"50W" Chinese laser squeal and low output while it does

[James L Barber]

We have what's claimed to be a "50W" Chinese laser, roughly 6 months old. I'm under the impression these are actually 40W tubes that test better than 40W.

A couple of months ago the power supply failed, dead stop. Since the outfit I bought the laser from is an importer only, I bought a replacement power supply from an eBay supplier in China.

Ever since I replaced the power supply, the machine tends to break out into a fairly loud squeal when cutting. When it does, the apparent power drops off, because it often won't cut completely through the 1/8" clear cast acrylic we usually cut with it. During the odd times that it doesn't squeal, we seem to be getting normal power from the power supply and tube.

Any ideas what might be causing the squeal? My *guess* is that something in the power supply is oscillating, or finding some kind of unwanted resonance. I've considered trying a different style of 50W power supply, but don't want to just start throwing money at it without at least some idea it could be successful.

Thanks,
--jim

[John Noell]

I am wondering if it is arcing causing the squeal. You might want to check for any signs of arcing and be sure to keep entire high voltage wire, esp. the insulated junction, as far away from grounded surfaces as possible. I was told to put my entire wire inside silicone tubing as a lot of Chinese high-voltage wire is not really all that well insulated.

[James L Barber]

Hi John,

An arc is always possible, but in this case if there is one I haven't been able to find it. No carbon tracks or marks anywhere along the HV lead, or where it passes through the sheet metal into the tube compartment.
It's a good reminder, though... I'll check it once again with the lights low.

Thanks,
--jim

[Dave Sheldrake]

Shenhui SH350D by any chance?

You need to check the residual current at the output end of the tube, you may well be over powering the tube with a new supply (they rarely arrive ready adjusted) if you have more than 16mA of residual then you are overpowering the tube (a common problem in the 350 when vendors set them high to make up for the lack of power from the shorter tube)

[James L Barber]

Thanks, Dave.

No, I haven't checked the current. I assume the best way to do that would be to set the "power" (PWM duty cycle) to 100% and put a suitable milliamp meter in series with the cathode of the tube.
Yikes... I have the meters, but I'll admit being nervous about breaking that cathode wire. If the connection breaks that would put 16-20kV on the meter circuit. I'll just have to make sure the connections are solid.

So... If the current is too high, how do you adjust it? AFAIK the controller only sets the PWM duty cycle, not any current limiting. Is there normally a pot or something like under the power supply cover?

Thanks,
--jim



Shenhui SH350D by any chance?

You need to check the residual current at the output end of the tube, you may well be over powering the tube with a new supply (they rarely arrive ready adjusted) if you have more than 16mA of residual then you are overpowering the tube (a common problem in the 350 when vendors set them high to make up for the lack of power from the shorter tube)

[Dave Sheldrake]

There is on some supplies Jim, usually a tiny screw hole, but an easier way is to set the maximum as a percentage in the machines PAD, the pad controls maximum %, the software controls the maximum % of that %.

I often use a simple 5 buck meter for testing residuals, they are usually fine (just don't set it to volts;))

Not all 350's are set to run PWM, some are set on the 0-5V circuit and run analog.

Attach meter, press laser button on the control pad, read off stable current (it will flick up high to start with as the tube fires :))

[James L Barber]

I'll give that a try. I think I have a 50mA panel meter in a box somewhere, I should mount it in the sheet metal on the back of the case and leave it in circuit.
I've attached a photo of the machine data plate. The company isn't one I'm familiar with, but then that applies to the great majority of Chinese manufacturers.

Thanks,
--jim

[Dave Sheldrake]

Made by the same company that cloned the redsail X700 , It's a Kehui, (KH) they clone many of the mainstream companies units and in general are really good value (the tubes are rubbish usually) given who made it I'd put serious $$$ on the fact the PSU is pumping way too much power. seeing how the 500 x 300 machine has a limit on width and a 40 watt genuine tube is 850mm long (a 50 watt is 1,000mm long) my first stop would be the PSU current setting :)

Once sorted out they are pretty useful things, poor build quality (rough edges and pidgeon poo welding) but functional all he same :)

[James L Barber]

Dave, it looks like I owe you a beer.

I haven't actually measured the current yet, (can't find the panel meter and don't want to risk my good Flukes) but I loaded a piece of scrap and started turning down the power down on the keypad 10% at a time. As I started to turn it down, the squeal became less and less apparent. By the time I got down to 40% (!!!) it had completely disappeared. The frosting on the cake is I'll estimate the cut line is only half as wide as it was going through 1/8" clear cast acrylic.

I reset the machine and started a production job. It takes about 25 minutes to scan/cut, so I thought I'd post early results here. So far, so good!

--jim

[Dave Sheldrake]

Seen it a lot of times, China will often run a 40 watt tube @ 21 -23 mA to get more output...it's seriously damages the tubes though and will take magnitudes of order off of the tube life though :)

[James L Barber]

Lesson learned, thanks.
At the minimum, I need to install a meter and tweak a few jobs... It doesn't appear that the "pad" processor re-scales against the preset Maxpower setting, it just limits the output to that value. I ran a couple of jobs last night and the only thing that appears to have changed is the 100% power cuts are now limited to 50%. (I went back to 50% for the Maxpower setting)

I'm not thrilled with the idea that all the jobs will need to be re-tweaked if I change a power supply. I wonder if it's possible to buy "matched" tubes and power supplies? If so I'd consider upgrading the machine to 60W, since 40W is the absolute minimum required for at least some of the work and leaves no margin. It would also be interesting to know if any of the available control pads actually scale the job settings against the overall Maxpower, or if they all behave like the one I have?

Questions, questions... ;-) Our previous machine (Rabbit 6040) didn't catch my interest in this manner, it just worked.

Thanks again for your help.

--jim

[Dave Sheldrake]

No problem,

Yup, I'd suggest an EFR CL1200 tube and matched supply from EFR, good quality beam profile and stable output ;)

[James L Barber]

Appears they're available via Alibaba, if nothing else.
Thanks for the suggestion.

--jim

[Bill George]

Appears they're available via Alibaba, if nothing else.
Thanks for the suggestion.

--jim
More power, longer tube = longer wider case.

[James L Barber]

Hi Bill,
Yes, thanks for pointing that out. The case on my machine has a cut-out for an extension box, which I can buy for $40 or so plus shipping. They look funny with the box mounted but they'll take a 1200mm tube after the "conversion". I've already replaced the 18mm lens with a ZnSe unit, supposedly everything else but the tube and power supply will work. I just need to figure out if it's worth the trouble... When I bought the "50W" machine I had no idea they weren't really 50W. 40W will work, but only just.

Thanks,
--jim

[James L Barber]

Thought I'd post an update. I decided not to mount a meter in the laser cabinet; more vibration than I wanted to subject the tube to. What I did do is install a 1/4" shorting-type switching phone jack in the tube extension cover plate, and wired the cathode of the tube through it. I then used the laser and a CAD-drawn template to mount a 50mA meter in a plastic box, and put a cable on it with a 1/4" stereo plug. A safety diode was installed across the jack contacts in case the meter circuit fails and it all works fine.

What might be more interesting is that even with the power dialed down to where the current reads 10 - 16mA, the squeal is still *mostly* (not always) there. I'm thinking something is oscillating or resonating in the tube power supply, so I'll probably change to a more expensive, "guaranteed stable" 40W unit, since it is is after all pretty much a 40W tube. :-(

I thought about putting a small series resistance in the tube cathode and using that I*R voltage to observe what's coming from the power supply with the scope, but I chickened out. I don't know how far the cathode is held above ground, (assuming it is) and I have no desire to zap the scope, myself or anything else in the room. (sounds of chicken clucking...)

Thanks to all that helped with this, I learned quite a bit about Chinese laser engravers while I was at it.

--jim