Go Back   DIY Chatroom Home Improvement Forum > Home Improvement > Electrical

CLICK HERE AND JOIN OUR COMMUNITY TODAY...IT'S FREE!

Reply
 
Thread Tools Search this Thread Display Modes
Old 05-16-2013, 08:26 PM   #16
Common Sense Common Guy
 
Join Date: Oct 2012
Location: So. IL near STL
Posts: 96
Rewards Points: 75
Default

Options for powering DC Xenon lamphouse...


Quote:
Originally Posted by mpoulton View Post
Please post your results! I'm very interested in how this works out.
Well, after purchasing one of those very welders, I hooked it up and it works... sort-of. The bulb strikes and maintains an arc, although it is flickering a bit. The lamp house's ammeter shows only 10A, no matter what the pot setting on the welder. The setting on the welder doesn't seem to affect the arc at all.

My clamp ammeter is on the fritz, so I need to obtain a new one before I can see whether the lamp house's meter is good or not. That would also allow me to test dead-short current on the welder, to see if it truly passes the right amount of current or not.

Any other suggestions?

EDIT: Photo of the arc image in the viewing window:
Attached Thumbnails
Options for powering DC Xenon lamphouse...-camerazoom-20130516195042980.jpg  


Last edited by flasherz; 05-16-2013 at 10:15 PM.
flasherz is offline   Reply With Quote
The Following User Says Thank You to flasherz For This Useful Post:
mpoulton (05-17-2013)
Old 05-16-2013, 10:38 PM   #17
Common Sense Common Guy
 
Join Date: Oct 2012
Location: So. IL near STL
Posts: 96
Rewards Points: 75
Default

Options for powering DC Xenon lamphouse...


One more update tonight... I got it working at a full brightness, it seems that the welder did not like being attached to a short 14 AWG extension cord. I swapped it out for a 12 AWG cord and I was able to adjust the output to 45A. The lamp arc still flared/flickered more than I would like, but I'll see what it looks like when I get it reconnected to the projector.
flasherz is offline   Reply With Quote
Old 05-16-2013, 11:07 PM   #18
Idiot Emeritus
 
Join Date: Mar 2008
Location: Fernley, Nevada (near Reno)
Posts: 1,618
Rewards Points: 1,030
Default

Options for powering DC Xenon lamphouse...


Capacitors across the DC lines might help with the flickering, but a choke would work even better.
micromind is offline   Reply With Quote
Old 05-16-2013, 11:41 PM   #19
Common Sense Common Guy
 
Join Date: Oct 2012
Location: So. IL near STL
Posts: 96
Rewards Points: 75
Default

Options for powering DC Xenon lamphouse...


Quote:
Originally Posted by micromind View Post
Capacitors across the DC lines might help with the flickering, but a choke would work even better.
Any thoughts on the right inductance / resistance for this application?
flasherz is offline   Reply With Quote
Old 05-17-2013, 12:18 AM   #20
Semi-Pro Electro-Geek
 
Join Date: Jul 2009
Location: Arizona, USA
Posts: 2,557
Rewards Points: 2,010
Default

Options for powering DC Xenon lamphouse...


Cool! Glad to hear it mostly works. If you are observing visible flicker, it's probably the result of control loop instability rather than actual ripple on the power supply. The ripple should be 120Hz, which is invisible, as well as tens of kHz, which is even more invisible. The control loop instability may be the result of the arc characteristics not exactly matching the designers' expectations for a welding arc. The operating voltage is probably significantly higher than most welding arcs, which are usually in the 10-20V range IIRC. Although the welder should be capable of operating at rated current in the 20-30V range, it is not really designed for it. The ideal solution would be to do a little reverse engineering on the welder and change the loop control parameters. That probably comes down to a few resistors and capacitors, some of which may already be variable. This would require quite a bit of hacking to figure out though. Adding a large choke in series with the lamp may help solve the problem by smoothing out the nonlinearity of the lamp's high frequency I-V response. As much inductance as you can muster would probably work well. Hundreds of millihenries should be the right ballpark. #8 wire wrapped on a core from a microwave oven transformer would probably be a good bet. I would NOT use a pi filter without knowing the loop control parameters of the power supply, since the frequency response of the pi filter can further destabilize the power supply under some circumstances.
__________________
I am a lawyer, but not your lawyer. And who cares anyways? We're here to talk construction. This is DIY advice, not legal advice.
mpoulton is offline   Reply With Quote
The Following User Says Thank You to mpoulton For This Useful Post:
flasherz (05-17-2013)
Old 05-17-2013, 01:17 AM   #21
Semi-Pro Electro-Geek
 
Join Date: Jul 2009
Location: Arizona, USA
Posts: 2,557
Rewards Points: 2,010
Default

Options for powering DC Xenon lamphouse...


I couldn't help doing some hacking. I took apart my welder. It's based on a UC3846 SMPS controller, the middle 16-pin IC at the top edge of the back of the main board. All the passive components that set its operating parameters are right near it. It is switching at 74kHz, which is higher than I expected for a power supply this big. It is a current-mode controller, operating in a symmetrical drive topology (probably push-pull; I didn't look at that part) using deadtime to control the power output. It has pulse by pulse current limiting as well as some kind of foldback protection which essentially kills the output if the load impedance drops below a certain limit. The open circuit voltage is much higher than the manufacturer says! It's 70-80V RMS, over 100V peak. There is a small ferrite output inductor, but that's all. It does not significantly smooth the 74KHz output ripple, and is probably only there to filter RF noise. There is a lot of filter capacitance on the input, and the 120Hz ripple seemed minimal but I was testing at relatively low power.

I'd try a huge inductor in series with the lamp first. I suspect that will improve the situation a lot. If it does not work, you may need to consider messing with the passive components controlling the frequency response of the error amp and current sense amp in the UC3846, and whatever they've done with the pin 1 current limit control, which is probably responsible for the foldback behavior and might be messing with your lamp operation.
__________________
I am a lawyer, but not your lawyer. And who cares anyways? We're here to talk construction. This is DIY advice, not legal advice.
mpoulton is offline   Reply With Quote
The Following User Says Thank You to mpoulton For This Useful Post:
flasherz (05-17-2013)
Old 05-17-2013, 02:05 AM   #22
Semi-Pro Electro-Geek
 
Join Date: Jul 2009
Location: Arizona, USA
Posts: 2,557
Rewards Points: 2,010
Default

Options for powering DC Xenon lamphouse...


And another thing: This SMPS topology is subject to "subharmonic oscillation", which is a form of chaotic instability at any duty cycle above 50%. It can be compensated for, but it's hard to do and the compensation depends on the load characteristics. You'll know you have subharmonic oscillation if the power supply makes a hissing noise, and the lamp may flicker randomly. I've run into this a few times on laser power supplies and have never been able to properly solve it when it happens. It's like electronic black magic, as far as I'm concerned. This welder might be designed to ignore the problem since it might not matter for welding. If your lamp demands greater than 50% duty cycle from the UC3846, that chaotic behavior may be causing your problem.
__________________
I am a lawyer, but not your lawyer. And who cares anyways? We're here to talk construction. This is DIY advice, not legal advice.
mpoulton is offline   Reply With Quote
The Following User Says Thank You to mpoulton For This Useful Post:
flasherz (05-17-2013)
Old 05-17-2013, 10:22 AM   #23
Common Sense Common Guy
 
Join Date: Oct 2012
Location: So. IL near STL
Posts: 96
Rewards Points: 75
Default

Options for powering DC Xenon lamphouse...


I'm having a hard time in various searches to find a large enough choke with enough inductance - have any supplier suggestions, or am I going to end up rolling my own here? I can find plenty of 50A, but usually only in the range of 1-10 mH or less.
flasherz is offline   Reply With Quote
Old 05-17-2013, 11:46 AM   #24
Semi-Pro Electro-Geek
 
Join Date: Jul 2009
Location: Arizona, USA
Posts: 2,557
Rewards Points: 2,010
Default

Options for powering DC Xenon lamphouse...


You'll have to roll your own for something this large. The core from a microwave oven transformer is great for this. Just hacksaw the windings and pry them out with a screwdriver, then wind your own with #8 wire. Disassembling a cheap AC welder should yield something usable that's even bigger, and you could probably just re-use the original secondary winding as the inductor.
__________________
I am a lawyer, but not your lawyer. And who cares anyways? We're here to talk construction. This is DIY advice, not legal advice.
mpoulton is offline   Reply With Quote
Old 05-17-2013, 06:59 PM   #25
Idiot Emeritus
 
Join Date: Mar 2008
Location: Fernley, Nevada (near Reno)
Posts: 1,618
Rewards Points: 1,030
Default

Options for powering DC Xenon lamphouse...


If nothing else, use the choke from the original rectifier.

micromind is offline   Reply With Quote
The Following User Says Thank You to micromind For This Useful Post:
mpoulton (05-17-2013)
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Inexpensive Siding options that that aren't Vinyl or Metal GumbyMan Building & Construction 15 01-06-2013 11:01 AM
Recommendations for heating and cooling options for my home? daluu HVAC 2 12-23-2012 01:31 PM
Shower options Chilidog Tiling, ceramics, marble 3 06-06-2012 04:28 PM
Maytag dishwasher has 5 cycles and three options. only one cycle works and no options wwejj0926 Electrical 1 04-22-2012 05:54 PM
Replacing undercabinet halogen bulb with xenon bulb gabbie14 Electrical 10 09-26-2011 10:46 PM




Top of Page | View New Posts

Copyright © 2003-2014 Escalate Media. All Rights Reserved.