DIY magnetic DRO

[ChrisBParticipant @chrisb35596]

I have been wanting to buy a couple for my mill and lathe for quite some time now, specially for my mill as it can get quite frustrating and confusing trying to keep in mind backlash, endmill diameter, cutting direction, the small handwheel scales etc.

I got a couple of quotes for a 2 axis and a 3 axis Dro sets with magnetic scales, but these cost roughly £1000 and to be honest I was put off for a while…Lately I have seen quite some posts on DROs, which re-ignited my interest in the subject.

I thought what if I could do something myself, at a fraction of the cost? I want magnetic scales as these are the smallest most compact there is and the least sensitive to dirt and oil, while offering very good levels of precision.

So here is my planned project:

I placed an order for five pcb encoders: **LINK**

and a pair of scales drivers from TouchDRO: **LINK**

I also purchased some miscellaneous items such as blank pcb boards and etching material, pins and sockets and wire, etc. I estimate the total cost to be not more than £400.

Will report on the progress and if it works or no…hope it does!

[ChrisBParticipant @chrisb35596]

Building my own

[Nick HulmeParticipant @nickhulme30114]

Quotes from where? –

**LINK**

[ChrisBParticipant @chrisb35596]

Posted by Nick Hulme on 30/04/2019 22:26:27:

Quotes from where? –

**LINK**

Yes Nick, I saw those, they are even more expensive at £1200 for a 2 and 3 axis sets. I was looking at EMS-i, price wise they are just under £1000.

[ChrisBParticipant @chrisb35596]

So until I get the encoders which are on order I'll start with the preparation. I need to print/etch a small board to mount the encoder on as I'm never going to be able to solder wires to the encoder chip (the contacts are less than 1mm apart!) Then I'll mill the encoder housing out of aluminium and try to make all fit inside.

Started off with some drawings to get a good I idea of how things will fit in.

Overall read head dimensions will be 25x20x7mm, compact size so it will not get in the way and hopefully makes installation easier.

Internals, encoder inside the read head,

The boards I need to build and the encoders which are on order.

I have a day off today (my birthday!) so I'll get on with these boards this morning – it's an experiment as I never did a board before but youtube and the internet is very helpful!

I'm using a photoresist pcb board, developer and a homebrew ferric chloride etchant… let's see how it goes!

[Les Jones 1Participant @lesjones1]

Hi Chris,
I have not noticed any mention of the magnetic strip to work with the encoders. You will also need to make a guide system to keep the encoders the correct distance from the magnetic strip.

Les.

[ChrisBParticipant @chrisb35596]

Hi Les, well noted. The encoder comes with a self adhesive magnetic strip and a protective stainless steel cover strip. It's about 1.6mm thick, the ride height is 0.1 to 0.7mm so there's a bit to play with. I designed the read head to get approx 0.5mm ride height. If you look closely at the first picture, you'll notice the read head sides extend beyond the magnetic tape, that is what will give me the 0.5mm gap.

I ordered the magnetic tape is over sized so I can cut to length as required.

[Colin HeseltineParticipant @colinheseltine48622]

Chris,

Having Looked at the link you supplied where do you get the £1000-£1200 from. The three axis M-DRO set is only £520 complete.

I have used the Yuriystoys Touch-DRO software with M-DRO magnetic read heads on my Cowells mill and lathe.

Colin

[ChrisBParticipant @chrisb35596]

Hi Colin, I need two DRO sets, a two axis DRO for the lathe **LINK** and a three axis for the mill **LINK**

If I had to get the read heads only, the cheapest they have is £140 each and those are 6" long I need longer ones. I'm sure the 5 individual read heads will add up close to the £1000 mark.

[ChrisBParticipant @chrisb35596]

Works in progress: The PCBs are developed and being etched, so far so good!

[Colin HeseltineParticipant @colinheseltine48622]

Chris,

I had not appreciated you required two sets of readouts.

Colin

[ChrisBParticipant @chrisb35596]

Posted by Colin Heseltine on 01/05/2019 11:24:12:

Chris,

I had not appreciated you required two sets of readouts.

Colin

Thought so Colin, no problem.

The circuit boards have etched nicely, I will need to trim them to size and drill 1mm holes at the 5 solder points for the wires. Not sure how to cut them to make them identical…might take them on the mill rather than sawing them off.

[Ian PParticipant @ianp]

I see there are instructions on the encoder data sheet that refer to soldering it immediately after unpacking it because it is moisture sensitive. I am curious why its not then moisture sensitive after soldering.

What was the actual part number of the RLC2HD encoder you actually ordered and how much do they and the magnetic strip cost?

Ian P

[Anonymous]

Posted by Ian P on 02/05/2019 13:00:10:

I see there are instructions on the encoder data sheet that refer to soldering it immediately after unpacking it because it is moisture sensitive. I am curious why its not then moisture sensitive after soldering.

It's a production issue. The plastic encapsulation absorbs some moisture. Soldering on commercially produced boards will be done with a reflow oven, where the board with placed components is heated up to soldering temperature and back down again over a few tens to hundreds of seconds using infrared lamps. If the encapsulation has absorbed water it may well boil, cracking the case. It's not an issue for hand soldering, nor after the part is soldered down.

Andrew

[Ian PParticipant @ianp]

Well explained Andrew. having been responsible for getting boards manufactured in a previous existance it is something I should have remembered, although they were mostly flow soldered.

Ian P

[ChrisBParticipant @chrisb35596]

As Andrew well explained, it's for production manufacturing, I'll be hand soldering so I don't think I'll have any major issues there. We'll see when it comes to soldering those miniature joints what difficulties will arise…got low temp lead solder which should help a bit.

The encoders I ordered are all the same and cost €36 and the magnetic tape around €50 per m

Below is the spec I ordered, you can get it with different resolution connectors etc

RLC2HD miniature linear PCB level incremental magnetic encoder

Product description

[HD] – Incremental, no line driver, 5 V [A] – N/A (standard) [2D0] – 2000 (1 µm) [B] – 0.5 μs (2 MHz) [00] – No connector [C] – Periodic reference as per scale pitch (2 mm) [00] – None (standard) [] – [B] – ±40 μm/m [M600] – Length in mm (less than 1 m) [B] – VHB back-adhesion tape, with cover foil [0000] – No reference mark [] –

Product part number

RLC2HDA2D0B00C00

[ChrisBParticipant @chrisb35596]

Some more progress today. Milled the encoder housings out of aluminium, all ready for mounting the electronics. The encoders should arrive over the weekend, but will have to wait a bit longer to try them out and see if they actually work as the TouchDROs have to come from all the way from the States – might take a while.

First I machined the read heads in to the final outside dimensions and then made a fixture to clamp them.

This is the fixture held in a precision milling vice.

Read head blank clamped on the fixture ready for machining. Clamping action is very simple, I tighten the screw, in turn the countersink pushes a "jaw" away which pushes against the work piece. I could have clamped the piece directly on the milling vice but was afraid the part would distort. Besides that, with this fixture I could get a good degree of repeatability ( had to make 5 pieces)

First read head ready, four to go!

Almost done.

Ready. Also machined the covers which will also act as a guide to keep a gap of 0.5mm from the readhead to the magnetic tape.

[Ian PParticipant @ianp]

Impressive progress in a few days Chris. You have got me tempted to go down the same (similar) route. My mill has three separate iGaging type scales and I think I will replace the three displays with TouchDRO running on a tablet. Once that's up and running I can replace the scales later.

How long can the cable be from the encoders? I see the description states 'no line driver' so presumably would not like too much capacitance. What type of cable will you use (and protect and anchor it to the encoder housings)?

Ian P

[ChrisBParticipant @chrisb35596]

Hi Ian, you can wait till I get my TouchDRO to make sure it works, I'll be the guinea pig!

I'm using a 5 core felxible shielded cable like this: **LINK**

I the diameter on the housing is a close fit to the cable so that will slide in a tight fit. Then I'm filling the cavity inside the readhead with sealant so that will seal and keep everything together. Cable length is approx 1m, can be much shorter actually as the TouchDRO can be mounted on the saddle (for the lathe) and transmit to the tablet via bluetooth. But I kept the lengths standard so I can move them around and swap if I need to.

I could have used armoured cable but to be honest I think it would be an overkill, I mean I'm not in an industrial environment and with some care and proper routing the cables will be protected – It is chemical resistant so oil and stuff like that should not affect it much.

Next step is mounting the encoder, left Slovenia on Wednesday, so I hope I get them tomorrow, if not it will be next week

[Ian PParticipant @ianp]

I am not in any rush but I have no doubts that what you are doing is going to work. A few minutes ago I installed the TouchDRO app on my phone just so I could see what it looked like. Obviously not anything I can use yet but can see it will be a big improvement on three separate displays (and on the two axis Sino on my lathe).

The cable you have selected, being polyurethane jacketed is pretty tough stuff and impervious to most things in the workshop.

Ian P

[ChrisBParticipant @chrisb35596]

Some more progress : received the encoders over the weekend, so next step was to solder the encoder to the pcb and then install everything on the readhead casing.

They are indeed small and was a bit concerned I might not be able to solder them as the contact points are really close to eachother, but I managed. I checked continuity and all seem to be fine, the the proof of the pudding will come when I receive the touchDRO, so fingers crossed!

I anodised the aluminium read head casings with chromic acid, the material is 7075 and has a tendency to corrode although it's probably unlikely given it will live in an oily environment.

Next step is to seal the encoder inside the readhead casing, I'm split between resin or sealant, not sure which way to go. I would prefer resin as it will become one piece. On the other hand with resin, once cured there's no way I will open the thing again if I need to.

[Ian PParticipant @ianp]

I would wait before your 'next step' so that you can test everything works before potting.

Are you connecting the cable screen at both ends? There is a case for connecting it only at the 'input' end of the wiring (the TouchDRO PCB connector) but you only seem to have 4 pins.

Ian

[ChrisBParticipant @chrisb35596]

Not sure if I did this right, I connected the shield to the read head casing, the casing will slide and contact the lathe and mill body (which are grounded) so I'm thinking I should have a ground, did I get it right?

In reality I don't know if cable shielding will be necessary as there will be no other electrical wires routed along the read head cable, so I would think noise should not be an issue – but I might be mistaken.

Good advice re. waiting before sealing everything…hope the touchDro wont take long to arrive!

[Ian PParticipant @ianp]

Connecting the screen at both ends introduces the 'possibility' of the screen carrying current. Only connecting one end means that the screen is purely a screen. In sensitive audio or electronic measuring instrumentation it is usually only connected at the amplifier end.

Current flowing through a screen could be induced to the contained (signal) wires by capacitive coupling, in a perfect world no current would flow through a screen connected at both ends because the equipment at both ends should be at the same potential. In practice your lathe motor and lathe metalwork might be plugged into one 13A socket and the PSU for the DRO electronics might be in another socket so there might be a 'loop' of ground/earth wiring that acts as one turn of a transformer that could couple surges and spikes into the signal wires.

To be fair logic level signals in this DRO application are pretty immune to interference and there are a lot of 'could's' and 'possibly's' in my previous paragraphs, in practice grounding the sensor end will be OK. Not sealing everything up yet allows you to test and if necessary make changes.

Ian P

[BeakerParticipant @beaker]

I just wanted to bump this thread back up. Chris B: This is really interesting stuff, please keep us updated. I'll definitely be doing this if you're successful. Thanks for blazing the trail!

[Author]

Posts