St0rmBringer

Updates (July)

Contact: tae@st0rmbringer.uk

June 2025

July 2025

  • 26/07/25 Well that’s been a couple of insane weeks for sure but heyho ,here we are. Got to get back to updating more often but between looking at another new machine, Medic work (yep I’m a part time Medic as well) and other sundry shit that takes up time I simply didn’t get round to it. Probably going to be more on tips and canned cycles.
  • Quick shout out to the “New guy” Ari, this lad is gonna be good, he has that sponge like interest in being an Engineer. A “good luck” to Jimmy in his new choice as well. Decent guy, although utterly shit taste in Music.
  • Bloody August soon so that means a new “Updates page”. Onwards and upwards.
  • 13/07/25 I post on quite a few forums when time permits, this post went up on one of them over the weekend.
  • “Greetings everyone, I am a mechanical engineer whose main job is CNC programming using CAM software, and in general running the machining department of my company. As this is my first job and I have no prior experience with CNC machines it can be kind of difficult to set proper speeds and feeds when programming our CNC mills”
  • We really aren’t going to make it are we? No insult intended to the chap but FML, THIS is what we face these days, how in holy fuckery is somebody with no machining experience being allowed near a CNC let alone being put in charge of it. That’s a perfect recipe to end up with a buggered machine and/or a dead engineer. Some may find this offensive but to be honest? I don’t care. I’ve seen somebody die in a machine, it’s not nice and never leaves you and THIS is a perfect way to cause such an accident.
  • 12/07/25 Well that was a scorcher today wasn’t it? G-Code page has now got G72 rough facing cycle added and the G70 finishing cycle added. I’ll start adding some longer program examples as each of the codes is added, a gradual work up rather than just bombing people with a shit tonne of stuff all at once.
  • 11/07/25 Bloody hot isn’t it?, not many updates during this last week as it’s been hot and I’m knackered. Heyho, there we are then. Right so the weekend? lot of Fanuc updates on coding (G-Codes) and then got some M-Codes to get up as I’ve been asked.
  • The most useful thing I’ll be posting this weekend are container programs, basically, shells of complete programs folks with a little bit more than introductory experience, can edit and modify then use without worrying they are going to crash their bosses new toy. Cut and paste and add in all your own shit to see what it does. With that in mind I’m also going to post some decent code readers, path tools. You load in your program and the software generates a tool path so you can see if you were about to make and your boss have a shit fit loading your efforts into his/her shiny new machine and battering it.
  • 10pm already and I’m old and tired so will get onto this is the morning although did manage to text an old friend/colleague I worked with in the past , Ronnie B, one of the most humble but knowledgeable engineers I have ever known and it was a privilege to work alongside him. Good to know he’s loving retirement. Enjoy Ron, you earned it.
  • 06/07/25 I’ve been asked if I can do some writing for the less popular but still available Siemens controllers and the Sinumerik system so that’s been added to the agenda, Both G290 and G291 will be posted and the associated methodology as soon as I get time. Not my favourite controller but if some folks want it I’ll do it.
  • 05/07/25 Speaking to one of my bosses last week and he raised the point that G-Code (ISO) programming looks complex or difficult to learn. Ergo wizzywig controllers may be the way to go, much as I respect him and understand his position I have to disagree. In reality, “good engineering” is by it’s very nature “Difficult”, it’s a fundamental feature of the industry. My main position is that reading code is absolutely essential, if you cannot read what is going up the screen then you have no idea what the machine is going to do next. Now I agree on smaller retro-fit machines like XYZ’s, Fagors or older Denfords etc they tend to move quite slowly so if you F**k up your code or what you are doing you will probably have time to react and fix things before it goes REALLY wrong. Even down to machine bangs, if you belt a Alpine, Colchester, DSG or Warco machine by clomping the jaws or leaving a chuck key in or getting a tool geometry setting wrong, it’s unlikely to do a massive amount of damage. The damage it will do is probably going to be fairly easy to repair. (Resetting the tool block straight, adjusting the tailstock, possibly even a realignment of the head to the bed) All of these repairs can typically be done in-house.
  • Now lets look at something like a Doosan Puma 2100.
  • 5,000 RPM spindle
  • 30M per minute rapids
  • 22kW spindle
  • 22kW is 29.5HP, that’s a LOT of power to try and stop quickly if you need to, now those rapids, 30 meters per minute. So picture your G28 position (tool change) being 250mm away from the job with std turning tools, 30MPM is 500mm per second, the turret is going to close that gap in 0.5 seconds on rapid moves. The average human reaction time to visual stimulus (something you SEE) is 0.25 seconds (given human conduction synapses velocity)
  • So if you are stood looking through the machine window as the tool runs down to the job if something goes wrong at the very instant it starts to move, you will have 0.25 seconds to rectify the problem before the tool hits the end of the bar. Reality? you are not fast enough, that’s not me be being a dick, it’s a simple scientific fact. Realising there is a big problem, deciding to act then moving your hand from rest, at speed to the emergency stop is not going to happen in the 0.25 seconds you have left.
  • Outcomes?, you can miss-align the turret, damage the tooling, damage the job, twist the turret mounting, damage the lead ball screw, twist the bed, crack machine castings. Just with ONE mistake any and all of these things can happen.
  • No? Ok lets talk about one I personally SAW (not with my current employer) . Nice £100k+ Puma, the engineer stuck it in manual, used the turret index to bring tool 7 into line, set the workshift off of that tool, then homed the machine, pulled the rapids up to full and hit the button.
  • So? Well using the turret indexer doesn’t call up that tools offset, you need to be in MDI and call the tool and offset from there.
  • Turret went screwing in 250mm different to where it should be (or thought it should be)
  • BANG
  • It tore the entire tool block off of the turret, twisted the turret in 2 axis (radial and parallel), knocked the head out of alignment. The tool block alone was £480 + VAT and you don’t want to ask what a new turret cost, a 3 month wait for a company engineer to schedule coming in to fix it, a £15,000 bill for the engineers time and attendance and spares for multiple visits and the loss of production for the time period. The conservative estimate was £30K – £50K in all. All because an idiot didn’t check what he was doing on a machine that moved faster than he could react.
  • So back to ISO code, you will never eliminate machine bangs, it’s one of those things that will happen, all we can do as engineers is minimise them and one way of doing that is to understand what we see on a screen another is by making sure appropriate people are given access to those machines. Wizzywig coding is fine and a lot easier for people who maybe aren’t as experienced but experience is what stops £100k machines being turned into scrap metal and/or people getting hurt.
  • 04/07/25 Ooo visitor counts are up by quite a bit, averaging -4,000 visits a day so the backend is reporting. Ok so a few email requests I’ve had as well. “Will you be doing anything for other controllers?”
  • Yep, as soon as I’ve finished with Fanuc it’s going to be probably Seimens (Sinumerik) and Deckel Dilaog, I’m not a Seimens fan and I really despise Deckel, as I’ve been asked though, I’ll do it. Along with this will be set up instructions etc if you have just purchased a CNC and really don’t know where to start or you have a CNC in the corner that would be great if you could run it.
  • Ahh well, back to finishing some designs and drawings and editing the G-Code page.
  • Ooo I just realised, on the G-code page I have moved G70 to AFTER G71–G72, there’s a good reason for this, unless you understand the roughing cycles knowing how the finishing cycles work is a bit premature hence the move in the list.
  • 03/07/25 Hectic week so far with very little time to do much on the site. That said been playing around with CAD again today. One of those “I need this” items that I couldn’t get easily so thought I’d make my own. Yep a “Thread counter”..so what’s that?
  • Well if you are cutting manual threads and use a DRO you will have a start point and the passes you have made, that’s great till somebody (looking at you Skin) speaks to you and you forget the setting from the last pass on your X and either end up cutting air or trying to take a 3mm pass. This usually means I have a notepad and write down the pass on each one so if I get press ganged away I don’t lose my last position. Given I usually have dirty hands “Pen and paper” is a pretty shit idea so decided to make a thread recorder for 3D printing. I’ll post the cad files up here so anybody can make their own if they wish to.
  • Onwards and upwards as I want to finish this design and get some sleep if that’s possible.