laser PSU flyback problems

[GerryS]

Hi,
My name is Gerry from the netherlands and I have an issue with my diy 40w co2 laser which I cant seem to solve. I hope someone in this community can give me a hand.
I build a 40w laser from the kit bought at buildyourcnc.com, known as the Blacktooth.
The problems started when I tried to test my psu for the first time, without having the tube connected. Capacitors were blown with a nice puff, some smoke and a mumbling OOPS... Figured out the psu was destroyed so I bought a cheap 40w psu from China. It worked perfectly for a few months. Suddenly in the middle of cutting some MDF the laser stopped. No sound, no sparks, it just stopped.

I opened up the box (after some time) but nothing looks fried. After some research I found out it should be the flyback. So I ordered a (few) new flybacks, installed on of them and the laser worked again. For about ten minutes before it died again. In the mean time I ordered a PSU from Lightobject. It lasted for about 1 second. I checked al my wiring, connections to and from the tube, installed a new flyback in the lightobject psu, managed to cut for 10 minutes and then it died again. As you can imangine I'm getting pretty frustrated by now.

The last time I destroyed a flyback I was monitoring the Amp output. I set it to approx 14mA. When I started it was at 14.3 but while i was cutting it slowly declined to 13.9 / 13.8 mA in about 10 minutes. When it was supposed to fire to start the next part of what I was cutting (the headlight of a volkswagen beetle), the laser was dead. Probably the flyback in the psu again.

I connected the PSU the way it is suggested in the pdf on this site.
WP to GRN;
5V,IN,GND to POT;
TL to breakoutboard;
breakoutboard 5v to 5v on laser PSU (this is the way it is connected in the diagram from buildyourcnc.com but I'm not sure this is correct since it doens't seem logical).


Any thoughts on what may cause the flyback to be fried time after time are more than welcome. I've been searching online for days but can't seem to find anyone with an answer.

[Techgraphix]

The flyback-circuit consists of a few components, normaly a diode and a relative small capacitor. I think you meen the capacitor?

As you replaced the supply with a new one, that also lived very short, and your wiring is OK, i suspect there is something seriously wrong with the tube itself.
Repairing high-voltage parts is not always satisfying: not seldom one component is damaged that much that it, obvious, measures correct but still isn't OK, think of a shortcircuit between a few windings in the high-voltage transformer. Only with dedicated equipment (in case of a transformer: a high-voltage squarewave-oscillator and a oscilloscope) you can proof that there is something wrong with it.
But that is not the case with youre problem as you swapped the entire supply..
BTW this is not a very common error with CO-2 lasertubes, at least, i did not read about it before... I did some repairs in HV from TV's, Microwave Ovens, and radartransmitters.

Kees
(from the Netherlands too )

[GerryS]

-font-b-High-b-font-font-b-Voltage-b-font-40w-font-b-Flyback-b.jpg (76.49 KiB) Viewed 3385 times

Hi KEes,

Thanks for the suggestion. I got a spare tube so I started installing it yesterday evening. Didn't manage to finish it so going to finish it today and hopefully that will solve the problem. I'll let you know.

I attached a picture of the component I replaced. Since it is completely sealed I can't look inside. I was wondering what is the purpose of the metal u-shaped thing on the black part of the flyback

Never suspected it could be possible that a faulty tube still could fire a laser beam.

I'm using the laser to cut wooden letters and figures for my webshop (mostly childrens bedrooms). Since the laser is not working properly, I'm forced to using the old trustworthy cnc router again. Quality of those cuts are acceptable but still ha ve a lot of sanding to do after they are cut. Takes a lot of time as you can imagine.

[GerryS]

I've been testing the new tube today. I managed not to blow up another flyback transformer todat so I guess we can call this a good day. I cut three figures today and while I was cutting something else came to my attention. The beam inside the tube was flickering and the mA level was fluctuating plus or minus 4 mA and sometimes boosted way over 20 mA. I'm not an expert but I would say that is not how it is supposed to work. Can this have something to do with poor grounding?

Could I solve this by connecting the wire coming from the front of the laser tube directly to the casing of the PSU? Since my laser is in the shed and I have doubts about the way the wires are grounded over there, I connected the laser to a socket in the kitchen which is properly grounded (I had this checked and fixed a few years ago after I was getting shocks from the dishwasher). Same result so I can rule out bad grounding in the shed.
Could the new tube be bad of should I start by rechecking my connection from the PSU to the tube?

[Techgraphix]

It's quite normal that the plasmabeam inside the tube "dances" around the cathode (the side where the beam leaves te tube)
Fluctuations in the current, while cutting is not common, a few tenths of a mA is OK but 4mA...
It can be that the corner-power is a lot less but as you work with a potentiometer, this is not regulated by a controller.
It can be that the potentiometer is dirty, so the voltage on the HV-powersupply-IN will fluctuate and so the output of this supply.
I think just grounding is sufficient but what is of upmost importance is that everything, low level, is shielded and properly grounded (shield connected to ground only one side and all groundwires should go separate to one common groundpoint..
Interference can cause a lot of unpredictable problems as well as bad contacts (oxidized or loosen screws..)

You should not connect the returnwire (kathode) to the case of the supply. Normaly there is a relative thin wire going in the supply. |It can be that, in there, the signal is used for currentsensing. When you connect the tubes Cathode to the case you will bypass this circuit (if it exists..)

The metal U-shape is to keep the two feroxcube parts together tight. It looks one piece but it will most probably be two u-shapes pressed in the highvoltage-transformer.

Kees

[GerryS]

Thanks again Kees.

I recently replaced the pot meter. It was the ultra cheap version. you mentioned a dirty pot meter could be a reason for unstable HV input but could a cheap (and possibly less accurate) pot meter be the reason for the bouncing of the Amps?

The plasma beam is not dancing in the tube. This only happens when I set the pot meter very low, but around 8 mA it is a stable beam, though it sometimes becomes brighter or dimmer. This is not a problem when I'm cutting but I want to start engraving too and that's when a steady output becomes more important.

I'll go over all my connections once again (you can never check to many times).

[Techgraphix]

About the potentiometer: With old radios, for instance, you can often hear a crackling sound when you turn the volume and even sometimes you have to turn it to get the sound back. This volume-pot is nothing more than a resistor-devider. The output is a ratio (position of the knob) of the input. Same as with your laser: the input is 5V and the output is a ratio of that. When the pot. cracks, it means the ratio is gone and the output will, most likely be 0V for a short moment. If 0V is no power-output and +5V is max power output, you can imagine that this cracking will affect the output power.
But there can be more issues with a potentiometer. Internal is a (carbon) runway connecter to the solderlugs. When the connection of the runway to the solderlug is corroded/loose/broken/whatever, the output will drop to zero or raise to +5V whichever side is bad.

When you check your wiring, it is best to take the wire from the connector and place it back again, one wire at the time so you won't be mistaken.
Sometimes people solder an end and put that in a screwterminal.. WRONG! crimp cable-end sleeves instead or put the bare clean wire in the terminal.
A fixed wire does not mean you have a solid electrical connection.. That's why you have to reconnect it (take it from the connector and place it back)

CO2 lasertubes have a treshold. Mostly this is around 10-15% of the max current. Under this value you can see some plasma but there is no output at all.
Can you monitor the |IN-voltage? (from the potentiometer)
If this is deadstable, then you have to go the the load-side.
And ofcourse it is still possible that your repaired HV-supply has some damage left from former defects..

Kees

[GerryS]

well the POT is new so I guess that shouldnt cause the problem. But I might have done somthing that may be the cause. I put some solder on the HV wire to make a better connection. I'll remove that piece of wire and test again. At least, i'll test as soon as my new waterpomp has arrived (hopefully today) cause it died yesterday. Bad luck comes in threes i guess. Well if that is the case, my luck should change for the better.

By the way, I bought the UC400ETH controller with the UCCNC software (as a replacement for MACH3). I'm going to try it out soon and will let you know what my experiences are. The UC400ETH is similar to the Ethernet SmoothStepper but cheaper. UCCNC is like MACH3 but cheaper (I'm from Holland remember! I like cheap stuff )

[Techgraphix]

(I'm from Holland remember! I like cheap stuff )

Alleen als je uit Zeeland komt...
(Only when you come from Zeeland, a province in Holland, for the non-dutch)

Kees

[GerryS]

Well I'm from Scheveningen so close enough to Zeeland to be infected

Cleaned up the soldered connection yesterday and now I get higher mAmps but still bouncing. Going to check the pot meter today.

Did a test with the new UC400ETH and after some tuning it works perfectly. Now I can use my laptop to drive te machine. This is a great improvement since I was using some old desktop computers in a dirty and moist shed. Computers dont like that so I used up three computers in two years. Now I can drive the machine from the warm and dry livingroom and monitor it by using a webcam.

[GerryS]

Well the problem is solved. I installed the new high precision pot today and the output stopped bouncing. I guess it is worth spending a bit of money on essential parts. Timing is perfect cause I had quite a few orders this weekend so I started lasering the MDF. Only problem I have now is complaining neighbors. The weather is great so every one is sitting in the garden enjoying the smell of vaporized MDF.
guess I call it a day and continue when everyone is out working

Thanks Kees for all the help!