Tidying Up An Analoy 1401 Alloy Analyser

[awemawson]

My recent resurrection of the Induction Furnace has encouraged me to dig out a pair of Alloy Analysers that I've had for ages - indeed in the Furnace thread there are examples of their use.

This has brought to my attention what a dreadful state they have got into - mainly due to the original case foam having degraded and bits crumbling all over the place. This isn't just cosmetic, it makes the instrument unpleasant to use and pack away, and physically difficult to get bits in and out.

So the plan is to entirely remove the instruments from their cases, give them a good clean up and replace the original lining with a modern neoprene closed cell foam - a square metre of which arrived by post this morning  There are also the usual age related frayed cables and general wear and tear to redress.

The green perspex 'arc shield' on one has broken - I remember supergluing it a few years back. Superglue scraped back to perspex and cover clamped together and gluing as I type, this time using a perspex glue - the same stuff that successfully repaired the lube oil reservoir on the little Mirac CNC Lathe.

Now I didn't think that I had much in the way of technical information on these instruments so have been in touch with the manufacturers - Arun Technology, where their Finance Director is very kindly trying to find out what he can. It seems my instruments date from the late 1980's and Arun have moved twice since then and their archive isn't digital. However he is putting me in touch with one of their long standing Service Engineers when he returns from holiday.

When I opened up the second instrument, not only did I find that I have an operating manual, but partial circuit diagrams and a spare EPROM !

So before I start pulling them apart have a few pictures:

[awemawson]

So the first thing to do is remove the contents of the case and clean off the original heavily degraded foam lining and it's adhesive.

The lid has a complicated bit of metalwork in it retained by four screws that makes various compartments to store accessories. - it must have been an expensive bit of metalwork to make - nowadays it would probably be a plastic injection moulding.

The foam is held on by double sided tape - I have yet to find a solvent that removes it satisfactorily without removing the underlying paint work. Zippo lighter fluid (refined petroleum) removes the remains of the foam but doesn't touch the tape. IPA doesn't touch either, I daren't use acetone which happily removes the adhesive as it's equally happily removing the paint  In the end I used Toluene, which softens the adhesive to a sticky mess that can then be scraped off, but it's not a very satisfactory process.

So the result of an afternoons work  is the case of instrument #1 cleaned off internally with the lid inner frame also cleaned, but there is one cavity in the electronics unit that is the storage for the probe that remains covered in the old foam. I had hoped to be able to dismantle the framing such that the foam covered bits could be removed to be cleaned but it turns out that the cheeks of the cavity are fundamental parts of the structure. The foam will have to be cleaned off mechanically as I don't want solvents and sticky gunge all over the electronics. But that's a job for tomorrow - I have a head full of Toluene vapour so time to finish.

As I thought, the electronics is built round an 8085 microprocessor and is all rather nicely made with 1989 dated test labels everywhere.

I will treat the probe head as a separate unit as indeed it is, as it just plugs in.

[Muzzerboy]

Yes, that is a product "of its time"! 1985 date codes on some of those old ICs, multicoloured ribbon cables, WIMA(?) film caps, TO-3 power transistors. Those were the days. As you say, seems to be from the late 80s judging by the date code on the micro.

I wonder how the sensor head works? Pretty sure you said it analysed the light spectrum but can't see any mention of it in the furnace thread now that I look - years of self abuse have clearly taken their toll on my eyesight.

[awemawson]

I think that the clever scanning bit is in the probe which I’ve never opened up, but I will be !

[vtsteam]

Andrew, you are a lucky man to own that level of equipment. 

[awemawson]

Years of hoarding, avarice, and scanning the small ads Steve 

So I couldn't leave it at that, and came back out to the workshop and cleaned the foam and glue off the probe storage chamber - horrid job but it's now done.

Then I realised that there are two miniature fans amongst the electronics - both with degraded foam filters - so those got removed and new filter material installed.

If you see the holes in the side of the probe chamber - I think that those were ventilation through the OPEN cell foam that was originally installed as padding so I now need to think how to provide the same ventilation but using my closed cell foam. I may just cut corresponding holes in it!

[awemawson]

The probe 'arc shield' had glued over night so it got re-fitted.

Today's job: start gluing neoprene foam on.

Should be fairly straightforward were it not for two issues. Firstly putting it in restricted spaces is tricky as the pre-glued sheet sticks to anything it touches, and secondly, like a prat I put the main instrument into the case to draw an outline where the base foam is needed, fixed the foam, then realised that I'd put the instrument in at 180 degrees  More foam removal and cleaning !

Eventually all back together so before I investigate the probe I wanted to prove that the main instrument still worked. I'm glad to say that it did and correctly identified 'LM2' as LM2 giving me confidence that I've not disturbed anything vital.

It was then that my  five 8.5 kg ingots of pure aluminium were delivered. Pleasingly the analyser agrees with the supplier that it's 'LM0'  so a nice confirmation that the results are reasonably accurate.

Irritatingly, having lined the 'probe pocket' in the case with what I thought was the same thickness (10 mm) foam I find that the probe no longer fits and I had to remove it - oh well you can't win 'em all !

So next item on the list is the probe itself - pull it apart and see how it ticks !

[awemawson]

So time to open up the probe head. Never been in here before - all I've previously done is make a repair to the umbilical cord using self amalgamating tape, and replace the window glass that protects the lens. This is no more than a lab slide 'cover glass'

So nothing ventured nothing gained - out with the Allen keys - carefully cut round the tamper seal - and see what we have.

Well  as expected, rather a lot of electronics. The emission from the arc shines through the cover window, passes through a lens and enters a 'long black box'. The far end of this box has a 16 pin ribbon cable. Now I assume that there is a prism and the emissions are spread out down the 'black box' and impinging on a sensor of some sort.

I strongly suspect that the sensor is a 'CCD' or charge coupled device as I found a patent by the company using those devices for spectral analysis . Or maybe they have a sensor dedicated to each element that they are interested in - I don't know, and I wasn't prepared to dismantle the box or I expect the calibration would be all over the place.

Anyway I thought it prudent to carefully re-assemble it before I break it. I replaced the optical window and tested that it still works - phew - it does !

[awemawson]

A bit more information has come to hand.

Studying the scraps of circuit diagrams that I have, which are incomplete and fuzzy, there is reference to the CCD !

Also the instruction book has a description of what can only be my 'black box' - glad I didn't open it !

[Muzzerboy]

Aha, so refraction to discern the different wavelengths (elements) and a CCD to place the position and amplitude of each.

Those electronics assemblies were the life's work of several people - many man years there!

So it was assembled and tested late 1990. Interesting / encouraging to see they are still in business. I wonder how their current products compare in terms of the technology in use.

[awemawson]

They've gone over to XRF analysis like everyone else, and from being a British company for decades I see from Companies House that their one Director is now Chinese.

The instrument that I've been pulling apart is amazingly well made and a work of art - I doubt that the bean counters were involved in it's development !

It's testament to the quality that it's still working at least 30 years on !

[vtsteam]

Andrew, what size and shape sample does it use? I don't have a good mental picture of how it is loaded for a test, or where the electrodes are in relation to the test sample. Can you take a pic of what it looks like ready to test a sample?

Oh, also, I just noticed that the "protective windows" seem to be standard microscope slide cover slips. Cool! 

[awemawson]

Steve I'm attaching a picture that may help. The 'heel' is rested on the sample and the graphite point is brought into contact to strike an arc.

So you can see it will work on quite a small  sample.

[vtsteam]

Thanks Andrew. I see. So you don't actually have to prepare a sample. This device is just placed against a metal object to do the analysis.

I've just read a really interesting explanation from 1922 of the then new science of quantitative spectrographic analysis, and since I know you enjoy reading this kind of stuff, too, here it is:

http://nvlpubs.nist.gov/nistpubs/ScientificPapers/nbsscientificpaper444vol18p235_A2b.pdf

[awemawson]

Thanks Steve,

You may have missed this thread that I posted some years ago about my analysers:

http://www.madmodder.net/index.php/topic,9244.msg102025.html#msg102025

[awemawson]

One of the issues that I find with the Analyser is that the sharply pointed graphite electrodes wear down very rapidly. The book say to use a new one for each test. Sometimes I can get away with using one electrode for two tests but not often. I think that the issue is that as the electrode wears down the discharge is no longer at the optical axis of the input lens.

It occurred to me that my other manual Analyser the Metascope uses a tungsten electrode that lasts a long time, so perhaps I can somehow create a suitable tungsten point. The graphite ones are  3 mm diameter and held in a tiny collet chuck, so I need to find a bit of 3 mm tungsten. Maybe a TIG rod. The Analyser limits the arc current to 6 amps and the voltage to 50V so I don’t think the difference in electrode resistance will be a problem

[philf]

Andrew,

I have several broken 3mm shank tungsten carbide end mills (or slot drills)

Would they work?

I could easily grind a point.

I guess you can't measure carbon or tungsten content of your material.

Phil.

[awemawson]

Phil thank you kindly yes that probably is a good way to go - PM incoming

(Picture attached of the  3 mm graphite ones to show the proportions - they are about 1/2" long)

[awemawson]

I came across this scholarly article on the use of pencil leads as electrodes for analysers - pencil leads were my first thought for replacement graphite points some years ago

[Sea.dog]

Have you read it all Andrew? 

[awemawson]

It's good to get you off to sleep, and the original in Romanian is even better 

Stick with the management summary and you'll be fine 

[awemawson]

Phil Fern very kindly hand crafted an EXCELLENT tungsten point for me that arrived in this mornings post allowing me to experiment.

Sadly it seems that carbon arcs and tungsten arcs are different animals. I cannot maintain the arc at all with the tungsten, whereas with the carbon you can strike it and draw it out to a few mm length. I'd be interested in the physics of this if anyone  has any knowledge that they'd be willing to share.

All is NOT lost though as this puts to bed a nagging thought that I've had for some years that this would be a possible way forwards. It isn't and I can put it in the 'shame but . . ' box and move on!

A big thanks to Phil Fern, you sir are a Gent 

[Muzzerboy]

Interesting, if disappointing.

Is the "carbon" electrode really just pure carbon? You'd imagine it might need to be although that would make it fairly brittle. That presumably makes material sourcing pretty difficult.

I know from my professional experiences that components such as motor brushes that are often referred to as "carbon" or "graphite" generally contain a range of constituents. In the case of brushes, this includes copper, ash, binding agents and other proprietary ingredients. Pity about the tungsten, as that is easy to procure in pure form.

[awemawson]

Murray, the genuine electrodes have a feel and consistency of a soft pencil. I've found a source of 3.1 mm '6B' pencil leads so when they arrive I'll experiment with those.

. . . it's amazing where these little project lead you !

[Muzzerboy]

You are indeed entering new spheres here. I see our artistic friends even use 8B and 9xxB pencils, although not generally in clutch / mechanical format. The 6B seems to be the best available in 3.1mm, presumably the Faber Castell brand. Probably rather brittle to be any smaller or softer.