Відео

Hi. Thanks for your interest. Are you from the Bestarc company? I'll be spending the next 3 to 4 months of my spare time making this CNC plasma machine so that will be my main project for now. I have looked at your website and it looks like you only sell plasma machines and MIG welders. I have just bought your plasma machine, and I already have an expensive MIG welder, what sort of project were you thinking about? Maybe I'll email your customer services with a link to this video and we can continue this over email, thanks.

Thanks for that. I did ask for thoughts and advice so it is all welcome. If you look at the CAD again there is a vertical 40x80 section joined under the top 80x40 section to form a very deep L shape. So it is much deeper than 40x80 alone, more like 80x120. I also wanted to keep the mechanical parts, especially the ballscrew out of the way of the plasma so put it on the back. I did look at the Langmuir and JD garage cantilever designs. I think I have much more section than theirs and the 20mm linear rails and carriages are very stiff compared to 3D printed bearing holders used on JD garage. Gantry beam is aluminium and Y&Z axis will have weight reduction where possible (holes, 3D printed parts etc.). When it is built I'll test the deflection and maybe measure the stiffness at the end of the gantry. I'm OK with it, especially with the THC but let's see!

@@routercnc9517 Right on! I figured you must have thought about this stuff in your design. I'm looking forward to seeing the build and the tests!

Probably! But it was one of those industrial places with a counter, I needed the parts to confirm it was all working and there was a queue starting behind me. The Rectus fitting feels nice and quality and there are a few adapters put together, and I needed an extension on the airline so went for it. Was hoping to pay <£50 but here we are.

I had another person’s comment arrive in my email but it was gone in the UA-cam comments. So I couldn’t reply which makes it look like I am ignoring them. Try replying to this one.

​@@routercnc9517 Yes, I know YT delete some comments. My yesterday's comment contained 3 links, so, maybe was that, no idea. Was only to tell you the 3 videos the "Cracktown City" guy did about his CNC plasma cutter. Not the same style, but have some good ideas. I suppose you already saw them, but just in case. I'm gonna write again the 3 links in my next comment, right below this one, we'll see what happen

ua-cam.com/video/T-vQkJGF1Ts/v-deo.html ua-cam.com/video/ggnd5RdA-Yo/v-deo.html ua-cam.com/video/BBJ6rFitng0/v-deo.html

@@routercnc9517 Yep ! Links deleted !

​@@routercnc9517 My comment with the links was deleted and also my next comment telling you was deleted, no links this time... Not sure this one will be deleted, too...

Me too!

I think I looked at that and it was quite expensive. I'll have another look to see what it offers and compare to my current direction. Thanks.

I've not looked at those so I'll check them out. I quite like the idea of the PC being inside the control cabinet and out of the way so the NUC is a nice option and there are some models around £100. I'll see how they compare for size and availability, thanks.

OK thanks. I'll look at those. The ones I had a quick scan over where some Banggood units around £90. Came with HDMI, USB, WiFi, network, and Win 11. I'll see if Banggood would be interested in a review!

OK, look nice but no prices on the iCubes website (for some reason could not find Automation Machines but could be me!). "Look" more expensive than my cheapy option, although those are new ones, probably don't need much performance as controller should do the work.

I think it is because I emailed them and said I was based in the UK. For the US model they would normally include an adapter to take the EU plug into a US plug, but they left that out of the box for me. The price point / features meant I was happy to make it all work.

The manual says the built-in one is OK for low volume DIY, but above that then consider another air filter. When I was in the industrial pneumatic shop the other day I did glance at the filter kits, but didn't ask about prices! I think the trade off is less life on the consumables. I did think about a water tray but didn't fancy the mess. I'll try the downdraft funnel type into a bucket, with a fan just above the bucket blowing the smoke out of the garage door and see how it goes. I'm leaving the design open so I could retro fit a water tray if that ends up being the better option.

What did you go for on the air drier? I’ve seen people use the type with absorbent granules inside which change colour.

OK that’s good to know. Hopefully still works the same.

Agreed and the idea of that was to not create a mist. It is similar to the fogbuster in principle but I find it a bit fiddly to set each time to get it mist free.

Always good to have your unsolicited advice Matt! I did consider running something more like that test and bought the anemometer to also add some measure of flow rate but decided I'd just go for how long you can run a tool for. I like your ideas though as it goes much further than my original plan. One thing I wondered was how fair it was because the hose attached to the compressor which went to the valve (for finding the balance point of pressure as you describe) would have a certain inner diameter which would also be a variable. Although I could make sure the pipe was larger than anything on the compressor. I have noticed the motor does get hot with prolonged running, not worryingly so but it does heat up. Anyway I like your thinking because flow rate at a given pressure is probably useful to match to the tool you want to use. I think this is because the max pressure spec is typically the max pressure at low/zero flow and the max flow spec is most likely at very low pressure. In other words 2 ends of the performance curve which are not that helpful since real use is in the middle of that plot. I'll think about something flow related, maybe one video for the future sometime, who knows!

​@@routercnc9517 Thanks for your reply, the thing to remember is that when the motor is running and the tank is not at pressure, essentially the motor is free spooling only having to overcome the internal friction of the drivetrain and moving the pistons etc. Only once the pressure builds up does the motor actually get loaded and have to produce any meaningful torque. It would be like testing the performance of your car by revving it in neutral, it is a pointless test that doesn't reveal anything. You can use compressor fill time calculators to get close to a representation of output, but in my opinion that isn't really relevant, when you are using air tools, especially like ones that use pneumatic motors, your compressor is going to be running pretty much all the time and you need to be able to maintain it at working pressure. And for the motor with the temperature increase, the resistance increases also, which reduces the power and torque of the motor, this along with having to compress hot air makes a surprising difference to efficiency. With my compressor in my home shop, I have 6 small compressors that are controlled with an arduino, they measure the rate of decrease of the air pressure and using something similar to a PID loop it switches on more compressors to compensate, there is also temperature sensors on the compressor bodies so the hottest compressor is rested, and the code makes sure they all get a similar amount of run time, was a great project, learned a lot! Pretty interesting subject really. Thanks for your videos, always enjoy them!

That’s an unusual setup- sounds like someone needs to start their own channel!

@@routercnc9517 Too camera shy!

Yes, it won’t run a sander so time to save up a bit more.

I think I saw some examples like that. Wouldn’t work for me as outside the workshop wall is the neighbour’s wall so it would be annoying for them.

I’ve thought about IPA but never got round to trying it.

Tempting but a fair bit of work in quill/spindle. Also the knee is only supported on 2 linear carriages so not very stiff for lateral forces.

Ah nvm, I think I get it now. If you had a 9V ac/ac adaptor it would essentially be an LDR in series from the adaptor going to the Harvi I think. Suppose it’s possible the resistor divider is also helping in some other way - sounded like it was very sensitive to alterations here.

Essentially yes. The resistor divider was needed to drop the AC voltage to a level the Harvi needed. They were 1 Watt because they needed to allow enough current to flow as that is what the Harvi needs to sense. They get warm but not hot.

I couldn’t find an AC /AC adapter which output a low enough voltage. Otherwise they would not be required.

@@routercnc9517 Thank you for the elaboration, I’ll do some experimentation!

OK, happy viewing!

Do you mean on the CNC machine? That motion control software is called UCCNC. I installed in around 2021 and the manual shows it to be software version: 1.2111. If you mean the software version for the microprocessor code uploader (Visual Studio??) into the TI board as part of this ELS project I used the exact same version James showed in one of his videos. I tried the current version at the time when I built mine and I could not get it to work. So I re-watched his videos and downloaded a the same legacy version as James and that worked fine. Of course because he made this a few years earlier that was the latest version for him a the time. It was too long ago and I can't remember the exact version but watch his videos and you will see it when he installs the code developer software and hooks it up to the microprocessor. If you meant something else, please let me know.

@@routercnc9517 I meant the Clough42 ELS. I wonder if my issue is the same as yours. I will try the earlier release. Did you report the issue?

Ok, ELS version 1.4. The other issue was with Microsoft Visual Studio and no I did not report it to Microsoft

@@routercnc9517 Thanks for the info. Troubleshooting is always a process of elimination.

Ooh! Yes that’s a great idea. Software backlash compensation on change of direction.thanks!

Yes they had a light coat of grease. Not much as they would attract dirt and maybe swarf.

Don't encourage that other guy!

Thank you, that is very kind of you. I'll just keep on doing what I'm doing!

OK thanks. I’ll see what I can do!

It certainly used to be!

Thanks. I've used lots of ferrules since this project was uploaded and tend to do that now.

If you mean tips for epoxy for self-leveling, then try not to mix any air bubbles in. One way is to mix it in one jug, then open a hole near the bottom and let it flow carefully into another jug. Then pour from the second jug onto the frame. This is to minimise air bubbles because these will upset the final surface. Or invest in a vacuum chamber to release the bubbles that way.

Many years later and no issues at all

OK, go for it!

Yes, very little left of the original drill - just how I like it!

I looked into a range of printers while researching and the new Bambu Lab printers seemed to be at the next level according to reviews. So then it was looking through the range - A1 mini, A1, P1P, P1S, or X1C. Mini was a bit small (180x180?), P1P was same as the P1S but w/o enclosure, and X1C had the same basic mechanics as the X1C but was considerably more expensive. I think the X1C with AMS was getting close to £2k, whereas the P1S with AMS was £900. When I asked my colleagues they said what do you want to do with it? Well as a beginner that is a difficult question because you don't know what is on offer and what could be done. So that's why in the end I went with the P1S because it was enclosed, so could print more materials, had a good size print area, had the AMS to allow multi colour prints (do I need that!?, yes is the answer it turns out) and I could buy this model plus lots of filament as still be much cheaper than X1C with AMS. I think the X1C had a few more sensors, and had hardened nozzles etc. but I think the P1S could be upgraded for not much money to have some of those parts. So on reflection I have no regrets on this model. It seems to do so much for you, prepping the nozzle, cleaning it, purging, auto bed levelling, that for the most part you can send a print and walk away. You can monitor progress on your phone app, and you get a message when it is done. The only times it has lost adhesion was when the part was getting quite large (fixed with adding a brim, and +5 deg on the bed), or the part was almost up to the edge of the build plate because it was a big part (fixed by printing 2 small pieces and clipping them together). It is also very fast. It is my first ever 3D printer so you will need to research a bit but I would buy it again without hesitation. I've even considered the A1/mini as a cheaper side kick to speed up iteration of parts, but I think I'll not get that past finance!

I think you are spot on with all those comments. For anyone that's interest Bambulab are having a sale until the 5 July. The p1s with the ams is 140£ off usual list

Hi Lewis, OK sounds good. I've had lots of suggestions over using different libraries for stepper control and there are quite a few. I did wonder if they take up more clock cycles to branch over to and in the end because I enjoy tinkering I hard coded the pulses, so good to know they run separately. As you can see it runs plenty fast enough though and I'll probably stick with it as it is working. But for future projects I'll look into the tone function. I wondered about the deceleration so I guess it could loose steps. On a CNC machine that would be a problem but maybe not too big a deal here as when running high speed you are probably just getting somewhere near you need to be. The knee DRO can track where you are in absolute terms. It might be a bit more sympathetic to the mechanicals though and stop the grub screw coming loose. I think 3 lines of code should do it . . . ?

@@routercnc9517 yeah a ramp down will definitely be more mechanically sensitive. Not sure how you are running your ramp up at the moment, but my solution was to use a while loop. While currentTone<targetTone; currentTone++, tone(currentTone, stepperPin);. Don't quote me in that, but it's close enough to give you the jist. In that case targetTone is the rpm to want converted to Hz, at the start currebtTone is zero, and then as the whole loop runs, the currentTone valuenisnincreaded.by one each time until it reaches the desired value. If that ramp is too slow, you can either increase by two or three or four each time, or divide the target tone by a set vale and use that, which then gives you a ramp over the same time period regardless of end speed i.e. the higher the target speed the more aggressive the ramp. Oh, the above is for ACC, just to the reverse for deceleration with the appropriate tweaks. The above code would also need to ensure that you are using interrupts for your inputs and once the ramp starts it won't stop until it has exited the while loop.

The interrupts fix that issue and are good practice anyway

I've come so close to getting this to work with HC-06 bluetooth modules I may give it another go. These Nanos are so cheap if I can make it work I can cover many such projects. However, there are other boards with WiFi and other systems already set up for SOTA so that is an option too. I'll also look into web based IDE as you mention. I do seem to like doing things myself (I learn more that way, but it can take longer!). Here is the history so far - I have managed to set up a Nano to act as serial host to talk to the HC-06 and I can use AT commands to talk to it. I used different pins (10 and 11) to keep the TX and RX free otherwise it does not work. There was a library which allowed data to be transmitted on these other pins. I then set the Baud rate from the default of 9600 to 115200 which is supposed to be the max for the Arduino but it stopped talking after that (I matched the Baud rate in the sketch to this value so it should have still worked). So I bought another HC-06 and left the default Baud rate at 9600 (I guess this means slow uploads, but let's get it working!), got it talking via the Nano using AT commands. Then plugged in the bluetooth USB into the computer and got it linked to the HC-06. It showed up as COM6 on my PC, which was a good start. Then I had to reconnect the TX and RX wires (which had been on pin 10 & 11) between the HC-06 and the Nano as we were now supposed to be sending data from the PC to the HC-06 and into the Nano via these proper channels. But IDE could not recognise the COM port and would not upload a sketch. Note this is HC-06 which is only a slave unit (not HC-05 which can be master and slave) and only has 4 pins, Vcc/GND/Tx/Rx. Feels like one step away but I'll come back to it another time since the software is working OK for now.

Hi MakerMadness. Welcome to the channel. Have a look around - I tend to get involved in long term projects so be prepared! Yes, happy the share the code. I noticed when editing the video that some of the variables were not used in the end, plus add-ins such as the EEPROM support (to save values when powered down). Also be aware it is flat code, no function calls, and does not make use of the many stepper type libraries out there, I wrote that code myself. As for Discord I've heard of it but have never looked into it. I don't have much time for many of the social media sites (videos take long enough!) but I have been a long time member of the 'mycncuk' forum. I'll put a link in the text box above to the long thread on this build series. If you check it in a few days I should have uploaded the code there as a zipped file. I'll post in this reply when I've done it. Hope that works for you. Thanks.

@@routercnc9517 would love to take a look at the code, do you have a avenue in which you may share it with us?

Hi MakerMadness OK, I've uploaded to the code shown in this video to the mycncuk forum. See link in the text description box above. Or search for 'mycncuk metalworking pillar drill experiences' and scroll to post #105.

I don’t own a milling machine. I have built my own CNC router though. Current project is a heavily modified drill press but it cannot be used for milling. You would get a good response from the mycncuk forum. Lots more mill and VMC users. Thanks

@@routercnc9517 thank you :)

@@ASW4567 If YT will let me put a link here the forum is here:www.mycncuk.com/index.php It can take a bit of time to get through the moderation and be able to post. If you get stuck give me a reply here and I'll nudge the site owner.

Thanks. I think shielded cable would have been my next step on more of the cabling if the capacitor had not worked (the stepper motor cable was already shielded to PE via the lower panel, but other cables were not). I do have some ferrites so I'll try those if it starts playing up in the future. For now the capacitor seems to be working.

I think I showed it in the video but yes there was a diode across the solenoid/electromagnet - it was at the power supply terminals. Yes the 24VDC circuit just goes from the power supply, through the relay contacts, and to the electromagnet, so separate in terms of the wiring at least. Looks like the capacitor sorted it anyway, thanks.

Thanks. It looks further on camera plus I was leaning out of the way to allow you to see the knee whilst I pressed the buttons. In practice it is fine for me to reach. Also saves me pressing it unintentionally and raising the knee into the workpiece.

Yes I have a storage case dedicated to wall fixings. Has brown, red, yellow plugs, 5, 6, 7mm masonary bits, small hammer (big one does not fit), AC voltage detector (to check for cables in the wall). So if I need to fix something to a wall it's all in there.

​@@routercnc9517I have something similar, everything ready in a compartmentalised box. However, I gave up using red plugs etc years ago. Now ONLY use various types of Fischer wall plugs. I know they can be 4 times the price, but I have never had the problems I used to get with those red and brown plugs. For the sake of a couple of pennies I know that I'm not going to be called back to a job because "Its come loose."

Yes, poor choice of wires for debugging! Only 24VDC used here for the electromagnet, but good advice if I was using mains.

OK thanks, I guessed there was a way to sort that. I'll do that next time.

I guess you could ask that about anything. Mountain walking, motor racing, horse riding. If it makes you happy then it's all good.

@@routercnc9517 lol! True story.

Thanks Pierre, good to see you still following along.

Still tempted. . .

Ok thanks for the feedback. The videos take a long time to edit so I probably miss how often I say something. I’ll watch out for that. Out of interest do you prefer just machining? Or just a quick intro and then jumping to finished parts? I’ve tended to edit as if you are following along and might do something similar. All feedback is welcome, thanks.