Help dialling it in please :)

[davek0974]

Having finally got a nice shiny Powermax 45 and some time to play with it I have some results that I may need a little help with?

This is 3mm HR steel, something I cut a fair bit of. As everyone says to do so, I reached straight for the book settings, naturally 3mm is not listed so I guessed on my starting parameters based on the upper and lower settings listed.

I set 4000mm/min, 45A, 3.8mm pierce, 1.5mm cut and 117v

I ran multiple straight cuts as usual, increasing speed by 100mm/min each time and after 12 cuts, looking at the back showed that 4000mm/min was perfect as the rest showed increasing signs of dross.

Doing straight line cuts on flat steel I found the voltage needed to be 105v and not 117v.

Part is a little tab and slot test I have used before. All pictures are straight off the table, no cleaning.

slot part, Top view

IMG_0865.jpg (348.34 KiB) Viewed 3024 times

IMG_0865.jpg (348.34 KiB) Viewed 3024 times

slot part, Bottom view

IMG_0866.jpg (308.83 KiB) Viewed 3024 times

IMG_0866.jpg (308.83 KiB) Viewed 3024 times

tab part, Top view

IMG_0867.jpg (261.94 KiB) Viewed 3024 times

IMG_0867.jpg (261.94 KiB) Viewed 3024 times

tab part, Bottom view

IMG_0868.jpg (294.96 KiB) Viewed 3024 times

IMG_0868.jpg (294.96 KiB) Viewed 3024 times

Tab part edge

IMG_0869.jpg (164.25 KiB) Viewed 3024 times

IMG_0869.jpg (164.25 KiB) Viewed 3024 times

Slot part edge

IMG_0870.jpg (136.83 KiB) Viewed 3024 times

IMG_0870.jpg (136.83 KiB) Viewed 3024 times

So, I see little dross which seem good, some odd angularity which seems bad, the corners of the tab are poor presumably too fast?
A little top dross, not sure what to do about that?

I have checked the initial height setting multiple times and it seems to be reliable and accurate.

Any views or tips??

The slot is 3mm wide and the short sides are very rounded, is there a way to help that???

[Mtw fdu]

What I can see at first is the plate itself is quite dirty. It seems to have a lot of rust in spots.

The work I do I always use clean material. Here is a pic of the quality I use.

DSCF0090.JPG (206.92 KiB) Viewed 3002 times

DSCF0090.JPG (206.92 KiB) Viewed 3002 times

Just my 2 cents!

Mtw fdu.

[Metriccar]

You're getting dross in the corners, so it is slowing in them already, but I would try slowing the acceleration even more.

I have a 65 so I can use Fine Cut, but apparently the 45 cannot?

I have found that averaging settings between thicknesses shown in the manual helps get the right setting.


Actually I tried this. 3mm thickness is 43% of the difference between 2.7mm and 3.4mm (3mm-2.7=.3. .3/.7=43%)
Cut speed is 4800 for 2.7mm and 3550 for 3.4mm, or 1250mm difference. Take 43% of that and you get 538 ipm. 4800-538=4262i[m.

So if you tried 4262 Ipm and tried again the 117v maybe it would hold the right torch height. And it may still need to have lower acceleration for the corners.

That's what I would try.

[davek0974]

Thanks,

the surface rust is more like a stain, these are skeletons of previous jobs i am using for tests so have been sitting around a while but its not scabby.

I am going to try lowering the acceleration today, never thought that it could be too high!

I have a feeling that the voltage is what it says but might not be what it really is I have a feeling the difference that I and others see between chart spec and real life is caused by math rounding errors - think about it - you are dividing say 119v volts by 50:1 so the system only reads 2.38v then the THC unit is multiplying it again for the display in Mach3 by 50 so any tiny errors in mere fractions of a volt will cause quite large swings in the display.

I guess I could measure the actual voltage but not sure if A:it will work with a digital meter and B:it's worth trying?

I think a lower ratio would help - the CandCNC RAV01 card uses a 7:1 ratio so has more 'real' volts to play with.

When I set my speed, i use flat steel so there is no bow when the IHS hits it (not ohmic yet) and also know that my gantry is very parallel to my slats.

I can try doing some mid-cut e-stops today and see if it really is maintaining my 1.5mm cut height or not.

[SeanP]

Do you think you are seeing more angularity on the 3mm with the 45 vs the 30 Dave?

You could be right with the voltage division theory, I usually start -10v to book and tweak it.

Did you calibrate the torch volts to show 126v? mind you I checked mine the other day it was 3v out, it didn't really alter much though.

I will try that part on my machine today with mild steel just out of curiosity.

I went the other way with motor acceleration to get better corners, will check what they are as well.

[davek0974]

Hi,

at the moment, yes a bit more angularity but its early days, this change has taken me right back to day 1 I guess.

I have not checked or adjusted calibration, might be worth doing.

I have more pictures from the lower acceleration tests today but the uploader is not working for some reason, will try again later.

Basically looks like the lower acceleration is NOT what it needs, the roots of the tab part are now perfect radii outer corners seem a little better though.

I am going to go try upping the acceleration now beyond the original settings and see what happens.

Any figures you can post would be interesting.

I did a mid-run stop and the 1.5mm cut height from IHS is spot on.

[SeanP]

yes I have some pics to upload, but struggling as they need resizing from the phone,
Motor tuning on mine is 9500mm velocity, 2500 acceleration on x,y
Z is 3800 velocity and 1400 acc
My x&y are 3.75:1 gearing

[SeanP]

Just to make you feel better with yours Dave

[davek0974]

Thats a relief, it was doing my head in, now I know its not just me!!!

I have a feeling the angularity in the slot is where it is cut anticlockwise - the first (left) side is good but when it gets to the second side there is no slug left in the middle as its only 3mm wide, this seems to pull the arc towards the centre leaving a sloping right-side??

I have had some success with sheet cams' "on internal corners" cut rule, slowing to something like 40% for 2-3mm seems to help a bit, will play with that some more now i know your results match mine,

I tried higher acceleration but my aluminium section gantry flexes and leaves ripples everywhere for no real benefit, i have slowed her down again.

[davek0974]

Given up for the day, 8 hours of burning tests is enough

As the tab test was not looking too shabby, I thought i'd try and reduce my backlog and run a job. Bad idea, it looks terrible, I have ripples, bad divots on lead-ins, and a really odd angularity fault which only affects the top edges?? Maybe a 30degree bevel on those but left, right and bottom edges seem ok.

The holes are 6mm cut at 60% of 4000mm/min

PM45 test job, yuk

30XP same job

PM45

30XP

PM45 back

PM45 front

Looks like I might have to return to the 30XP and send back the 45 I can't see any tweaking fixing this myself.

[acourtjester]

Sean’s number point out something I used as a rule of thumb, the ratio between velocity and acceleration. It seems like acceleration work better at 25% of the velocity number. Cutting corners you are changing directions and there for the acceleration really shows up. And will also effect the flexing of the table due to the directional change of the moving weight. Something you could try is changing the orientation of the cutting so the side cuts are not in line with the table IE 45 degrees not 90 degrees.
Looks like your pierce delay is to long.

[SeanP]

Oh heck,,, that's not sounding good!
Are those 3mm as well? The cuts from the 30 do look better, I've struggled with angularity to be honest from day one, but put it down to maybe using non genuine consumables, that can't be the case on yours though.
Maybe it's not ideally suited to cutting thinner stuff, 6-8mm is really good.
I've never tried the 30 consumables in the 45, it might be worth a try though.

[davek0974]

Yes, I'm not too impressed yet.

I could try dialling it in at 30A on the 45 consumables, don't have any 30A ones yet.

Typically, after a week of nothing to cut, i pull the machine apart and three orders come in

I couldn't honestly sell the one i tested, too poor quality.

Putting the 30XP back is possible but will mean cutting the plug off the CPC cable as i used my existing CandCNC leads to make that up with a plug from RS components. It also refused to cut 3mm stainless very well, this was one of my reasons for the upgrade.

I'll have another play tomorrow night but then need to make a decision I think.

I thought it would be easier TBH.

[acourtjester]

With your CNC cable connector I would combine the 2 on the cable end so you could swap between the 2. Have the cable with 2 ends that are parallel

[SeanP]

If I'm carefull these turn out quite good,, 3 & 10mm.
To start with I thought they were past the limits of my setup, but can do them ok now.

[davek0974]

Thanks again,

hopefully tomorrow evening I'll have some time to try again, will try lower currents and speeds, maybe get lucky with some dross but better cut.

I can't believe it won't work, just not hit the right settings yet.

I also fail to see how my machine is that bad, there is no slop and i mean no slop at the torch mount in any direction, you can wobble it but only with firm hand pressure. There *is* a possibility my ripples are caused by belt stretch, the belts are T5 pitch (IIRC) but only 10mm wide.

I would build a steel table if I was 100% certain it would be better, but looking your excellent build and the very similar results you had on my tab test part, I am not convinced.

Maybe I was just lucky with the 30XP but I can't produce what I need to on the 45 yet, I know that now.

I need to make a decision soon though as the 30XP is due to be returned as part of the trade-in deal I struck with my supplier.

[davek0974]

Bit more info...

The PM45 was a big purchase for me, I did my homework, saw lots of stunning samples of small, detailed work, just like where I need to be making stuff, struck a good trade-in deal on the 30XP and went for it, backed up safe in the knowledge that nigh-on every smaller shop plasma cnc post on the internet uses a PM45!

My drive belts are steel cored and I can already play a tune on them , what I am trying to explain is that the errors I see appear far worse than the quality of my machine leads me to believe - pen tests appear ok, not perfect at the top end of speeds but good enough.

I spent last night watching videos on youtube of cnc's running a PM45 and many of those were built with far less apparent stability than my one - I mean very long torch drops below the gantry, one machine had a single 1" square tube gantry and was still making work, most if not all are stepper driven with Mach3.

I have no idea how Mach3 handles corners, it's a bit of black art it seems and the best advice always comes back as "just let it handle it", however, I do know there is a relationship between speed and acceleration whereby things can go badly wrong - the internal corners turning into radii when you set acceleration too high for example.

I have just downloaded a little tip sheet from Hypertherm and will take a day off the day-job tomorrow to get this thing working OR refit the 30XP (which I really don't want to do), I have jobs backing up now (typical!) and I hate that predicament.

The top spatter is too fast/too high - this is good as I would prefer slower always.

The mess I got on the back of the test job I finished with yesterday I am identifying as high-speed dross as its very hard to remove, not bubbly and quite fine - again caused very high speed - this is good also as I want to go slower.

So, for some reason, the book specs do not seem to work for me - I am sure Jim Colt will be stating that they should, but, not it seems for me?

Is there a way my machine could actually be running faster than it thinks?? (will ask on Mach3 forum)
This would back up many faults I have had/have got???
I might devise a design made up of many zig-zags etc of a known length and try timing it.

So, tomorrow I am going in with a fresh eye and calm manner and will try going slower, come the afternoon, if I am not progressing then I will have to refit the 30XP.

Sound like a plan??

[SeanP]

Regards the belts Dave you should be better than mine I only got 9mm htd belts, wish I had gone for 15mm now.

I seem to remember thing maybe my setup wasn't calibrated right once as well, it doesn't take a lot to check say a 1000mm/min even with the tape and stop watch

It's good that you have the option to go back to the 30 if it does come to that.

I should have looked at your website before, some nice parts there, did you cut the Westie in stainless with the 30? it looks great detail, not sure I could get that with mine.

[davek0974]

Yes the westie keyrack was done on the 30XP - 25A, 1.2mm 304 SS, this is the level I need to reach on the PM45 to be happy

I have made up a 500mm test line, programmed at 500mm/min and hope to try that tonight, just to verify Mach3.

It does seem odd that quite a few users have to run below book specs, maybe there is an undiscovered issue with certain PC types and Mach3 at play here???

The voltage issue, I am still putting down to the maths involved in dividing a relatively low voltage by 50:1 and then multiplying again for display - that never seems a good idea to me. Maybe the 20:1 option would be closer to standard. I did check last night and my DRO shows the required 126v when on test, I did't go any further as I have enough issues to play with

[SeanP]

Yes I don't think the voltage discrepancy is anything to worry about, not sure why we have to go below book spec for speeds though, I have used 3 different pc's on mine, there was no variation at all with the toolsetting on each one.

I think your only chance to equal the Westie cut 'might' be to use the 30a parts in the 45, then again I see a post on Candcnc about discontinuing those parts so not sure

Hope you have better luck with the next runs.

[davek0974]

Hmm, i will need to try the 45 at all setting from 20 to 45 on the SS, I don't have any 30A consumables yet and I have read that the only benefit is a slightly smaller kerf, but at the cost of shorter consumable life.

What the world really needs is a PM45 with the new style torch etc so we can run proper fine-cut consumables

Looking at some of the work done by others on the 45, I am sure it can do a usable job, trouble is I was spoilt by delicate nature of the 30XP The big downfall is that it is a pain to get it running CNC as you need to modify the shield or you can only really drag-cut and that sucks on CNC. Even modified I only had 0.5mm to 1mm of THC travel, fine-cuts were killed at a goodly pace

This job of yours seems pretty stunning to me
http://www.plasmaspider.com/download/fi ... &mode=view

[acourtjester]

You have stated the belts are very tight (you could play a tune on them). The edges in the images you posted look odd. I wonder if they are tooooo tight?? They look like there is an isolation in the movement of the torch.

[davek0974]

Thats an interesting point and very observant of you I was baffled by what could cause that serrated cut.

I can easily remove or reduce the tensioners to see if it gets worse or better.

Is there a way to know if they are the 'correct' tension???

[steel 35]

Keep Schooling us Acourtjester

I am off to test with a pen.

[acourtjester]

It is hard to say how tight it should be but if you look at many belt driven assemblies they use a spring tensioner for the correct tension. Even the main accessory drive belt on autos have a spring tensioner to increase belt and shaft bearing life too. I'm not saying I'm right it is just another thing to check when working out a problem it help to have other views. Great to have Plasma spider to give us the contacts with other CNC users for assistance.