I have recently upgraded from a 50w to a 60w laser tube, after initial setting up problems with mirrors, I now have it so that it is really sharp on the line setting but it very blurred on the fill settings. I have tried switching off the bidirectional setting , this made it a bit better but still not good enough. Is there some setting the the Ruida controller i need to change?
Thanks
Might be a difference in the response time. In HVDC-excited glass tubes, when the Ruida’s “fire” pin goes from on to off, there’s about a 1ms response time before the laser output actually shuts off. And a different delay from when the Ruida fire command goes from inactive to active and when the beam output actually reaches full power.
It’s not just a lag either, the power ramps up and down as it goes on and off. That power ramping means the edge of a higher speed raster isn’t going to be a sharp vertical edge but a blurry slope.
This varies by tube and power supply so maybe you had a faster response with the older setup?
RF-CO2 lasers are great because the power change is much faster- essentially instantaneous.
Did you also replace the power supply?
If so, then you must measure & set the Scanning Offset Adjustment values to match the new power supply.
The nightmare scenario: getting a bad “new” tube or “new” power supply, as detailed in another discussion:
Testing the beam pattern as described there will be informative.
I have not done this, cannot see where to do it but will keep looking
How do I know what the response time is?? Thanks
You don’t – you either have to measure it. The next best option is to use the specification of most lps which is a response of <=1mS for 90% placard voltage. This is a common response time value.
Did you upgrade the tubes power supply to handle the higher wattage tube?
Most people don’t really have a problem with a 1mS response time, unless you’re running fast… how fast are you scanning and what’s the interval?
Hi, yes fitted new 60w tube and power supply at the same time. wheredo I check/ set this please?
Check/set what?
response time
Sorry you get what you get… That’s why the specifications are written as <=, so as long as the lps is within the specified parameters it passes, generally speaking. If it fails, it could end up on the market via unscrupulous dealers.
It’s difficult to measure since it requires you to time the duration from laser enable assertion on the lps to when it starts to lase. Few, if any of us have the equipment to measure the high voltages that are present when the tube reaches trigger voltage. The only other means is to use some type of photo detector to know when the machine actually starts lasing.
Most of us assume it’s the 1mS. This inherently brings up some physics about how fast you can actually toggle the tube of a dc excited co2.
If it responds in 1mS or 1/1000 of a second, then with a speed of 1000mm/s the best I can do is to toggle the tube once every mm. Meaning a dpi of 25.4. Cut it in half to 500mm/s and the dpi increases or doubles to 50.8 dpi… at 250mm/s it doubles again to 101.6 dpi. Depending on what you’re doing, speed can be a problem. This is also assuming worst case response time, but few know what their response time is anyway…
it’s not just toggling. the power ramps up over ~1ms, so if you try to do a deep fill on acrylic at 1000mm/s, 1ms is 1mm of distance. so the edge of the acrylic cut won’t be vertical, it will have a slope across about 1mm and the slope’s profile roughly matches the beam’s response profile.
also:
look up your controller settings for max x axis velocity. most often it’s 1000mm/s. also note the max accel.
go into LB, make a horizontal line 10" long. set speed to the controller max (let assume 1000), max power to 50%, min power 50% too, and override pwm to 0.2khz. that means once it gets to full speed the Ruida’s fire command is 2.5mm on and 2.5mm off.
as this is much larger than the focal spot size, it will come out as dashes on the actual work.
look at the x accel parameter to calc how long a line has to be to reach 1000mm/s. your line has to be at least twice that long because it also looks ahead and starts braking that long before it needs to be at a stop
so, excluding the accel zones, the middle of the cut where it reaches the specified constant velocity, you can think of distance as time, interchangeably.
1000mm/s at 50% 200hz should have that 2.5mm on/2.5mm off period. and 1ms of time happens across 1mm of that cut.
lay down notebook paper, the 11" long side horizontal, and hit that. ignore the ends, you want to look at the middle where it’s reached constant velocity. you can see the power output ramping profile as browned regions between the cut and uncut areas on the dashed line. note whether the cut went left to right or right to left. the profile for turning on is not the same as turning off, the edges will look different
@jkwilborn & @Dannym may be making it a bit more complex than necessary. 
Measuring & setting the Scanning Offset Adjustment to match your new power supply should get you almost all the way to the goal:
Scanning Offset Adjustment - LightBurn Documentation
Do that and see what happens …
1 Like
I was able to sharpen my bidirectional engrave fill image by trying settings until everything became sharp. It helps to scan at a low scan per mm setting so you can see how the lines from both directions are aligned.
It would be nice if you can explain what you did and how it fixed anything…?
Overscan adjustment doesn’t change the response time, only when the Ruida asserts laser enable.
The delay is inherent in the electronics. We can make up for certain aspects, such as when we’re scanning, but the delay remains no matter what operation you’re doing.
There is also a high voltage present when the tube turns off, as the HV isn’t bleed off by any low impedance path to ground in such a short time. When it lases, after the first pulse, it should have a faster response as the voltage is already partially there.
I have a hv meter, so I have a little bit of a clue, I think, on what it’s doing. 
Yes, but it will adjust the delays at various speeds to match the new power supply, which is pretty much guaranteed to be different than the old supply.
Worrying about anything else is a second-order concern when the initial problem is “very blurred on the fill settings” with a new tube & supply.
I can’t imagine how this happens if everything is set up properly in the first place.
I don’t know know where I’d change something in the Ruida or elsewhere if I wanted to blur the fill settings on purpose.
Easy:
- Install new tube & power supply
- Run similar jobs as before
- Get crappy results
- Post question on LightBurn forum
Now, I may have missed something again, but so far in this whole discussion the only change after installing the new parts was “initial setting up problems with the mirrors”.
After verifying the Scanning Offset Adjustment settings, then a whole bunch of other possibilities rear their ugly heads. But we ain’t there yet.
Spent all mornig doing this, it did not get me to where I wanted to go, Still get bluur fills!!
Guess I will need to get another new tube.
hi, only thing that seems to help is turning off the bi directional setting but then it takes longer to engrave!!
It might be helpful to post a good quality photo of what you mean by blurry. My test pattern was a 10mm square
Also, show a pic with and without “bidirectional” on your fills.
With bidir, the error in start/stop points don’t line up between even and odd raster lines and that could be described as “blurry”.
If you don’t use bidir, then, as long as the error on start/stop points is consistent (and it should be) then the quality of the fill itself will be fine, it’s just offset in the wrong place.
This is readily apparent if you vector engrave vertical lines that should line up with the edges of the raster. The up-and-down vector movement will make a line at the correct X location for sure, even if there’s something wonky with the timing of the firing of the tube. But a horizontal raster’s error will shift the fill left or right and it will be readily apparent that it does not line up.