Bluemalu - Rostock MAX build

[Bluemalu]

Finally got some time for myself and sorting out the build.

Please be patient with my ramblings and metric measurements...

The kit I bought is from http://creativemachines.co.uk/store/ind ... er=product in the UK. It differs slightly from the US versions in the fact that the MDF is not coated in Melamine.

From various sources my options are:-

(1) Leave it as is and build. Unfortunately, the MDF is easily contaminated by oils, moisture and general use, the picture below shows it as it came from the packing.

(2) Seal the MDF. Choices available are:-
(a) Thin solution of PVA - adv. Cheap and cheerful, dis. Water content and a bit off hit and miss on quality of finish
(b) Rustins MDF clear sealer - adv. Solvent based and clear dis. None I can think of
(c) Acid Catalyst - 2 part lacquer /sealer - adv. The best solution (Morrells) dis. 2 part spray method - I don't have a HVLP gun
(d) MDF primers - adv. coloured dis. coloured, would cover the laser etching info and requires a top coat.

As maybe gathered, option (B) is the one I'm going for. So today is going to be sealing day.

One question for the experienced builders, having had a dry fit of the base unit parts, the width of the MDF in UK is 6mm, the slots are around 7mm, is this too much, should I be looking for some thin plastiic shimming strips / channel (like those used for edging strips on steel boxes?

See also:

http://forum.seemecnc.com/viewtopic.php?f=54&t=1392

Thanks for you time.

[Bluemalu]

Completed the coating yesterday, around 400ml of Rustins MDF sealer used. Took about 1 hr to apply each side, coated both sides and edges twice. It did raise the fibres but not too much, more like a matt varnish.

Now to start the build proper, having read through the v1.13 of the build manual and Mhackney's all looking good. Rather than cut the countersunk I used an inherited tool. Made light work of the MDF.

Fitting the feet went well, just for the pedants and fellow OCD suffers the plastic nuts do have a handing, there is a slight lip on the face towards the surface being clamped.

All going fine until a first fit up of the #68377 T slot Support and the #68352 Tri Support. Given how simple the system is how could I have failed at the first hurdle. The issue is the tolerances of the components. All of the sheeting has been cut from 6mm uncoated MDF, the slots cut into the sheets are 7mm, a 1mm gap seems rather large.

Given it's Easter Sunday, there's not much I can do, so I've thrown open the forum for advice, is a 1mm gap acceptable or should I look at installing some 0.4mm birch shims?

[foshon]

I don't know that anyone else had experienced the issue. I would think that you'll run into issues with the areas where the uprights mount. Have you contacted SEEMECNC and informed them of the issue?

[MorbidSlowBurn]

It sounds like the cut file was not adjusted for the lack of melamine. There may also be issues with the tri-supports. They "T" into each other so you may have issues with the alignment of the mounting of them to the top and bottom. I highly recommend contact seemecnc for their opinion.

[mhackney]

I recommend emailing SeeMeCNC and the vendor that cut your kit and explain what's going on.t

[Bluemalu]

The joys of B&Q on a Bank Holiday, to celebrate we had a good argument about magnolia in the paint aisle.

Anyhow, I've emailed my supplier and he seems happy with the demo unit he's built up, I'm awaiting a reply from SeeMeCNC head office.

From the little jaunt to the shops, 3 possible solutions came to hand. Original size of stub - #68352 - 24.6mm * 6mm

(1) Masking tape, couple of wraps round. New size: 25mm * 6.5mm

(2) 50mm * 6mm labels from Staples. New size: 24.8mm * 6.4mm

NB: If I started the tape before the corner rather than the long face the lshort edges would have a couple of layers rather than the one it currently has.
(3) Iron-On Edging tape - 21mm * 10m - B&Q - New size: 25.4mm * 6.8mm

Cut using a sliding guillotine to 6mm lengths and 6mm squares. Millie's hair straighteners for the long edges and a soldering iron with "plumbers" 6mm tip with a piece of paper to protect the melamine.

Choice number 3, though being long winded and time consuming, produces a nice snug fit with a small degree of slop.

[Bluemalu]

Many thanks for all the comments, I contacted Charl (creative machines) and SeeMeCNC's guru's (thanks for the swift reply from John Oly). The outcome was that the shims won't be required but also won't cause any issues if fitted. In the end I went with the melamine iron on shims. Applied with the judicious use of my wife's hair straighteners and some brown paper (very important as it stops the glue going on the straighteners and also stops the melamine from burning).

Spent most of Saturday watching the IPL (cricket) and cutting and mounting various 6mm shims.

After that a first fit up.

All was going well but I should have read further ahead as more cutting of shims was required.

After that the actual building of the base went well, the LCD and Electronics doors have been swapped around.
As yet I haven't cut the threads for the wrap around plastic as I don't have any 10/32 bolts and all my taps are metric. All ready to fit the top. I used blue masking tape to hold the nuts so I wouldn't forget to take it off.

After a little bit of moving and pressing, the top went on very smoothly, tightened up all of the joints and had a nice cup of tea. The actual building time was surprising quick. Time to read up on stage 2, the Onyx heater bed...

Looking back on the build of the first stage, I only needed to shim the long edges, on some of the larger pieces I applied 6mm*6mm pieces on the ends of the tabs (using the soldering iron and brown paper) but I think these are overkill.

[Bluemalu]

Been busy building for the past couple of days, no posts, I won't bore you with the well tread paths of the build covered by Geneb and McHackney.

The first minor change in my build was the T slot hardware, 2 sets were sent with the 2nd bag reading a cryptic "Use instead of kit supplied screws".

The square nuts do have a right and a wrong way, the curved side is inward, this is due to the fact that you can push the nut into the track without having to slide it through the track

For building the extruder the plastic gears were warmed in a bowl of hot water first (wouldn't be the first time plastic has cracked), the bearings were then pushed in using a rather Heath Robinson press, rather than the socket I used the large flat washer in the kit to avoid over pressing it.

Everything else went to plan until the fitting of the belts. Over a nice cup of tea and a home made panettone a less frustrating method was devised. This is a combination of Gene's and McHackney's with my own twist at the end.

(1) Feed the belt though the lower slot first (Section 18.6), keeping the belt (teeth inwards) in the T section, gently slide it down the T (two went smooth the last required a gently prod with a screwdriver).

(2) Once it's between the two lower rollers, route it through the outboard section of the T rail, round the top and back down.
(3) Raid the bathroom for some dental floss and cut a 5cm (2") section and super glue it to either side of the belt (I covered the first 3 teeth edges). With bonding to both edges of the belt a degree of wiggle is possible. To get the belt through, rather than pulling it, keep a touch of tension on the floss and gently move the carriage up the T section.

That seems like a good time to leave it and enjoy a well deserved Timothy Taylors.

[MorbidSlowBurn]

On your bearing insertion into the gears they are supposed to be recessed into the gear. When I first installed mine I went flush. This created friction and misalignment in the gears after assembly. I went back and used a socket to recess them and all the issues went away. Not sure if you checked to see if the gears ran smoothly yet or not.

[Bluemalu]

Thanks for the info, I used the flat washer just to recess the bearing (I've seen bearings over pressed (probably not possible in this case) using a socket). When I tried the gears one of the gears had a damaged tooth, I tried to reshape it and make it run smoother but in the end, ended up taking the whole unit to pieces to swap them over, now it's on the primary side it runs fine, although the proof will be in the printing.

[Bluemalu]

Too late to the party...

Finally, after what is nearly a year I've got round to completing the build of printer, it's been glowering at me from the corner of the room every time I passed it.

Went a bit berserk with the label printer and the nylon sleeving.

A KROY label printer, like most of the printers on the market, the printer isn't too expensive (£80) with the cartridges being about £25 each, I used them on a previous project so had a couple of sizes of heat shrink spare from that.

Sleeving - eBay

Started on the Onyx heat bed, as everyone was putting the component underneath it was the way to go.

As can be seen the Kapton tape makes an excellent stop for solder and silicon to stop it pooling.

I like the possibility of removing the hot end completely, allowing for other tools to be fitted (currently looking at a scanning probe similar to the one on my Modela MDX20), also, might help with finding out where the base is and if it's flat. I decided on banana plugs as I had some kicking around. However.....

Quick rethink.

Visited the local glass shop for a piece of 310mm toughened float glass, I think I may have over done it with 10mm glass. Glad they only had 4mm Aluminum. Cut a circle out 310mm in dia, used a hole in the centre and a trammel rail on the jigsaw. I tried the manufacturers recommended bit, didn't work too well (I had the speed wrong), chose a coarse cut and slowed the speed down to as low as it could go and cut away. Not as difficult as I expected. To get the hole for the snowflake correct I used the snowflake as the template.

Completed the wiring and now it's ready to see if we can free the smoke from the electronic components.

Time for a cup of tea, then start looking at power up and calibration.

[Eaglezsoar]

The sleeving makes for a nice, neat look.
Very good build. Many of them looks like a rat's nest inside (including mine),
yours looks fantastic. Good job!

[Bluemalu]

Cup of tea over,

Downloaded the latest version of Reptier Host V0.95F which has slic3r Version 1.0.0RC2, then Arduino 1.0.5-r2. So far so good.

Now to take advantage of all these, I need the firmware, but which one? I can find multiple repository with firmware in with Rostock MAX mentioned.

My hardware is a RAMBO pcb 0.9F that I suspect I can call a RAMBO v1.0 board, the LCD is supplied but as yet not fitted as I ordered the wrong interface PCB (ramps no RAMBO) so that's on it's way.

Checked the teeth, definitely 15, and with the RAMBO v1.0 will be 8 micro steps.

I went for the Repetier-MAX-Dev-Master firmware having looked around the forum, before I plough headlong into calibrating with this, could someone please confirm I'm on the right path.

[Bluemalu]

Thanks Eaglesoar.

The picture with the teeth wasn't zoomed in, if you squint or zoom in using paint, I wrote down the numbers as I couldn't remember how many there were and I wasn't going to strip it all down!

[Eaglezsoar]

Cup of tea over,

Downloaded the latest version of Reptier Host V0.95F which has slic3r Version 1.0.0RC2, then Arduino 1.0.5-r2. So far so good.

Now to take advantage of all these, I need the firmware, but which one? I can find multiple repository with firmware in with Rostock MAX mentioned.

My hardware is a RAMBO pcb 0.9F that I suspect I can call a RAMBO v1.0 board, the LCD is supplied but as yet not fitted as I ordered the wrong interface PCB (ramps no RAMBO) so that's on it's way.

Checked the teeth, definitely 15, and with the RAMBO v1.0 will be 8 micro steps.

I went for the Repetier-MAX-Dev-Master firmware having looked around the forum, before I plough headlong into calibrating with this, could someone please confirm I'm on the right path.

The latest firmware is Repetier 0.91 and is available here: https://github.com/seemecnc/Repetier-091-ROSTOCKMAX" onclick="window.open(this.href);return false;
Obviously you are going to have to change it so it knows you have 8x microstepping and 15 tooth pulleys and the steps per mm will need changed to whatever you are using now.
After installing you will need to use the assembly manual to do a complete calibration.

[Bluemalu]

Many thanks, thought I might have the wrong one, hopefully the calibration I've already carried out won't be wasted, would I be correct in saying I only need to copy the relevant points from the configuration.h file?

Quickly figuring out I might need to up the voltage for the heat bed if I want to use ABS, it's crawling it's way to 100 degrees.

[Eaglezsoar]

Many thanks, thought I might have the wrong one, hopefully the calibration I've already carried out won't be wasted, would I be correct in saying I only need to copy the relevant points from the configuration.h file?

Quickly figuring out I might need to up the voltage for the heat bed if I want to use ABS, it's crawling it's way to 100 degrees.

You will need to do a complete calibration after the installation of 0.91
If you put a towel over the bed it might crawl a little faster. Some have had good results doing that.

[Bluemalu]

Setup and Calibration.

Spent the last couple of days reading up on the various setup and calibration methods and also finding my way round the arduino software and configuration.h

So, checked the base of the tower for rotation (piece of paper on heatbed and project lines using a straight edge from each tower to to the centre), using the protractor and the sketch in the configuration.h found that all were as they should be.

#define DELTA_ALPHA_A 210 /** Equivalent of X Tower */
#define DELTA_ALPHA_B 330 /** Equivalent of Y Tower */
#define DELTA_ALPHA_C 90 /** Equivalent of Z Tower */

Checked the towers for being vertical, used an inclinometer on each face of each tower and averaged the difference to get a final result. (Inside face then working counter clockwise, so in theory 1st/3rd and 2nd/4th should be identical)

Tower X 90.1, 90.2, 89.8, 89.8
Tower Y 90.0, 89.7, 90.0, 90.2
Tower Z 89.7, 90.2, 90.0, 89.7

Interesting, looks like the T section is not quite as true as it should be. I'd like to say the values are "good enough for government work" but with no experience of the minutiae of the software and hardware I'd like a second opinion from the more experienced punters.

Next up was confirming the delta calibration settings, this is definitely one thing I could have done during the build just to confirm the measurements. The rods were 269mm each, a nice easy one to start with.

My dilemma comes with the CARRIAGE_HORIZONTAL_OFFSET (35.0mm default), where is the datum for the tower, is it either;-
(1) Centre of tower (39.0mm as measured)
(2) Centre of the pulley system ie offset from the centre of the tower inwards by 3mm => 36.0mm

Not thought of a good method to measure the END_EFFECTOR_HORIZONTAL_OFFSET (33.0mm default)

PRINTER_RADIUS (198.25mm default), from the Rev 2 build log, I know I need to use this to get the centre correct, having gone through the procedure for a couple of hours and not seeing much improvement (currently at 195.75mm), I decided to start this dimension check.

Hotbed - Once the towers are vertical, I am going to set the E-Stop screws all to an identical length, then commence "levelling up the heatbed". As I've got an aluminium heat plate on top of the Onyx, a couple of questions for others.
(1) Any issue in using IC heat sink compound between the plates (I seem to have a glut from previous PC builds)?
(2) Should I tighten the screws down once it's up to temperature to reduce thermal creep?

Unfortunately, my time at home has come to end so I'm off to work on Thursday so if anyone had any interesting article for calibration of a delta printer it will give me something to think about during the next two weeks. Hopefully, by then the PLA will have arrived and my last excuse to make a nest will be gone.

[Batteau62]

Many thanks, thought I might have the wrong one, hopefully the calibration I've already carried out won't be wasted, would I be correct in saying I only need to copy the relevant points from the configuration.h file?

Quickly figuring out I might need to up the voltage for the heat bed if I want to use ABS, it's crawling it's way to 100 degrees.

You will need to do a complete calibration after the installation of 0.91
If you put a towel over the bed it might crawl a little faster. Some have had good results doing that.

This cake pan trick has been working pretty well. http://forum.seemecnc.com/viewtopic.php ... 572#p26572" onclick="window.open(this.href);return false;
I'm not running an aluminum heat dissapator or 24V, but I think restricting the amount of ambient air you have to heat up really helps.

[Bluemalu]

Thanks for the pan trick, my silly question for today (even after googling it), which way up, does the tin enclose the whole heatbed assembly?

[Batteau62]

No such thing as a silly questions, just silly answers And Yes, the pan encloses the whole heat bed assembly. So for clarity, the pan is upside down, with the cork not showing. In my pic it is wrong because I was showing the cork insulation.

[Bluemalu]

Back home after a mobilisation of a ship in France, it gave me plenty of time to 'study form' about getting my Rostock calibrated. First mistake I had made was with the EPROM and not setting the mode to zero to adjust the radius.

Yesterday I had a lovely evening with my engineers square, vernier calipers, micrometer and some rather nice Ledin calipers checking all the dimensions for square and accurate. From it all I noticed the distance for ny carriage horizontal offset was out by about 3mm from my actual setting, also when I checked the pivot points on the cheapskates they were all slightly different so a bit of tightening and slacking of the screws bought them all into alignment.

Set the EPROM to mode 0 and checked within the Reptier software that the EPROM wasn't available and spent the next two hours sorting out the end stops and radius. After all that managed to get it repeatable to 0.05mm on my feeler gauges across the hotbed which I hope is acceptable.

I should have stopped there but being so close I decided to load up with some 1.75mm PLA and try the extruder. I have the original Steve's Extruder so the extruder steps per mm = 292, ran out 100mm and was pleasantly surprised to find it was around 100mm. Still on a roll I decide to extrude for the first time, set the temp to 190 degrees C and jammed it all up. Decided to quite whilst I was ahead and read up on setting up of the extruder.

Couple issues I have is I don't know (can't remember) what diameter extruder nozzle I have. I think it's 0.35mm (?) as the PLA I managed to extrude this morning measures 0.5mm.
Is the supplied tube actually PTFE? It seems a touch 'tacky' for PTFE and very clear rather than cloudy like most of the PTFE tube I've seen.

So going to have a read through the archives for the various methods for setup of the extruder for PLA, I think various temperatures to find the lowest usable, also, need to fit a fan (I was going to use ABS but need to improve my bed PSU).

Speaking of which, has anyone tried a 24V transformer with a full wave bridge rectifier to drive the Onyx.

[Eaglezsoar]

Speaking of which, has anyone tried a 24V transformer with a full wave bridge rectifier to drive the Onyx.

The transformer would have to rather beefy, at 24V the Onyx uses 17amps. Those of us that have done it have used a SSR whose input
comes from the Onyx heatbed output. There is a 30amp 24vdc power supply on Ebay that most of us have used.
So to answer your question, yes it will work but the transformer would cost as much as the power supply I am talking about. If you
don't have a transformer that can supply around 20 amps you are much better off using a switched power supply that can.

[Bluemalu]

Hmm, transformer is out, I don't particularly like switched mode PSU's but as I have to buy it rather than someone else I may have to rethink.

To bring the hotplate into calibration does anyone know the settings. I took a guess and put in a large THERM3_BETA but it doesn't appear to make any difference, also, why do thermistors start at 1 but extruders start at 0?

//#define USE_GENERIC_THERMISTORTABLE_3
#define GENERIC_THERM3_T0 25 // 170
#define GENERIC_THERM3_R0 100000 // 1042.7
#define GENERIC_THERM3_BETA 5000 // 4036
#define GENERIC_THERM3_MIN_TEMP -20
#define GENERIC_THERM3_MAX_TEMP 300
#define GENERIC_THERM3_R1 0
#define GENERIC_THERM3_R2 4700

edit: 50-50 chance, should have chosen GENERIC_THERM2_BETA, put in 10 and uploaded, hotbed temp was -20 (min value). So to calibrate GENERIC_THERM2_BETA is modified.

[Eaglezsoar]

Hmm, transformer is out, I don't particularly like switched mode PSU's but as I have to buy it rather than someone else I may have to rethink.

To bring the hotplate into calibration does anyone know the settings. I took a guess and put in a large THERM3_BETA but it doesn't appear to make any difference, also, why do thermistors start at 1 but extruders start at 0?

//#define USE_GENERIC_THERMISTORTABLE_3
#define GENERIC_THERM3_T0 25 // 170
#define GENERIC_THERM3_R0 100000 // 1042.7
#define GENERIC_THERM3_BETA 5000 // 4036
#define GENERIC_THERM3_MIN_TEMP -20
#define GENERIC_THERM3_MAX_TEMP 300
#define GENERIC_THERM3_R1 0
#define GENERIC_THERM3_R2 4700

Enter M303 P1 S100 to start the Pid autotune for the heated bed. The S100 tells it the tuned temperature is 100c

I don't know the answer to your second question, it's just the way the programmers decided to do it.