.org. They are deserving of my most heartfelt thanks.
Forums
4hv.org :: Forums :: Electromagnetic Projectile Accelerators
<< Previous thread | Next thread >>   

Another Coilgun

Go to page  [1] 2 3 4 5
Author Post
DerAlbi
Sat Apr 11 2015, 11:53AM Print View
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
As i am more active here in the last days.. i present my Coilgun-Progress here.
  • Its a multistage halfbridge design.
  • Projectile is 9.7mm in Diameter.
  • Optimized for the figure of merit: Ekin.Eff^2
  • I work on it since last August..

As some may know me.. i like numbers over pictures. So.. here are the numbers:

results.pdf

Since this is quite record breaking and makes the leap forward to possible commercial production i am sorry for holding back the detailed specifiactions of the design... i kind of post this here to hope that people gain motivation to investigate more than the old SCR-based design..
Halfbridges rock. Check the Data, throw away your SCRs. i actually would have some pucks for sale
Back to top
aarpcard
Mon Apr 13 2015, 04:01AM
Registered Member #2848
Joined: Tue May 04 2010, 05:19AM
Location:
Posts: 44
Wow, those numbers look fantastic! I've always wanted to try using half bridges for switching, but I've had a hard time finding appropriate modules for affordable prices.

Just curious, what halfbridge or IGBT modules are you using?
Back to top
DerAlbi
Mon Apr 13 2015, 08:53AM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
200x IRGB30B60K. IGBT-Driver is self designed. Handles short circuit overload and unexpected timing-faults.
I am currently winding coils. I used my Data to extrapolate the next coil properties... wish me luck.
I am not understanding why my optimization of Ekin*Eff^2 leads to the linear rise of velocity... i wonder when this trend breaks down.. i cant really beliefe this is practical in the long run.
Back to top
hen918
Mon Apr 13 2015, 05:23PM
Registered Member #11591
Joined: Wed Mar 20 2013, 08:20PM
Location: Alfreton, UK
Posts: 391
DerAlbi wrote ...

200x IRGB30B60K. IGBT-Driver is self designed. Handles short circuit overload and unexpected timing-faults.
I am currently winding coils. I used my Data to extrapolate the next coil properties... wish me luck.
I am not understanding why my optimization of Ekin*Eff^2 leads to the linear rise of velocity... i wonder when this trend breaks down.. i cant really beliefe this is practical in the long run.


200! in parallel! that must have cost a bit, and taken some time to solder!
Back to top
DerAlbi
Tue May 05 2015, 02:28PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
Extreme set back One puck style diode failed and caused a short circuit. At least one of the paralleled IGBT failed, shorted the gate to the Collector and held all others open. 20 IGBT blew up. Today i cleaned the PCB, and i am currently at resoldering new IGBTs.. near tears
I think the Diode had not enough pressure at it. common failure Stupid Albi.
Back to top
DerAlbi
Wed May 06 2015, 05:24PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504



While waiting for a replacement diode, i wrote to Kickstarter-Support what they count as weapon (since they prohibit weapons)...
I am kind of extremely worried and helpless and i dont know what to do. When my prototype blows up, i have to build a new one (which is not planned right now). All i want right now is a technology demonstration (proof of concept) to get a Kickstarter-campain going.
In principal i am done with the project. I achieved insane efficiency and the only thing left is to add coils until i reach my output power with ever increasing efficiency. I dont know if its wise to take shortcut and start immediately with a project demonstration before the prototype vanishes in smoke completely. If this happens it would set me back for 6 months and some big money -.- i would hate this. On the other side.. showing an unfinsihed prototype is kind of odd.. specially when you ask for >100k$ for a medium series of the first 100 devices...

The problem is the failure: its pure mechanically (the diode proessure) which causes the IGBTs to blow - i cant avoid that... Paralleling IGBTs is stupid. But it was the best option at the time. I hate myself. Should have spent more money. Should have done the math much earlier. Stupid Albi.
Back to top
BigBad
Wed May 06 2015, 08:36PM
Registered Member #2529
Joined: Thu Dec 10 2009, 02:43AM
Location:
Posts: 598
That looks like quite a severe failure

Maybe you could try redesigning it more along the lines of my massive staging architecture.

That could be a lot cheaper.
Back to top
DerAlbi
Thu May 07 2015, 02:00AM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
"Redesign" is a word i like to avoid right now Given all i know now and all the experience i gained from that prototype i have so much to improve and to simplify that a resdesign would affect everything. Thats why i would throw me back for such long time and why it would cost additional money i am not willing to spend due to such shitty mechanical failure..
There is a redesign planned; however more as part of a 2 stage approach to a commercialized solution (with the help of Kickstarter)... doing a third iteration would really change it from hobby to work. I woudnt like that right now
The cost point here.. issnt that "BigBad" . Its around 30$ only. the wok involved and the lost trust to the design is much worse. And the PCB wears from all the heating.. it wont withstand a second repair.
In the meantime i did some research and i still do not understand why my IGBTs blew up. It should have been safe.

Here is the evil life sucking diode


Notice the burnpattern inside the casing where the die touched. its only outside of the contact plate and only at one side.
The die is only damaged at one side too.
Notice also the yellowish stuff inside the casing. I understand the black stuff, but the yellow?? hmmh. It also smelled like rotten egg when i opened the case. If someone knows that failure pattern... you are welcome to share your experience.
Was the case damaged before ground zero ?
Back to top
BigBad
Thu May 07 2015, 03:39PM
Registered Member #2529
Joined: Thu Dec 10 2009, 02:43AM
Location:
Posts: 598
Have you done a ltspice on the circuit?

When I analysed mine I found I had a big resonance; I was putting in 300 volts, and getting out 1000+ volts.
Back to top
DerAlbi
Thu May 07 2015, 05:48PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
I have done extensive simulation including a 3D FEM-Simulation of the IGBT-arangement. I use active clamping.. there cant be any spikes above 450V. Anyway the IGBt would break down first since they are only 600V rated while the diode is 1kV rated.

I got new diodes today.. i am currently assembling the whole thing again. i suspect my driver is still damaged.. i will see that later.
And i did some super macro shots with my webcam.... check out the contact surface within the puck-package where the die was pressed togehter:


All those little marks.. i suspect that are contact burns due to missing pressure until the die was damaged sufficiently. Or my interpreatation is wrong and the marks are from the exploding die.. however the metalplates covering the die are still intact.

anyway: this time i add some more pressure. lets hope i replaced the IGBTs correctly and built in no new mechanical weakness.... (like a bad solder joint, missing gate contact and all such idiotic stuff that simply happens even if you check twice)
Back to top
DerAlbi
Sat May 09 2015, 07:23PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
What shall i say... KaaaBooOOOoommm! again. Yeah. really. Different IGBT module and no diode failure. It was a result of the repair.
The design was perfect for demonstrating all my optimizations.. but it wont take me to Kickstarter as usable prototype.
I am currently on searching parts for a redesign which is then way closer to a final design which might also make it more believable that i can make the gun into a usable formfactor. This may actually help a Kickstarter-campain..
At the end of the day this is just one prototype more. I will go up in voltage a bit just to see if that helps some issues.
But this thread will go inactive for some time. Sry guys.
Its so frustrating when the project is a complete success but its simply not enough to be presented.
Back to top
DerAlbi
Mon Jun 08 2015, 04:25PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
Just to give an update to anyone who is interested... i do not have specific progress.. its more a general overview...

The last few weeks were pretty intense.. I redesigned the whole Coilgun in one big go and took everything into account what i learned from the old design.
I optimized the design in formfactor, complexity, cost and weight.
This includes that i drastically reduced my available capacitance (stored energy) so that the whole energystorage weights only 1/6th of the previous version (now expected 860g). The IGBTs current handling capability is scaled down to values which i actually expect (<1kA).
PCBs are made with only 70um copper which - simulation said - is reasonable or in other words "good enough" (saves PCB cost and weight).
I am currently also fishing for suppliers for square enameled aluminum wire for my coils. But i found only chinese suppliers which ship in multiples of 10kg or if the wire is not in stock the minimum order quantitiy is 500kg
Since this seems currently unreachable i will continue to use round copper wire.. which will give better efficiency but tripples the weight. That tradeoff is one to come in a later design. Since ive got world leading efficiency anyway i am happy to sacrifice it to have a portable version (overall weight is expected to be <6kg including a casing and everything - still heavy.. but its simply what you need to have for 100J+ kinetic enery )
The circuits are also made easier... the IGBT drivers are much more staight forward and safetyfeatures are not implemented at driverlevel as before.

This all will be a real testrun for a commercial design. I also try a new optical projectile sensing system.. i do currently not expect it to work, but its worth "a shot", haha, sry^^

Wish me good luck.I will need it. Driving IGBT at such overload and capacitors so far over their ratings will provide some challanges. I am not sure if the IGBTs surive (didnt buy the fattest ones, but shopped for price) and the Capbank will actually have 50mOhm ESR which is quite a lot. I am extremely excited to finally get to know what happens with cheap consumergrade capacitors at constant current-overload usage.. i want to see if they degrade over time or if they surive the abuse. (hope so) I am currently "believing" that the capacitory should surive as long as the electrolyte does not boil inside the caps which should not happen at pulse loads.
The whole new design will have less efficiency compared to the original overengineered one... but thats all right.. as said, i have to keep weight in mind too. I still think that i will reach >30%eff easily... but not approaching the 40% as before extrapolated. Using aluminum coils later on will degrade efficiency further... but i actually dont think that it will drop too much. But i am not sure......
Back to top
hen918
Mon Jun 08 2015, 06:21PM
Registered Member #11591
Joined: Wed Mar 20 2013, 08:20PM
Location: Alfreton, UK
Posts: 391
DerAlbi wrote ...

Just to give an update to anyone who is interested... i do not have specific progress.. its more a general overview...

The last few weeks were pretty intense.. I redesigned the whole Coilgun in one big go and took everything into account what i learned from the old design.
I optimized the design in formfactor, complexity, cost and weight.
This includes that i drastically reduced my available capacitance (stored energy) so that the whole energystorage weights only 1/6th of the previous version (now expected 860g). The IGBTs current handling capability is scaled down to values which i actually expect (<1kA).
PCBs are made with only 70um copper which - simulation said - is reasonable or in other words "good enough" (saves PCB cost and weight).
I am currently also fishing for suppliers for square enameled aluminum wire for my coils. But i found only chinese suppliers which ship in multiples of 10kg or if the wire is not in stock the minimum order quantitiy is 500kg
Since this seems currently unreachable i will continue to use round copper wire.. which will give better efficiency but tripples the weight. That tradeoff is one to come in a later design. Since ive got world leading efficiency anyway i am happy to sacrifice it to have a portable version (overall weight is expected to be <6kg including a casing and everything - still heavy.. but its simply what you need to have for 100J+ kinetic enery )
The circuits are also made easier... the IGBT drivers are much more staight forward and safetyfeatures are not implemented at driverlevel as before.

This all will be a real testrun for a commercial design. I also try a new optical projectile sensing system.. i do currently not expect it to work, but its worth "a shot", haha, sry^^

Wish me good luck.I will need it. Driving IGBT at such overload and capacitors so far over their ratings will provide some challanges. I am not sure if the IGBTs surive (didnt buy the fattest ones, but shopped for price) and the Capbank will actually have 50mOhm ESR which is quite a lot. I am extremely excited to finally get to know what happens with cheap consumergrade capacitors at constant current-overload usage.. i want to see if they degrade over time or if they surive the abuse. (hope so) I am currently "believing" that the capacitory should surive as long as the electrolyte does not boil inside the caps which should not happen at pulse loads.
The whole new design will have less efficiency compared to the original overengineered one... but thats all right.. as said, i have to keep weight in mind too. I still think that i will reach >30%eff easily... but not approaching the 40% as before extrapolated. Using aluminum coils later on will degrade efficiency further... but i actually dont think that it will drop too much. But i am not sure......


Well Done!
I wish I had the time and the ability to stay interested in one project for that length of time.
Back to top
DerAlbi
Mon Jun 08 2015, 06:40PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
I do actually kind of design my life so that i have the time. I have a job at the local university here, where i only work 19h/week (24h/week soon) and there i am mainly involved in teaching. That leaves my head open for my technical problems at home and thats ok for now since that may produce future income, if i finally commercialize the accelerator.
And i do strongly feel that i need to grab that opportunity.. i know myself and i am a terrible worker for anyone other than myself (I simply have no motivation to solve the booring problems of others)
So this is a strong motivation too and i do what i takes...
And yes it compromises social life sometimes... but that comes with the territory i guess. I do actually know no one who, in my eyes, is really gifted and is socially healthy..
Back to top
DerAlbi
Sat Jun 20 2015, 12:39AM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
Ok. I just finished cleaning up my room. After another 500eur spent, ive got at least some progress.
I redesigned my capbank and my IGBT-setup. It is untested so far (under high load), however it looks promising so far.
The old cap bank gave me around 0.67J/g (Engergy/Weight) energydensity while the new design delivers 0.61J/g however that includes all IGBTs and drivers. The overall weight is now 1.25kg... Formfactor is extremely compact... measuring 5x6x43cm

At the other hand ive also reworked the coilgun main controller.
I was fed up by the shitty lighttrap-stuff (all the LEDs, IR-receivers, and optical cable) that i want to get rid of that concept alltogether.
So what i tried right now is to put a laser at the back of the barrel and build a laser distance measurement system to track the projectile position.
Today i got actually a position dependend voltage out of the measurement system. The laser has to be aligned quite precice but it worked somehow. The deisgn is currently extremely faulty and patched up on all ends...

I am currently not sure if that will work at all, or if it will be precise enough but i will give it another try.. i expect it to fail actually, even if i redeisng it with all i leanred... but hey.. its interesting!

Here some pictues of the testsetup...





Aaaaaand the Halfbdrige of course..



And the driver board alone..


Back to top
Signification
Sat Jun 20 2015, 01:46AM
Registered Member #54278
Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 285
I've had luck with this optical method when lasers and alignment became issues...



Back to top
DerAlbi
Sat Jun 20 2015, 02:49AM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
I thought about something like that, but i dismissed it. i dont want to have so much coil spacing anymore (assuming a the 3mm LED) that significantly reduces energy density. I also wanted a method that does not only give me discrete steps, but a continous position tracking...
With the light traps (discrete position resolution) i had to do a 4th order quadratic regession every light trap event in order to track the projectile. i would like to get rid of this computational nightmare and the guessing errors that the regression made...
Back to top
DerAlbi
Fri Jun 26 2015, 07:19PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
Small progress:
1.8MB stupid animation just for fun



normal Foto


I assembled the CapBank completely. These capacitors are connected in a 2s8p configuration. The capacitance of the 2 series branches differs sligtly so that at 600V bankvoltage the 2 halfes are ~7V different. There is also a asymetric leakage current that drives the voltages further apart over time if the bank is constantly connected to a 600V voltage soruce. Luckyly my ballancing circuit takes care of that perfectly. But i think for later designs the cap charger flyback should have 2 output transformer windings to charge the 2 halfes seperately. That would ensure same voltage no matter how unmatched the caps are.

Another trap for youngplayers (yes, i am young):
I, being stupid as fuck, bought the wrong parts from digikey.
Problem: my IGBTs are protected by TVS diodes that clamp the collector voltage to the gate (that topology is called active clamping) to prevent excessive voltage spikes due to ESL and strayinductance.


Note that the current TVS diode is not bidirectional. So it behaves as a normal diode in one direction.
What happens? If the IGBT turns on, (assuming light load) the collector goes down to 1V while there is 15V at the gate. So the TVS actually conducts. Thus R1 (47R) draws some current from the gate drive and whats even worse: the base-emitter-diode of Q1 goes into reverse breakdown (-7V typical). That limits the gatevoltage effectively to around 10V making the gatedrive completely unusable.
At the end of the day: never use unidirectional TVS diodes in an active clamping circuit.
As allways: bi offers more possibilities

Concerning the projectile position: A new 4 layer board with improved circuit and layout is on the way. Ask me again in 10 days. Prediction: i will work better, but ultimately will not be usable. That revision decouples laserdrive and photoinput much better and should offer higher overall sensitivity.
As for couriosity there are also 50mW laser on the way.. in the hope that i dont get problems during import... but i do not plan to use those lasers (50mW IR is simply too dangerous), as long as there is any possibility to get along with 5mW red dot china crap.
Back to top
Signification
Fri Jun 26 2015, 08:50PM
Registered Member #54278
Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 285
You didn't state the "C" and "V" values of the caps--are they all the same?

Please--don't fool around with 50mwIR laser where the visible red will work. There is an abundance of medium-power long-wavelength-visible red that will work fine. 50mw ir will get hot to the skin and instantly vaporize your fovea! If you burn this tiny spot on your retina, you lose the only high resolution (greatest cone-concentrated) pinpoint area retinal region sufficient to focus with enough resolution for things like reading. If you damage this tiny spot, you will be able to see these messages just fine, but will NOT be able to read them!! The fovea is most likely to be damaged when you are working with the ir laser since you will be focusing on dangerous places. See the article "COMPUTER GENERATED ANAGLYPHS" by Dale Nassar in Circuit Cellar INK: the computer applications journal.
Back to top
DerAlbi
Fri Jun 26 2015, 09:13PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
Use that tiny spot in the eye of yours


380LQ/102M/315/K052

And as i have written: i bought the lasers - but no intention of usage if its not absolutely unavoidable. 780nm arent completely invisible.. and i am actually most afraid of eye damage due to indirect/reflected light. Since there is no immeadiate pain at one moment one would dream about boobies and watch the glowing spot at the wall and thats it allready.. within seconds...
I hope at work i could borrow saftety equipment. But 780nm is rare and therefore expensive. Not sure if its available.
The 780nm are interesting for me because it offers nearly 50% improvement in terms of photodiode sensitivity. If anyone knows IR lasers in that package: (6mm brass case) i would be happy to try smaller diodes than these blasting 50mW.
Back to top
Signification
Fri Jun 26 2015, 09:24PM
Registered Member #54278
Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 285
DerAlbi wrote ...

Small progress:
1.8MB stupid animation just for fun



normal Foto


I assembled the CapBank completely. These capacitors are connected in a 2s8p configuration.


OK, I see on closeup the caps are each 1000uF @ 315V = 50J ea.

Dosen't 2s8p mean a pair of 8-in-parallel connected in series: = 4000uF @ 630V (794J)? This indicates 16 caps (8x2)--I count 20 (4x5) = 992 J total energy.
Back to top
DerAlbi
Fri Jun 26 2015, 09:41PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
Yes yes.
1) 4 for spare parts, destructive tests or replacements, in case i burn the insulation by wildly swinging the soldering iron.. also the cute chinese sales woman was only able to ship multiples of 10 caps.
2) actually only around 650J
3) because i do not use the full votlage rating. The IGBTs are 650V rated. I need some headroom for active clamping TVS diode tolerances and stuff so that i dont see how 630V could be my working voltage.. even if that would be desireable. Currently i plan 580V or maybe 600V.. but not much more. I think such working voltage utilizes the caps enough.. i mean technically there is a surge voltage rating upto 350V.. but.. aawwww nooohhhh.
Since i was heading towards (exceeded) 30% efficiency my deisgn goal of 100J would consume 300J electrical energy which would result in ~400V cap voltage. At that voltage the last coils should have a current that is hopefully still within specs for the IGBT pulse rating. I am not sure however how that cap bank will perform. The combined ESR will be about 50mOhm which feels extremely high.. but math says that its ok. What is most interesting to me is how the cap bank performs over time with the pulse load way over spec...
4) Those caps have only 940uF. They cheated! Suckers.
Back to top
Signification
Fri Jun 26 2015, 10:21PM
Registered Member #54278
Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 285
I have twenty of 10,000uF @ 500V (525 surge) and some 12,000uF--all electrolytics in the standard blue pack that looks like yours. That's over a kilo-Joule per cap! They are 3" x 8-1/2" cylinders. The truly worthy caps have the very massive electrode terminals. I also have some monster IGBT bricks with snubber boards...don't remember the exact specs but you can look on ebay: seller ID:amitela I may just keep them for my smaller coilgun.
Back to top
DerAlbi
Fri Jun 26 2015, 10:46PM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
Hmmh.. i did select those Caps on purpose.. they are cheap, readily available, there are other supliers with similar formfactors and the caps are small. Smaller caps provide the possibility to shape my capbank in a formfactor that suits me best. I do also like the distributed nature of smaller caps which keeps the ESL down since i have multiple half bridges to supply thus reducing switching losses - speking only about the conncection between caps and halfbridges. Additionally of course the SFR of smaller caps are inherently higher. If that is only due to smaller capacitance - honestly i dont know.. but i think smaller cans must have smaller internal ESL and paralleling them also helps.

Lager caps also provde less Joules per Cost since they are simply not high volume production junk.
I considered larger caps.. but the tradeoffs never outweighed the profits.
Back to top
DerAlbi
Tue Jun 30 2015, 12:11AM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
Hooooooooooooooollllyyyyy Shit.
I just got a reply from a wire manufacturer... they want to charge 880$ for square aluminum per kilogram. whoooooooooooooooooooot
Maaaan I am seriously thinking about using heavy coated round aluminum wire and making the coils physically larger. Then pressing them in shape, thus increasing the conductor density in the coil. Since there are no sharp edges the insulation should survive such abuse... but i need a hell of a press.
Any thoughts about those desperate measures?
Back to top
Signification
Tue Jun 30 2015, 03:59AM
Registered Member #54278
Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 285
Why would you be interested in Al?
Back to top
DerAlbi
Tue Jun 30 2015, 08:07AM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
Because i have such high efficiency, that i can sacrifice it in favor of halving the gun weight. Aluminium coils would be awesome!
Back to top
Dr. Slack
Tue Jun 30 2015, 08:18AM
Registered Member #72
Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1554
DerAlbi wrote ...

Hooooooooooooooollllyyyyy Shit.
I just got a reply from a wire manufacturer... they want to charge 880$ for square aluminum per kilogram. whoooooooooooooooooooot
Maaaan I am seriously thinking about using heavy coated round aluminum wire and making the coils physically larger. Then pressing them in shape, thus increasing the conductor density in the coil. Since there are no sharp edges the insulation should survive such abuse... but i need a hell of a press.
Any thoughts about those desperate measures?



Thoughts about those desperate measures - if you have such high efficicency that you can sacrifice the cinductivity of copper for alli, your efficiency is high enough that you can sacrifice the packing fraction of square for round.
Back to top
DerAlbi
Tue Jun 30 2015, 10:46AM
Registered Member #2906
Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 504
I wish it was so easy.
My copper coils are made from round copper. Their weight / volume indicates that (if there was no epoxy inside the coil) the copper volume is 78%.
That is (by accident due to the epoxy) the theoretical maximum for round copper (pi/4=0.785)
By using square aluminum wire i would gain quite some conductivity and still reducing weight. Infact in theory square aluminum wire has 80% of the conductivity of roud copper wire. The sacrifice issnt that big. But only if one uses square wire.
Sacrificing efficiency by 20% would still give me the chance to reach 30% overall efficiency. maybe 27%. Sacrificing more efficiency would result in bigger capacitors (more weight again, but that is also reduced in coil weight due to lower aluminum fill factor) and higher currents (more expensive IGBTs, which is the main problem here).
There is not much room to play with. Capacitors, IGBTs and the coils are the most problematic things to design. The sweet spot combining all 3 things with respect to weight and efficiency is not that easy.
I cant get too sloppy with either of thise design issues.

Edit:
I just received 100 projectiles. Amazing. they are made of steel and perform nearly equal to the pure iron that i had before. the new projectiles have a little different shape.. they are a little bit longer and their weight is 25.84g. But they are standardized and readily available.
That makes my goal to reach 100m/s a little harder since that implies 130J kinetic energy Given my cap bank is charged to 600V initially would need an efficiency of 29.15% to stay within the iGBT current limits. Since my cap bank has such high ESR that might still be possible with copper coils, but not with aluminium.. Addins 200g of capacitors might enable the possibility for aluminium however. Then i would need 23% efficiency - possible.
grrrrr. but... thats just math right now. need actual data!
Back to top
klugesmith
Tue Jun 30 2015, 05:49PM
Registered Member #2099
Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1464
Good discussion of fill factor and Al vs. Cu. Where do you get Al magnet wire? Where do you get square Al wire?
Don't forget that the Al coils will have a greater temperature rise per shot.

[edit]Round Al magnet wire probably comes annealed, for best conductivity. Maybe you could reduce it to a nearly square shape and keep the insulation, by drawing it between steel rollers (as opposed to ordinary wire drawing ). Also look into DIY application of insulating paint onto inexpensive plain wire. Help, somebody please stop me before I edit again! The Ring Lord company has square & round anodized aluminum wire , and we know that some vendors make Al magnet wire insulated by anodize coating instead of organic resin. [/edit]

Your pictures of blown-up IGBT array made me sad, but that's part of the energetic things business.
Does not mean the engineering or manufacturing teams are weak.


Back to top
Go to page  [1] 2 3 4 5

Moderators: Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Bjørn, Dave Billington, Steve Conner, Wolfram, Kizmo, Mads Barnkob

Jump:     Back to top

Powered by e107 Forum System
 
Legal Information
This site is powered by e107, which is released under the GNU GPL License. All work on this site, except where otherwise noted, is licensed under a Creative Commons Attribution-ShareAlike 2.5 License. By submitting any information to this site, you agree that anything submitted will be so licensed. Please read our Disclaimer and Policies page for information on your rights and responsibilities regarding this site.