• Electronics and Embedded Programming V3
1,545 replies, posted
[QUOTE=VistaPOWA;36244007]Could someone explain how a Tesla coil works?[/QUOTE] It's using the principles of resonance. The secondary coil forms an LC circuit with the topload (the ball or toroid (donut shape) as the capacity), which has a natural frequency. If you excite the coil at it's natrual frequency, the voltage inside it will swing higher and higher, until it's so high that it creates lightning. A good analogy is a swingset, as it also relies on resonance: you have to push in just the right moment to make the swing go higher. If you push slower, or faster, you can't achieve as great a height.
DrLuke: This is the trigger circuit I'm currently using to fulfill that purpose (thanks to [url=http://forum.allaboutcircuits.com/member.php?u=143625]jimkeith[/url]): [img]http://eagle.undo.it:8083/img/cdwelder_12_trigger.png[/img] Effectively dumping a 0.47uF ceramic capacitor at 12V into the SCR gate.
Actually the main source of the capacitance is the distance between the primary and the secondary coil, if you take a look at a Tesla coil you'll notice there is quite a significant gap between the primary and secondary. The same principle is used in flyback (aka LOPT) transformers which are commonly used in CRTs to generate the 5kV to 30kV or so needed, but can also be used at lower voltages.. [QUOTE=Night-Eagle;36248394]DrLuke: This is the trigger circuit I'm currently using to fulfill that purpose (thanks to [url=http://forum.allaboutcircuits.com/member.php?u=143625]jimkeith[/url]): Effectively dumping a 0.47uF ceramic capacitor at 12V into the SCR gate.[/QUOTE] Err correct me if I'm wrong but should not the switch be on the gate instead of the anode ? Also where do you get 12V from. Actually why are you even using an SCR there. [t]http://eagle.undo.it:8083/img/cdwelder_08.jpg[/t] This is pretty poorly designed imo.
[QUOTE=chipset;36239257]Okay I won't post them in thumb tags. Better?[/QUOTE] Strangely enough yes. Chrome does not like resizing high resolution images and displaying multiple at a time. Runs like shit.
[QUOTE=Xera;36248830]Strangely enough yes. Chrome does not like resizing high resolution images and displaying multiple at a time. Runs like shit.[/QUOTE] Well if you reply to him and quote all the thumbnailed images it will run like shit as well.
[QUOTE=Chryseus;36248539]Err correct me if I'm wrong but should not the switch be on the gate instead of the anode ? Also where do you get 12V from. Actually why are you even using an SCR there. [/QUOTE] Q2 is a programmable unijunction transistor. R8 and R9 are "program" resistors used to set the voltage at which current can flow from anode to cathode. The program resistors are effectively a voltage divider that halve the +24V. The SCR is used to stop the 1.0 farad capacitor from discharging into the work prematurely (i.e. as soon as the welding electrodes are applied to the work). At any rate, the question is what to replace the blown SCR with (50RIA20 blown). [editline]8th June 2012[/editline] [QUOTE=Chryseus;36248539]This is pretty poorly designed imo.[/QUOTE] Please feel free to elaborate.
[QUOTE=Night-Eagle;36248949]Please feel free to elaborate.[/QUOTE] Where do I even begin to start, Q4, R11 and R12 are absolutely useless, it appears to be some kind of current limiting which should kick in at around 40A, not at all needed and it's sucking a ton of current from the base of Q3 requiring R3 and R5 to be ridiculously large. Q2, I have absolutely no idea why you would want to use a PUT, these are usually used in oscillators not power devices, simply connecting the switch through to the SCR should do perfectly fine. R10 has no purpose and D3 is not required unless you happen to be driving something inductive. If your going to use an SCR replace it with a decent SCR, preferably 200A or above. Also that PUT is completely the wrong symbol.
With what program do you guys make these circuits?
[QUOTE=Staneh;36249790]With what program do you guys make these circuits?[/QUOTE] People who don't know what their doing like to use the [url=http://www.falstad.com/circuit/]Java Circuit Simulator[/url] (just don't expect accuracy). Personally I use [url=http://www.linear.com/designtools/software/]LTSpice[/url] or [url=http://ngspice.sourceforge.net/]NGSpice[/url] to design circuits As for plain schematics or PCBs I use [url=http://www.altium.com/]Altium Designer[/url].
[QUOTE=Chryseus;36249841] Personally I use [url=http://www.linear.com/designtools/software/]LTSpice[/url] or [url=http://ngspice.sourceforge.net/]NGSpice[/url] to design circuits [/QUOTE] So LiveWire not as big as with LT or NGSpice? (I've had luck with ExpressSCH for drawing them out)
Livewire is pretty awesome for small to medium circuits, certainly much better than java circuit sim. Nothing though really beats a proper SPICE based simulator, they may be slightly harder to learn but the results are very accurate if you use proper device models. Qucs Gnucap NGSpice PSpice (non-free) Spice Opus Spice+ Multisim 10 (non-free) Are all good simulators.
While we're on the subject, Labcenter Proteus is based around Berkeley-Spice; the demo's surprisingly full featured and quite intuitive to learn. [URL="http://www.labcenter.com/download/prodemo_download.cfm#professional"]Have a play.[/URL]
Me when I discovered the amazing functionality of AVR Studio5's AVR simulator: [media]http://www.youtube.com/watch?v=EKy3qcbwnA0[/Media]
[QUOTE=DrLuke;36251249]Me when I discovered the amazing functionality of AVR Studio5's AVR simulator: [media]http://www.youtube.com/watch?v=EKy3qcbwnA0[/Media][/QUOTE] Didn't they just release AVR Studio 6? [url]http://www.atmel.com/microsite/atmel_studio6/[/url]
[QUOTE=ddrl46;36251714]Didn't they just release AVR Studio 6? [url]http://www.atmel.com/microsite/atmel_studio6/[/url][/QUOTE] I'm pretty happy with studio 5, it does everything I need.
[QUOTE=Chryseus;36249718]Q4, R11 and R12 are absolutely useless, it appears to be some kind of current limiting which should kick in at around 40A, not at all needed and it's sucking a ton of current from the base of Q3 requiring R3 and R5 to be ridiculously large.[/QUOTE] How did you get 40A? Seriously don't know. [QUOTE=Chryseus;36249718]Q2, I have absolutely no idea why you would want to use a PUT, these are usually used in oscillators not power devices, simply connecting the switch through to the SCR should do perfectly fine.[/QUOTE] There was concern that due to the amount of current being pulled through the thyristor, it might blow if it spent too long in its non-fully conducting state. [QUOTE=Chryseus;36249718] R10 has no purpose and D3 is not required unless you happen to be driving something inductive. [/QUOTE] I realized R10 served no purpose and did not include it in the build. D3 was suggested just for the reason you mentioned. I actually chose not to post this schematic on this thread because of the inclusion of these two components (and was too lazy to remove them). Can't remember if I actually did post this schematic, now. [QUOTE=Chryseus;36249718] If your going to use an SCR replace it with a decent SCR, preferably 200A or above. Also that PUT is completely the wrong symbol.[/QUOTE] Just looked on the [url=http://www.ebay.com/sch/i.html?_nkw=200a+thyristor&_sacat=0]Hong Kong Online Marketplace[/url] (i.e. eBay), and damn those 200A thyristors are expensive. Should I get one with a heatsink? Ugh, that's going to be a bitch to mount, heatsink or nay...probably easier than mounting the heatsinks for four MOSFETs, though. [url=http://eagle.undo.it:8083/img/cdwelder_13_kp200a.png]This[/url] appears to be the specs for the KP200A. Also, thanks.
I got my pcb shipped today. This is my complete weather balloon tracking system with transmitter, buzzer, and inside and outside temperature sensors :) [IMG]http://kj6wmj.torba.us/img/ardutrack_nano_front.jpg[/IMG]
[QUOTE=Night-Eagle;36255926]How did you get 40A? Seriously don't know.[/quote] My bad I screwed up the calculation, it's actually around 1.5A, that will cause a 0.75V drop across R12 thus turning on Q4, while this works to limit the charge current the way it's implemented is not very good requiring you to use high power base components. Good old Ohm's Law at work V = IR [QUOTE=Night-Eagle;36255926]There was concern that due to the amount of current being pulled through the thyristor, it might blow if it spent too long in its non-fully conducting state.[/quote] I see, however in practice I don't think this is an issue since the current will decay fast enough to avoid any excess power dissipation. [QUOTE=Night-Eagle;36255926]I realized R10 served no purpose and did not include it in the build. D3 was suggested just for the reason you mentioned. I actually chose not to post this schematic on this thread because of the inclusion of these two components (and was too lazy to remove them). Can't remember if I actually did post this schematic, now. Just looked on the [url=http://www.ebay.com/sch/i.html?_nkw=200a+thyristor&_sacat=0]Hong Kong Online Marketplace[/url] (i.e. eBay), and damn those 200A thyristors are expensive. Should I get one with a heatsink? Ugh, that's going to be a bitch to mount, heatsink or nay...probably easier than mounting the heatsinks for four MOSFETs, though. [url=http://eagle.undo.it:8083/img/cdwelder_13_kp200a.png]This[/url] appears to be the specs for the KP200A. Also, thanks.[/QUOTE] That thyristor will work quite nicely, however you will need to redesign the trigger circuit to provide a high current gate pulse, dumping for example a 1000uF cap into it should do the job.
Awesome - you scared me when you said 40A. Thought I really messed something up. [QUOTE=Chryseus;36257632] That thyristor will work quite nicely, however you will need to redesign the trigger circuit to provide a high current gate pulse, dumping for example a 1000uF cap into it should do the job.[/QUOTE] How do you figure the values for C2 and R7? I've been reading [url=http://www.littelfuse.com/data/en/Application_Notes/01_trigg.pdf]this, page 1-36+[/url] and I seem to be missing some fundamental knowledge here. Does this (SCR trigger circuit in "my"/jimkeith's schematic) simplify to a RC circuit to find how much current is being put through the gate? And [url=http://eagle.undo.it:8083/img/cdwelder_13_kp200a.png]this[/url] means that the KP200A must be triggered with at least 3V & 150A on the gate, right?
Anyone interested in a digitalized build document from the 80's about a little FM oscillator, that has been said to be the most stable design? Very easy to make and a fun introduction to RF! Just let me know if interested!
[QUOTE=Chezburger;36265507]Anyone interested in a digitalized build document from the 80's about a little FM oscillator, that has been said to be the most stable design? Very easy to make and a fun introduction to RF! Just let me know if interested![/QUOTE] Sounds interesting! Wouldn't be able to mess around with it just yet but when I get this project done, sure
So I got one of these geiger counters as a gift. [url]http://mightyohm.com/blog/products/geiger-counter/[/url] I have ideas, but most of them I can't do until I get used to opengl and opencv. Any ideas people been dying to try with a gieger counter and arn't that hard to implement?
[QUOTE=Trumple;36265956]Sounds interesting! Wouldn't be able to mess around with it just yet but when I get this project done, sure[/QUOTE] Allow me while I translate the document from Dutch to English, and finish up some kind of PCB design! I'll let you know when it's done!
[QUOTE=Chezburger;36269767]Allow me while I translate the document from Dutch to English, and finish up some kind of PCB design! I'll let you know when it's done![/QUOTE] Could you send me a copy to? Mag in het Nederlands zijn hoor.
0lenny0, LoneWolf_Recon and Trumple, I sent you the file. Let me know if you have it ;)
[QUOTE=Chezburger;36270938]0lenny0, LoneWolf_Recon and Trumple, I sent you the file. Let me know if you have it ;)[/QUOTE] Got it! This looks really interesting, thanks very much
I've ordered 3 meters of 40-way rainbow cable today
[QUOTE=DrLuke;36271199]I've ordered 3 meters of 40-way rainbow cable today[/QUOTE] Bah, should've gotten [url]http://www.dealextreme.com/p/40-pin-ribbon-wire-cable-rainbow-10-meters-124485?item=24[/url]
[QUOTE=Chezburger;36270938]0lenny0, LoneWolf_Recon and Trumple, I sent you the file. Let me know if you have it ;)[/QUOTE] Just downloaded the document, thanks!
[QUOTE=chipset;36271369]Bah, should've gotten [url]http://www.dealextreme.com/p/40-pin-ribbon-wire-cable-rainbow-10-meters-124485?item=24[/url][/QUOTE] I eyeballed with it, but 30\$ was too much for me, so I bought only 3 meters for 9\$.