Making a Diode Bridge, Help!!

admin

im building a rectifier for the buzz box. ive got all the stuff to make it DC exept the diodes which is where i need help.  i have the alum plate for the heat sink but i want to know which of these diodes would work and how many i need.  the welder is capable of 200 amps.http://cgi.ebay.com/New-1N4045-Recti...item2ea99bf898http://cgi.ebay.com/International-Re...item518daf8a9dRealgear 140 MIGMore DeWalt tools that I prolly needDIY 160 amp 40 OCV MOT Stick Welder
Reply:I like the second post better - 150 amps at 200 volts... but $8 to ship one $15 diode = get real!Anyway, if you use those diodes, you will need 8 of them.You will need to parallel two in each arm of the 4-arm full wave bridge.I'd recommend a Peak Reverse Voltage (PRV) of at least 200 volts.The first diodes were only rated at 100 volts - dicey in my opinion.Rick V
Reply:Originally Posted by Rick VAnyway, if you use those diodes, you will need 8 of them.You will need to parallel two in each arm of the 4-arm full wave bridge.
Reply:The 200V peak diodes should work just fine.  100V is indeed a bit dicey.  With an AC 80V RMS OCV, the peak to peak voltage would certainly go over 100V.  200V should be safe.You should NOT need to double up those diodes.  Four 150A diodes in a full wave bridge are capable of carrying 300A.  In any case, diodes do not parallel well without additional elements.
Reply:Agree, also. If one parralel diode goes open all curent will go through the other and pop it so the rig is only as strong as it weakest link anyway. Also can get into heating problems if the heat sink does not remove heat evenly one heats more than the other and breaks down faster.I never had much luck with a diode bridge on an AC buzz box, just dropped the voltage too much and made it hard to start and maintain an arc, arc length got really tight. Did get around it some by going with 240v though the 208v tap raising the voltage ~10%. Your on your own risk there...Suppose it would be possible to up the voltage using capacitors with the bridge but that would take some expensive equipment and likely would just be better and cheeper to get a DC welder. Also a problem with a cascade voltage multiplier type bridge is the curent goes down.I blew up a 400 amp set of diodes nearly the size of hockey pucks messing around with 250 amp buzz box (likely did not have enough cooling), that along with a couple shocks was pretty much enough to convince me there were better and safer ways to get DC.The rating of the diodes is only good if the temperature can be kept within operating specs.Last edited by Timberwolf; 04-28-2010 at 11:30 AM.Lincoln PT185 TIGLincoln 175 MIGLincoln 240 amp AC StickLind Needle Arc plasma welderPlasma cutter, soon???
Reply:You might want to look at my old post, "Converting Stick Welder from AC to Full Wave DC".I keep hearing the argument about not paralleling diodes.Why not?The manufacturers do it all the time.Example 1I used to have a Canadian Tire MIG/Flux Core Wire Feed Welder (Campbell Hausfeld WG206000AV) 115 Volt input with 70 Amp DC output.  It used a bridge of 8 diodes - like a conventional bridge 4-rectifier set up but with two diodes in parallel in place of each diode of the bridge.Example 2My Lincoln MIG-Pak 15 (SP-175T) uses a bridge of 16 diodes! - apparently in pairs of pairs!  How do you like that?  Double-double! So, commercial welding equipment is using diodes in parallel - and no balancing resistors.Rick V
Reply:Maybe so, also they spend a few $$$ on electrical engineers to design and test the equipment. If it is an inverter with some sort of chopper powersupply then those diodes may be phased too, would have to look at the circuit board to see.To me it is a question of if the goal is to weld or mess around with science fair experiments. If you want to weld with DC a home built rig hooked to a buzz box is a poor and frustrating way to get there. Got that T shirt and thats even with a couple years education in electronics. Sure it can be done, but the parts and effort required to do it right and make it work are going to be a greater cost than what it would be just to get the right tool for the job.  If its just about experimenting and messing around ,then its all good, we all have hobbies.Lincoln PT185 TIGLincoln 175 MIGLincoln 240 amp AC StickLind Needle Arc plasma welderPlasma cutter, soon???
Reply:I'm with Timberwolf.To convert a used ($100 on Craig List) 225 amp transformer type buzz box to say a 150 amp DC transformer type ($250 on Craig list), you will need:- 4 or more power diodes- 1 big inductor (arc stabilzer)- heat sinks and mounting hardware- lots of timeAt the end of the day, it may be more efficient and satisfying to simply sell what you got and invest in a used DC welder that you know works.  Think about resale: you can get your money back selling a brand name used DC welder; what can you expect to get back from a home-built unit?Timberwolf said it best...Sure it can be done, but the parts and effort required to do it right and make it work are going to be a greater cost than what it would be just to get the right tool for the job.
Reply:Originally Posted by Timberwolfalso they spend a few $$$ on electrical engineers to design and test the equipment.
Reply:Originally Posted by rlitmanThe 200V peak diodes should work just fine.  100V is indeed a bit dicey.  With an AC 80V RMS OCV, the peak to peak voltage would certainly go over 100V.  200V should be safe.You should NOT need to double up those diodes.  Four 150A diodes in a full wave bridge are capable of carrying 300A.  In any case, diodes do not parallel well without additional elements.
Reply:Not a bad write up here on a mig set up. See some good stuff and some questionable things.http://blinkenbyte.org/welder_conver...onversion.htmlIt was a few years ago when I messed with the AC/DC rig, before I had a digital camera else I would post some pics. Now when ever I start one of these brain wave projects I take pictures for later reflection and explaination. If you have four 150A diodes connected in a bridge, they are NOT capable of handling 300A. Only two diodes are conducting at a time, one taking the positive half, and the other the negative. The current capacities don't add.
Reply:Originally Posted by Rick VI keep hearing the argument about not paralleling diodes.Why not?The manufacturers do it all the time.
Reply:Actually the amp rating of the diodes is based on the heating of the die element in the case.The individual diodes in a bridge only run half of the time.