Page 3 of 3 FirstFirst 123
Results 31 to 43 of 43

Thread: Engraving with a RECI tube - what's the general consensus power that actually works?

  1. #31
    I have a redsail and engraves great. Slower though. I had a uls for years but had to sell it. Engraving on my redsail does not engrave as good but its close

  2. #32
    Join Date
    Sep 2016
    Location
    Georgia, USA
    Posts
    383
    >>As in, a lens or aperture or some other device that could simply block or absorb (which may be a bad thing) some of the incident beam before it gets to the regular lens?<<

    If you have a Ruida controller, when in scan mode (engraving) you can select between "common mode" and "special mode." Special mode might do what you are seeking but the manual seems to warn this mode can shorten the life of the tube.
    700mm x 500mm Ke Hui KH-7050 Laser
    80W EFR F2
    S&A CW5000 chiller
    Chuck style of rotary attachment

  3. #33
    Thanks very much for all the suggestions.
    Details so far: I decided that it would be a worthwhile effort to get soime figures down that would be useable, on the big Goldenlaser, because the you-beaut flat-bed machine I've ordered from China is in go-slow mode due to their new year holidays there right now.

    I tried the Goldenlaser 150w on assorted settings of power, speed and acceleration, and also resolution. And focus.

    Conclusions:
    1. get rid of the 4" FL lens and use a 2" lens - that was the best help for starters.
    2. I tried power variations from 18% to 32%. Of course the depth or effect of that varies depending on speed and acceleration... coinsidering the firing and turning off of the tube in a hurry.
    3. I tried assorted speeds and accelerations - the faster speeds at a set acceleration needed a lot of landing and turning around room, (up to a foot each side) while the slower speeds needed barely an inch of turning space past the image.
    The faster speeds gave a more blurred image, and lower accelerations didn't help either.
    Changing too many parameters at once was useless as you didn't know what one had the greatest effect.

    IN the end, I established that although I could set the acceleration at up to 5000 unspecified units, 1500 worked well. 700 did not.
    Speed - I tried at 1600 mm/sec, 800. 600. 400, 200, 100. The slower speeds were a heavier burn to a degree, and to a degree they were not.
    They were also a sharper image.
    IN the end 400 was nice and clear, but overall too light in tonal density, while 200 had too much white etching blotting the picture more than I wanted, so I went for 300 mm/s speed. (at 1500 acceleration)
    I had tried assorted power settings, but settles on 30% as giving me what I liked the look of the most.
    It may be a linear scale such that I could go lower, and slower, but I'd put enough time already into it, and the 18-26% trials were just too light for my fussy preferences. And there reaches a time when you don't want to wait all day for it to end...

    Resolution: I'd created the original file as a photoshop composite of 6 images,( for various reasons- that's another story not important for now), and made it at fullsize ( 8' x 3') at 144 pixels per inch. After some judicious sharpening (unsharpmask, contrast tweaking, blurring of unwanted JPG artifacts, resharpen edges, blurring background more to remove detail where it would be a distraction, enhance detail where it was going to be appreciated, then fiddle the curves or gamma, for a nicer blend of greys in between the darks & lights that were necessary, and fiddle the curves so there were no 100% blacks or 0% whites blowing out, but instead 92% and 6% at the ends of the range, for detail in the shadows and in the lights... and more fiddling and more... plus I had to fake the train smoke, then delete the background altogether, flip it, and make it a negative. Then convert to BMP with diffusion dither and save. 8.5 meg total size of BMP.

    The small 1/4 length samples ( 2' x 8") I had the resolution at 288 ppi, but the laser could not handle the memory needed - weird- as it was only a 2 meg file... so I enlarged the scan gap from 0.088 to 0.167 ( double) and it was happy.
    The samples allowed me to test the various power, speed & acceleration options.

    As I said I settled on E=30%, S=300mm/s, A=1500, Scangap 0.167.
    Later I'll upload some sample photos.
    After that I cropped out a 26 x 12" square at fullsize, 144 ppi, and ran it on the above settings - absolutely BEAUTIFUL!
    Again the machine's RAM could not handle the 0.088 scangap, but was happy with 0.167

    So, for a 150 watt machine, chinese, the engraving standard is/was in my opinion, on this job, phenomenal for what I need- after I'd figured out the best tuning of the variables.

    Whether I've damaged the tube or not, I can't say, but the slower speed gave it more time to rise and fall, compared with the higher speeds.
    The negative end of the tube is purple - and the RECI website says that is a sign that the tube's life is largely spent, but I'm not to blame for that colouring!
    I ordered another tube from the same mob who're putting the big laser together - as freight would be $1k or more on a single tube Reci W6, but freight will be about zero when combined with the big unit. Their price was great - $680 USD, because I'm already buying the rest from them.

    In the mean time, this RECI tube has improved a bit, and now runs on about 20-20.5mA when I set the software power at 100% - and it was only on 21-21.5 mA when we took delivery last year. (it had been sitting down around 18mA for a while, cutting).
    __________________

    So, my next query will be about stitching pictures together to engrave something larger than the bed - but I'll start a fresh thread for that.

    I'll add photos here when I get them off the phone.
    Thanks!
    Last edited by Ian Stewart-Koster; 02-02-2018 at 3:02 AM. Reason: typo
    Best wishes,
    Ian



    ULS M-300, 55w made 2002 with rotary. Goldenlaser 130 watt, 1300x700 made 2011.
    Flat bed 2500x1300 150/90watt 2 tube laser, 2018 model.
    Esab router, 1989, 4.5 x 2.0 m, conv. to Tekcel, and modded a 2nd time.
    HP L260-60". Roland PNC-1410. Mimaki GC-130 SU.
    Screenprinting carousel 6x4 and 7x4 ft 1-arm bandit vac table.
    Corel Draw X3, Illy, Indesign & Photoshop CS2 & CS5, Enroute 4
    Pencil, paper, paintbrush, airbrush & dagger-liners & assorted other stuff.

  4. #34
    This is the actual FULL-size sample, beside my hand, for scale, of the settings I ended up being most happy with.
    20180202_173121_Medium.jpg

    Below are steps along the way to achieve those settings:

    This is the 1/4 length, 1/4 height sample of what the finished product will be like- this is 2 ft long: I didn't like the train smoke so changed th at part of the file & edited it. Also added subtle detail to the bridge supports.
    20180127_224858_Medium.jpg

    One of the earlier ones...
    Offhand it looks like the good final balance of light and dark, but crummy resolution.

    20180202_173231_Medium.jpg

    This was a bit too light/faint- at 400 mm/s speed. Not enough whiteness.
    20180202_143058_Medium.jpg

    This was too slow at 200 mm/s and this was with the 4" FL lens- way too much white/burn and blown out detail)
    20180202_173210_Medium.jpg
    Attached Images Attached Images
    Last edited by Ian Stewart-Koster; 02-02-2018 at 7:48 PM. Reason: typo
    Best wishes,
    Ian



    ULS M-300, 55w made 2002 with rotary. Goldenlaser 130 watt, 1300x700 made 2011.
    Flat bed 2500x1300 150/90watt 2 tube laser, 2018 model.
    Esab router, 1989, 4.5 x 2.0 m, conv. to Tekcel, and modded a 2nd time.
    HP L260-60". Roland PNC-1410. Mimaki GC-130 SU.
    Screenprinting carousel 6x4 and 7x4 ft 1-arm bandit vac table.
    Corel Draw X3, Illy, Indesign & Photoshop CS2 & CS5, Enroute 4
    Pencil, paper, paintbrush, airbrush & dagger-liners & assorted other stuff.

  5. #35
    Join Date
    Jul 2014
    Location
    Norfolk, UK
    Posts
    94
    Nice work!! I think you discovered one thing I forgot to mention - massively over sharpening (unsharp mask) the image really gives massive edge contrast for the laser to work with. It looks horrendous in photoshop but amazing in print - this is something I never learnt before I sold my chinese laser. Even on the Trotecs I had access to in Canada it changed a muddy picture into a work of art pretty quickly. I really like your idea of going 6%-92% on the levels, thats very smart I'll make sure to remember that (hoping to have another chinese laser delivered in 2 months)!

    That first pic is definitely something to be proud of. Well worth the time experimenting.

  6. #36
    Thanks for the compliments, Mark.

    Yes, unsharpmask is the key! I learned that when lasering a photo of our clydesdale team in plywood. You need unsharpmask to create detail where there mightn't ordinarily be any. I often go up to 500%, and then a rerun at 180% after that.

    The 92% black maximum & 6% white minimum learned when editing photos 18 years ago for halftones for printing in a book.
    I apply the same principle to photos when we have to screen print them.
    100% black becomes a detail-less puddle of ink when you account for dot gain - so does 95-96%. 92 is dark but allows for detail in the shadows.
    (I'll only leave text at 100% black)
    0% white is a blowout of nothing- no image- and 5% when you account for screen blockage is often also printed as nothing. 7-8% or so becomes a faint pale grey to show you what you need to see, and that it's not just paper background but actual picture.

    Ctrl-M in photoshop edits the curves to apply the percentages most easily, I find.

    The actual photo I said was a composite- the best bridge pic was a setting sun one - and anpother of the carriages was the same, but the resolution was woeful - a 65k photo! The train in that pic was running the wrong way, and the clarity was zero. So I git the train from another pic, but had to get the needed detail and needed light and angle. I visited a few trains with the camera to get the right one, and right lighting - many trains are just all black and dirty, and many appear like that in photos too. The best train pic was stationary, so I had to fabricate the smoke for it... The bridge posts/piers were poor in all pictures, so i went and took my own- although the bridge is gone, the posts remain.
    Just had to get the light right, and then Photoshopped the shadows in on the lowres pic...

    Long story but worthwhile project! Just have to finish it! The attention to detail is appreciated in the end
    Best wishes,
    Ian



    ULS M-300, 55w made 2002 with rotary. Goldenlaser 130 watt, 1300x700 made 2011.
    Flat bed 2500x1300 150/90watt 2 tube laser, 2018 model.
    Esab router, 1989, 4.5 x 2.0 m, conv. to Tekcel, and modded a 2nd time.
    HP L260-60". Roland PNC-1410. Mimaki GC-130 SU.
    Screenprinting carousel 6x4 and 7x4 ft 1-arm bandit vac table.
    Corel Draw X3, Illy, Indesign & Photoshop CS2 & CS5, Enroute 4
    Pencil, paper, paintbrush, airbrush & dagger-liners & assorted other stuff.

  7. #37
    Join Date
    Jul 2014
    Location
    Norfolk, UK
    Posts
    94
    Thanks for the tips, I'll remember those and try some pieces I attempted last year again once I get my new laser. The trotec just made a big black blob of all the shadow areas, so sounds like your tips will help a lot!

  8. #38
    The other useful tool in Photoshop, Mark, is the 'dodge' (looks like a lollipop) took. You can set it to just lighten the darks only, and then mouse over areas to lighten and get the extra detail in blackish shadows. Like extreme use of unsharpmask, it does not hurt in this process, but helps.
    Best wishes,
    Ian



    ULS M-300, 55w made 2002 with rotary. Goldenlaser 130 watt, 1300x700 made 2011.
    Flat bed 2500x1300 150/90watt 2 tube laser, 2018 model.
    Esab router, 1989, 4.5 x 2.0 m, conv. to Tekcel, and modded a 2nd time.
    HP L260-60". Roland PNC-1410. Mimaki GC-130 SU.
    Screenprinting carousel 6x4 and 7x4 ft 1-arm bandit vac table.
    Corel Draw X3, Illy, Indesign & Photoshop CS2 & CS5, Enroute 4
    Pencil, paper, paintbrush, airbrush & dagger-liners & assorted other stuff.

  9. #39
    Join Date
    Jul 2017
    Location
    New Zealand
    Posts
    27
    Quote Originally Posted by Ian Stewart-Koster View Post
    Thanks for the replies, and info.
    Dave, by 80w, do you mean eg a 100 watt tube that masquerades as 80 watt because you should not go up to 100% - like the Reci W1 ?
    Or a lesser 80 watt tube?

    The Reci W1 says 80-100w, but the Reci W2 says 90-100... That's weird sort of semi-deceptive marketing ploy, I'm thinking...
    Reci aren't deceptive - they are the most honest and thorough of the tube makers.

    They give you four ratings: max, rated, operating and long-term power.

    Reci, EFR and SPT are all reliable and honest with their descriptions.

    As to 'not go up to 100%', where did you get that nugget of misinformation?

    IF you don't calibrate your machine properly, you're bound to have issues.
    Homemade 6W 900x500mm diode engraver
    600x900mm Reci W4 with Ruida 6442G
    300x200mm 50W desktop hacked to bits with mods
    ...waiting on a 1325 Reci W8...

  10. #40
    My Triumph is 'calibrated' by mathematics --

    I have NO idea if my power supply can be adjusted to match actual power percentage relative to max mA output to what the software thinks it is, and honestly, I really don't care- From day 2 (day one was a nightmare ) I figured out that (a) I should not run my laser at higher than 26-27mA, and (b) 72% as sent to the laser to vector nailed 27mA dead on (and still does). And the down slope is fairly linear, 36% power hits around 13mA, etc., which should equate to about 40 watts, and it's output is very close to that of the 40w Synrad in my LS900...

    It's my understanding that most fareast lasers act much the same way, around 70% power hits the safe mA max, and 100% is going to hit the MAX max, with mA's well into the 30's...

    That, in my mind, explains the 'not go up to 100%'...
    ========================================
    ELEVEN - rotary cutter tool machines
    FOUR - CO2 lasers
    THREE - fiber lasers
    ONE - vinyl cutter
    CASmate, Corel, Gravostyle


  11. #41
    Join Date
    Jul 2017
    Location
    New Zealand
    Posts
    27
    Quote Originally Posted by Kev Williams View Post
    My Triumph is 'calibrated' by mathematics --
    Versus mine being calibrated by tools. I guess guessing is OK for you - for me, not so much.

    It's my understanding that most fareast lasers act much the same way, around 70% power hits the safe mA max, and 100% is going to hit the MAX max, with mA's well into the 30's...

    That, in my mind, explains the 'not go up to 100%'...
    Your understanding is flawed. The maximum current you can put into a Reci W6, for example, is 32mA. That is what they call the 'test current', used to calibrate the tube and to QA its maximum power for the rating sticker they put on it. At 32mA that gives you ~160W output. If you put more than 32mA in you will kill your tube in tens of hours.

    If you run it at 32mA for any significant length of time, you will run it out in a thousand hours or so.

    Their recommended operating current is 30mA, which gives you the power they rate the tube at for marketing purposes - 130W. You can run it at that for ~3000-6000 hours. If you mostly engrave or your cutting needs are less than maximum, you can extend the tube's life significantly by knocking a couple of mA off the maximum operating current and get up to 8,000-10,000 hours operation. For a tube that costs ~$1000, 8000 hours+ operating at ~120W is pretty damned good value for money. At the far end of that period, expect to still be getting 85% of the rated output.

    All laser tube makers make tubes and rate them at X power for Y mA input. The '70%' BS has infected the industry from those hundreds of thousands of 'K40' analog-controlled rubber-stamp engravers that have been adopted wholesale by hobbyists and forced to do things well outside their design specification.

    Laser power supply manufacturers make PSU's and give them a range of output, adjustable via a potentiometer. Through the use of accurate tools, you set the peak output of the PSU to match your desired tube current. It's not rocket science. It's as complicated as when I set the motor speed on my wife's 30 year old sewing machine.

    If you don't match one to the other, then I guess you have to use 'mathematics' to guess at your settings. I prefer the use of a calibrated laser power meter and an accurate RMS Multimeter.

    If you use accurate tools to calibrate your tube, you never have to do guess and can safely and reliably use the full range of your tube's output, as well as the full range of power steps from 0-255 to fine-tune your output.

    Disparaging those manufacturers that strive to create better products for us to use because you can't or won't be bothered calibrating the equipment you have is, I find, borderline ignorant.
    Last edited by Mike Thornbury; 11-30-2019 at 7:36 PM.
    Homemade 6W 900x500mm diode engraver
    600x900mm Reci W4 with Ruida 6442G
    300x200mm 50W desktop hacked to bits with mods
    ...waiting on a 1325 Reci W8...

  12. #42
    Micro-measuring the power output of a Chinese glass laser whose output changes significantly with cooling water temperature changes is beyond overkill in MY real world. It is absolutely not necessary in the least that I know the exact power output of my supposedly 80w RECI at XX.XX mA... What IS necessary to me is to get the freakin' work out the door so I can get to bed before 3:30am. I've been running C02 lasers since 2001 and fibers for 3-1/2 years. I've never had to replace a single C02 tube or fiber source yet.
    ========================================
    ELEVEN - rotary cutter tool machines
    FOUR - CO2 lasers
    THREE - fiber lasers
    ONE - vinyl cutter
    CASmate, Corel, Gravostyle


  13. #43
    Thanks for bringing this thread back to life!
    The big new 2 head, 2 tube machine has been extremely useful.
    Going by a tube current max of 24mA, the w2 tube (100w) fires as low as 5%, and maxes at 100%, on 24mA.
    The w8 tube fires at as low as 6% power, and maxes on 24mA at 50% power.
    So those are the limits I have set in vendor settings.
    Calibrating different substrates took time, for best look versus optimal time, especially in 6 to 8 hour job runs.
    (Thanks Kev for waking me up to doing some engraves as vectors instead of as rasters)

    The other two machines hardly get much use, but every now and then they earn their keep, or I have all 3 running at once, from 3 different PCs !
    Best wishes,
    Ian



    ULS M-300, 55w made 2002 with rotary. Goldenlaser 130 watt, 1300x700 made 2011.
    Flat bed 2500x1300 150/90watt 2 tube laser, 2018 model.
    Esab router, 1989, 4.5 x 2.0 m, conv. to Tekcel, and modded a 2nd time.
    HP L260-60". Roland PNC-1410. Mimaki GC-130 SU.
    Screenprinting carousel 6x4 and 7x4 ft 1-arm bandit vac table.
    Corel Draw X3, Illy, Indesign & Photoshop CS2 & CS5, Enroute 4
    Pencil, paper, paintbrush, airbrush & dagger-liners & assorted other stuff.

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •