Huanyang Inverter VFD NC Studio Settings

I have been putting up non vacuum stuff up here for the last few posts because I have nowhere else to put them! The vacuum system is in stasis at the moment as other more daytoday things has been taking up my time. So my apologies to you high vacuumers out there.

This post is motivated by a day and a half of frustration which finally got sorted through much googling, testing, reading and rereading and logic-ing out. My cnc machine is now armed and fully operational! I have no doubt that there are others in the same position as I was yesterday...

Be warned: I am not an engineer and don't really know what I'm talking about here, so use this info at your own risk!

How I learned to love my new Huanyang VFD spindle for NC Studio. This post is just about getting NC Studio talking to the VFD, not motor settings. There are other good tuts out there on that bit.
I'll assume you've done a bit of work on this already and get right into the nitty gritty.


Please read it carefully, every step is important.

**some changes you make may need the VFD to be restarted, it seems...**

**with NC Studio we are dealing with '4-speed external control'. You don't really get the values in your gcode. boooo!**

Speed1 being stopped/no rotation. Defined by PD003/PD011(see below)
Speed2 definded by PD0086
Speed3 definded by PD0087
Speed4 defined by PD0088

To begin!
0.PD000 set to 0.  Allows changes to happen in the settings...

1. PD013 set to 8. Factory reset (Thus may not be nessasary if you have a matching VFD and spindle.)

2. PD001 set to 1. This selects external terminal/multi-input as the signal input.(this over rides PD002[1] setting I assume)

3. PD003,PD004,PD005 ... this is motor specific settings I'm not covering here.

4. PD011 set to 0. Now this is the interesting one. This is going to supersceed PD003 that you would think is the defined speed for Speed1.

But there is an assumption that as long as [FOR]ward is enabled (which it is, because its permanently tied low, isn't it?)the motor is turning and doing something and so would burn out if the setting is too low. But in GCode the spindle is turned on by its speed setting ie: S18000 or what ever...so it's a cheat we programme into the VFD firmware to turn the spindle off with a zero speed setting.
(granted there is an Mcode for spindle on but with NC studio you just have 4 speed options on 3 pins. 000 being off, 100 being low, 010 being...etc.)

PD044 is set to 02. Sets FOR connector to enable forward rotation when taken low.
PD047 is set to 22.  Sets SPH connector to enable setting in PD0088 when taken high.
PD048 is set to 24. Sets SPM connector to enable setting in PD0087 when taken high
PD049 is set to 23. Sets SPL connector to enable setting in PD0086 when taken high

PD080 set to 2. That's the magic setting.

This allows signal from NS Studio to select the three(4) settings for speeds you will define below...
We've already looked at the PD003/PD011 issue so that is Speed1 ie: 0Hz
PD0086 is set to 133 for 8000rpm
PD0087 is set to 300 for 18000 rpm
PD088 is set to 400 for 24000rpm
I may change these settings depending how they work in practice.


If I'm wrong with what I've done here please let me know. If this has been a help please comment. If I made some typos please let me know.

ER11 drag engraver

This post is a little off topic so I'm sorry about that. It will be relevant to somebody I suppose.

spring loaded ER11 diamond drag engraver

Above is a spring loaded diamond drag engraver. It's 3mm with a 120 degree angle tip. The issue with it is it needs to be spring loaded to allow for uneven surfaces, but there is very little space to build a mechanism inside an ER11 size collet. The commercial ones work externally to the collet. My thinking was if it is in the collet it would be more ridged and more accurate.

Drag engraver apart

So here it is taken apart. An ER11 collet with a 7mm diameter.
A hard brass pipe(Albion Alloys, metric K +S!)
Polished 6mm steel shaft from an inkjet printer with a spring, I'm not sure where that came from!
I chopped the steel shaft with a hacksaw and stuck it in the lathe. I wrapped the shaft in masking tape to protect it from the jaws of the lathe, then drilled the centre to receive the diamond tip bit and trimmed the other end to take the spring. The spring is squeezed on. This is important for a couple of reasons. It ends up >6mm diameter therefor holding the shaft into the brass tube and so the collet.
That's morealess it. The extra lug at the end is just because I cut the main body of the shaft too short. The grey crap is PTFE/graphite dry lube.
To use...(engineers close your ears), I put it onto the spindle as usual then tighten the nut until the collet pinches the brass tube enough to stop the shaft moving, then loosen the nut a smidge. It should return when pushed in. So I'm using the collet slightly loose...

The State of Matter as it stands.

For SEO your supposed to add a caption! DIY Metalization.

A very quick update...this is the state of the system now. The steel belljar is working well. Sight glasses seems to be sealed ok too. After buying a very nicely priced butterfly valve.( Thanks to JF and his wife for the delivery!) I can now cut off the diffstack from the belljar. I got down to below the range of the pirani gauge. Oh yes, I'm sorry I really coped out. I bought a diff pump on Ebay...Interestingly the urge to get the DIY one is still strong in my mind. I do intend to revisit it in the near future. If only to get some closure!

Belljar Hoist Mount

 The jar is heavy so I will put it on a pulley and hoist eventually.

Vacuum Chamber Interior

And the interior..stainless steel M10 threaded bars tapped into the base plate.(gas escape grooves cut accross the threads that are in the tapped holes. To try and minimize virtual leaks.). A rolled aluminium ring 5x25mm. Soldered at the join (my first aluminium soldering. Very possible. Probably not a good idea for sealing... Must fly!

A big jar of nothing.

 

 There it is with the view ports welded in. I went for one to view filament and on to view object being coated. (I didn't do the welding, I thought, as it was critical I leave it to the expert!)

 There was some distortion when the welding happened so I had to sand the flanges flat again. I made that little sanding thing to help. It took a while but I got then flat enough for the o-rings to take up any of the remaining irregularity.

 

 The strangest thing was the bottom of the jar was no longer flat after the welding process. It shrank upwards towards the welded pipe. I was very surprised. It was across a section of at least 50mm. I spotted it when I did a test pump down and the bottom gasket was buzzing as air rushed in through the gap! Back to the counter top with sand paper as described before.

 Above is a view of the o-ring seal on view port. The o-ring are viton(€16 each woops). 6mm cross section. The glass is a square in 12mm, normal type. I just got a glazier to cut it for me.

So the tube sits proud of the flange by about 3mm. This keeps the o-ring in place. I cut the clamping rings from 3mm aluminium. Drilled the top one tapped the underside one. It is split as you can see from the side view.
Oh and I pumped it down today, its tight! Yipee!

Cheapo (metal) Bell Jar the beginnings...

After a long stretch of solid work I got a chance to get back to vacuum for a while.

I was getting a bit nervous of my glass belljar imploding so I decided to go for a metal one instead.

After seeing a guy on youtube who built a mildsteel box to coat telescope mirrors in I decided to give the gas cylinder a second try.

I bought a used gas cylinder 380mm diameter just right for my base plate at 400mm.

I let the remaining gas out and unscrewed the valve. Don't forget to fill it with water to displace the remaining gas before you go a cuttin'

 

I marked the line by wrapping a yoga at around the cylinder to get a parallel line. That wasnt a good idea... you need to use some thing not so stretchy, that defeats the purpose! I marked away and used the bottom end with the complete dome.

I put the cylinder on it side and cut it with the water still inside. It was pouring out all over the place. I trimmed much of the foot ring off too. I wanted to get rid of excess weight. It is heavy.

Once it was cut I dried it straight away. We don't want any rust forming, very bad for outgassing.
Then on to leveling. Even though I paid great attention to the cutting I still spent 3 hours leveling the cut to get it nice ond flat. For this process I spray glued strips of al'oxide paper to a scrap of countertop. Its nice and flat and very ridged. I did a couple of hours of 80grit and an hour with 120...

Very tiring on the arms! During the sanding I regularly ran a black marker along the edge so I could see my progress clearly. I would watch the colour sand off, and the black patches get smaller and smaller. Eventually I got down to a consistant edge. I went then, by hand with a finer sand paper to get rid of the 120 scratches. I rubbed along the edge insted of across it. If there are scratches left they are better not to be radial across the edge as it is more likely to leak in that case.

The whole interior had a dark crust on it. I don't know what it is, possibly passivated with phosphate? I diligentlly sanded this away. There was still a bad smell coming from the tank. This got less and less as the crustyness came off.
I cut a new flat rubber gasket and greased up and sure enough it pumped down to 40micron the limit of my pump....

A Little Closer!

This is the casating I wanted to try a bit of evaporation on to. Its made of fastcast polyurethane and then spray-painted with gloss black.As the Al coating has no filling qualities the gloss of the coating is a result of the texture of the base.

 

I picked up a few aluminium prefused tungesten filiments from a company called Midwest Tungsten in the US of A. This is a bit overkill for a DIYer but the were cheap, from their surplus bin!

 

So I hung the plastic part from a wire inside the bell and began to pump down...

When I got to 40micron or so I turned on the MOT power supply.(I wrote this up in the blog a good while ago)

After 20 seconds I got this! I was delighted! Just mechanical pump, no diff. Now, its not up to scratch yet but these little victories help me keep the faith.(Ignore the electrodes poking through, they are just sealing the holes here)

A strange thing at this vacuum is the fact that the back of the part got a covering of Al. as well. It must fly around in the belljar at these crappy vacuum levels.

 There is the front. Opaque but not bright. It has greyness in person. Getting there...

I found a leak!

After a long absence from my metalisation project I decided to get back to it. I was having a real nightmare with my belljar not getting the vacuum level I was expecting. Or that I needed to even start a diff pump. There were so may potential leaks I was overwhelmed for a while. So I decided to go step by step and eliminate the possibilities starting at the pump.
Flexible pipe tested fine, under water with compressed air. The second section however had a f**king leak. Right there, bubble, bubble, bubble. I couldn't believe it. The solder joint looked fine but it was leaking.
So I Glyptaled it under a vacuum and I was all fixed up!
More to follow...