Explorer II 100W Power supply problem

[Constant Laubscher]

I was engraving at 100% power and 38% speed when all of a sudden it was loosing laser power from 100% to about 30%. I could also hear that all the fans was running at half speed.
I measured the power to the tube and find that when I run the laser it only measures 31V and not the 46-49 Volt.
I disconnected the tube completely from any power and found that actually only one of the five power supplies was working.
Four had blown fuses and even if I replaced the fuses it still popped them!
Is there anyone out there that might have had the same problem in the past?

The machine is just out of warranty and I am looking for some ideas on what might have caused it to happened.:mad::confused:

[Richard Rumancik]

Five power supplies? That is an unusual configuration. I am assuming that they are interconnected connected in some kind of series/parallel arrangement. Hard to visualize this with 5 supplies. If you can get some kind of wiring diagram from GCC that would help. If you post it maybe one of our resident electrical types can help. Usually GCC will supply a general wiring diagram. (You probably won't be able to get any schematics for the supplies themselves though. Good chance GCC buys them.)

I'd guess that there is a fault somewhere that is overloading the supplies but it's pretty hard to speculate. I doubt if 4 supplies would go bad at the same time, but maybe a single short is causing 4 of them to blow fuses.

If they are parallel connected then perhaps one supply went bad taking down the rest. Maybe you can isolate a bad one by trial and error. I'd try to put on a dummy load however like a resistance heater when troubleshooting.

One thing you could do is trace all the harnessing from the power supplies to the laser load and see if you can detect any wiring defects (pinched wires, melted insulation, etc.)

When the fuses blew the second time, did you have the laser tube load on them?

[Constant Laubscher]

No I removed all power connections.
I spoke to GCC but they don't fix them they just replace the power supplies.( so much for a technician) a plug and play exercise that would cost me $612 per power supply.
A tube repair if that would be the problem would cost between $6000 - $7000.
I shall do more investigating today and see if I can find the problem.
When your machine is out of warranty, you pretty much on your own or you can pay big money for there lack of skills to repair components.

[Constant Laubscher]

I have now ordered the components on the power supplies that popped.
I am still looking for any suggestions on what may have caused the problem in the first place? I hope that will be all that needs to be replaced.
Any advice?

[Kim Vellore]

Constant,
Could you post some specs., model number or pictures of the power supply. I am assuming it is a bank of switch mode power supply modules connected in parallel. You could disconnect the parallel connection replace the fuse, turn it ON and check the fuse to see how many fails and replace the failed module. You could try to identify and replace components in the module but usually there will be more than a few components that will fail.
The SMPS fails for many reasons and will not be related to specific laser manufacturer.

Kim

[Dan Hintz]

I agree with everything Kim just said...

[AL Ursich]

The Full Wave Rectifier is my first item to check. They go bad by opening or shorting. When I was repairing Power Supplies for Sony 1 in 10 shorted. That could account for the 4 blown fuses if they are hooked in parallel. One shorted to ground will blow the 4 at the same time.

Then look for bulged Electrolytic Capacitors.

Photographs of the inside of the Power Supply would be good... Depending on the amperage of the power supply there could be 4 large semiconductors screwed into a heat sink with braided leads that could be bad.

I really miss fixing stuff....

AL

[Constant Laubscher]

The same component is blown on all four ( thermistor).
I was waiting for the components to arrive, replace them and then see what else has failed. I do not have any access to a oscilloscope but will go with what is visible.
I shall post photo's later today.
Thanks for the advice from everyone .

[Richard Rumancik]

The thermistors on the front end of the supply are usually for limiting inrush current. They start off at high resistance allowing the caps to charge slowly and then go to almost zero resistance after a short time.

So why did 4 of them blow? Seems like all the power supplies saw something they didn't like. Maybe a shorted output causing excessive load? Did you have any power line disturbances on the input end (eg. lightning storms?) Do you have any power conditioning on the front end (surge suppressor)? If so check that the surge suppressor is still healthy. (per indicator light or else discolored surge suppressors) If you don't have one, you really should get one . . . and don't skimp. You want 2500 joules or more rating.

[Bruce Volden]

The Full Wave Rectifier is my first item to check. They go bad by opening or shorting. When I was repairing Power Supplies for Sony 1 in 10 shorted. That could account for the 4 blown fuses if they are hooked in parallel. One shorted to ground will blow the 4 at the same time.

Then look for bulged Electrolytic Capacitors.

Photographs of the inside of the Power Supply would be good... Depending on the amperage of the power supply there could be 4 large semiconductors screwed into a heat sink with braided leads that could be bad.

I really miss fixing stuff....

AL

C'mon Al...you have 3 CW's:D :D

Bruce

[Constant Laubscher]

I am sorry was busy with some projects. Here are some pictures of the power supplies & component that popped.

Kind Regards
Constant

[Richard Rumancik]

Interesting . . .You said the fuses popped when you initially replaced them. So either there was an overload or transient that blew the fuses AND the thermistors. I assume the thermistors are in series with the fuse. But I would have thought the thermistors would fail open-circuit. In which case the fuses would not blow a second time as there would be no input.

If they failed shorted (which is less likely) then they are out of the circuit and not doing their job as inrush-current protection. In this case, the fuse could blow a second time. Maybe it's too late to see if they are shorted (approx. zero ohms). Compare to a good one if you have any new ones left.

Maybe one of our resident technicians can shed some light on this.

[Constant Laubscher]

The parts have just arrived:):D
Same result, I replaced the visible blown components but when I switch the power on they all blow again.( Thermistor & fuse)
I also see on one of the four power supplies that 2 semiconductors are blown.
All the tests I have done so far I have tested them individually and did not use the machines power source.

http://www.sawmillcreek.org/images/misc/progress.gif

[Constant Laubscher]

Same result, I replaced the thermistor and fuse but they still blow as soon as you switch the power on. There must be something major wrong.

[Constant Laubscher]

Same result, I replaced the visible blown components but when I switch the power on they all blow again.( Thermistor & fuse)
I also see on one of the four power supplies that 2 semiconductors are blown.
All the tests I have done so far I have tested them individually and did not use the machines power source.

[AL Ursich]

Bruce,

Yes the 3 CarveWrights do keep me busy fixing them.... :)


I would focus on the semiconductors attached to the heat sinks. Suspect 2 sets are a full wave rectifier and most likely the problem. The other heat sink most likely has the pulse width modulator to adjust the output.

Using a Multi meter in diode check mode would be the way to check the semiconductors.

Close up pictures of the components on the heat sinks would be good.... Did you look at the underside of the board yet for burnt copper traces.

Still think you will find a semiconductor on the heat sink shorted. Only takes one shorted to side rail you.

Some power supplies have a Crow Bar Circuit too.... A faulty crow bar circuit could do the same thing. Crow Bar will kill the power to protect the device.

It almost looks like there might be 2 power supplies on that board from the components.

AL

[Constant Laubscher]

Can you give me any idea of what the full wave rectifier is?

[AL Ursich]

A full wave rectifier is a set of 4 diodes that have AC voltage on 2 terminals and + Positive and - Negative on the other. It turns AC or Alternating Current (House power) into DC or Direct Current (= Battery power )

http://www.allaboutcircuits.com/vol_3/chpt_3/4.html


http://www.eecs.tufts.edu/~dsculley/tutorial/diodes/diodes3.html


If it were me, I would order and replace every component that is attached to a heat sink or the chunk of aluminum. Be sure to be careful as sometimes the mounting screw could have plastic insulators as part of the screws and a insulating mica thin strip between the metal parts of the semiconductor and the heat sink. There will also be a white paste that is a heat sink compound between the component and the heat sink. That acts as a conductor of the heat from the component to the heat sink. Missing heat sink compound can cause premature failure of the part.

A tip to replacing the components on the heat sink is to carefully cut the legs from the component and un solder each leg one at a time. It is very common to damage the copper pads that the legs are soldered to trying to unsolder 3 or 4 legs at one time as they can be soldered on both sides of the board and pulling will just rip the pad off of the board making the power supply just more scrap.

Taking close up pictures can help with the polarity issues after the parts are removed since you don't have a schematic.

Close up clear pictures of the components on the heat sinks would help or the letters and numbers on each with a description like 3 leg device...

Just had a thought..... The power to the Laser could be AC or Alternating Current..... Rather than DC...... I am thinking just power supplies.... Not Lasers..... Either way... Bet is is components on the heat sink's.
AL

[Constant Laubscher]

Thanks for all the advice. I got some results back from a place who tested all the power supplies and his conclusion was that they are toast.
What caused it is a big concern at this point?

I did talk to a person in Taxes http://www.powerclinicinc.com Brain Hill
and they specialize in Power supplies big and small. He gave me a quote of min $200 - max $300.00 per unit. That is half the price of a new.

One more question, why do they connect 5 units in parallel instead of using one unit to drive the tube?

Thanks again for all your advice it is greatly appreciated.

[Leo Graywacz]

Thanks for all the advice. I got some results back from a place who tested all the power supplies and his conclusion was that they are toast.
What caused it is a big concern at this point?

I did talk to a person in Taxes http://www.powerclinicinc.com Brain Hill
and they specialize in Power supplies big and small. He gave me a quote of min $200 - max $300.00 per unit. That is half the price of a new.

One more question, why do they connect 5 units in parallel instead of using one unit to drive the tube?

Thanks again for all your advice it is greatly appreciated.


They do that to increase the current output. The power supplies would have 5x the output of one PS. Probably cheaper in the long run to do it that way instead of one large current producing power supply.

[Richard Rumancik]

Is there a manufacturer name and model number on the supply? If so, one option is to try and source replacement supplies through a US distributor. I know the small LaserPro's use MeanWell power supplies, and there are some MeanWell distributors in the USA.

If you can track down the source you might be able to get new ones for $300 each. Sometimes the manufacturer can be traced from a UL or CSA file number if it is not written on the supply.

Now if the supply is a "special", then its a different story. GCC may have ordered a supply configured especially for them. In that case it may be more difficult to find an off-the-shelf replacement unless you can determine what is different.

[Dan Hintz]

If I knew the output current and voltage, I would just order supplies from somewhere else. Laser companies always want to make 200%+ profit off of replacement parts... personally, I think 10-20% profit off of parts that failed is more than adequate.

[AL Ursich]

Does the laser manufacturer have a core charge that you must send in the old power supply? If they do? Ask if you can purchase a re manufactured power supply at a reduced price.

That power supply company in Texas looks like a nice place to work. With many companies still using older equipment having access to a repair service like that is an ace in the hole.

AL

[Constant Laubscher]

The output on these are 48V DC.

[Richard Rumancik]

It seems like it is pretty much an Asner "off the shelf" more or less.

Did you find this pdf?

http://www.asnerpower.com.tw/pdf/ALS800Ver7.0.pdf

It explains the part number system. ALS800-48XYZ is the basic unit to which you change XYZ for unique options. So you have to find out if you can easily get the ALS800-48C21 in quantities of 4 or 5.

If it were me I'd contact the TW office and ask them what the availability would be for 5 supplies in North America. I would change all 5 and keep the working one as a spare.

Note that there may be some possiblity of subbing if they don't have an exact match in stock. For example, it would not hurt you if it had a power switch on the front. You would just turn it on and leave it on.

You might be able to use another brand but it may cause headaches in mounting etc. Depends on how much reworking you are comfortable with. Also the form factor needs to be very similar. This is not the most desirable option.

[Richard Rumancik]

Another idea: you might try and see if either of these companies could bring the supplies in for you:

http://www.pma-power.com/index.php

http://www.sopranotech.com/products-power-industrial-power-supplies-als600800.php

Also, it wouldn't hurt to tell Asner what happened to your initial supplies and see if they have a hunch as to what caused it. Also, I'd ask if there is anything that you need to check on the load side before connecting. You sure don't want it to happen again. Maybe GCC can tell you what you should be seeing impedence-wise on the load side and how to take a measurement.

[Constant Laubscher]

I have ordered some power supplies from http://www.trcelectronics.com
Should have them by Wednesday and it has a three year warranty.
I did some testing this weekend with just the one power supply installed and the laser work just fine but only at max 35% power. ( $40,000 35W machine.) At 35 % load equals the output from the power supply 46VDC.
When you up the power to 40% the voltage starts to drop to 42VDC and so on as you increase the power. I shall update the information as soon as the new Power supplies arrive.

Thanks for all the input it helped a lot!

[Richard Rumancik]

Good to hear . . . if the one remaining supply runs okay then it seems there isn't a problem on the load side.

Does the control panel monitor for failed supplies? Let's suppose that you have 5 supplies and one fails. Then you are running at 80% power max.(4/5) but this would only be evident when you select 100% power for a job. So how long before you realize that one supply has popped on this machine? If there is no feedback/monitoring, then this is not the best situation . . .

When you ran with one supply, did the control panel detect a "problem"?

BTW how much did the supplies cost?

Maybe you should consider some power conditioning if you don't have it. The power utility may have a staff engineer that can offer some advice. Somehow I think that maybe you got hit with a line disturbance. Also, during severe lightning storms, I'd unplug the unit entirely if you happen to be around.

[Constant Laubscher]

Cost were $330.00 per unit with a three year warranty.( Mean Well) not the same as factory installed.

The funny thing is that when the tube and all power supplies were disconnected and I ran a diagnostic test the machine did not even pick up that all those components were gone never mind faulty.
The PC board has it's own supply.

[Constant Laubscher]

Finely have my machine up and running!:):D:p

I installed 4 new 1000 W power supplies( $ 330 each ) instead of 5 OEM at $625 each. ( Had to drill new mounting holes )
I think the OEM power supplies were rubbish from the word go!
With only two of these 1000w running I can run the machine at 100% power and double the speed I would normally raster engrave with much better results then before I had all these problems.

I am just happy to be up and running without any more trouble.

[Richard Rumancik]

If you notice such a change in performance then maybe you had one or more supplies blow a long time ago - if they blew one by one it finally got down to a point where it could not deliver enough power for the tube.

Since there seems to be no feedback or monitoring the system doesn't know there is a problem. Some supplies have a "output okay" connector or LED. Too bad GCC didn't incorporate some kind of circuit to do this. It doesn't seem like it would have cost a lot.