Member postings for Andrew Johnston

Hi David,

If I read correctly you're recommending a silicon carbide wheel for grinding steel? I'm trying to surface grind some steel (low carbon, hot rolled) at the moment, and while the opposing faces feel flat and I cannot measure any difference in thickness there is definitely a fine, but regular pattern on the surface. I'd assumed that it was due to the wheel bearings being knackered, but might it be due to the wheel skipping? The wheel I'm using is white aluminium oxide, 46 grit, I can't read the rest of the description but I think it was medium hardness.

Tractor man: A single phase induction motor has a pulsating magnetic field, which may, or may not, affect the surface finish. A 3-phase motor does not have this pulsating effect. There are those who state that a single phase motor can affect the finish when using a lathe: I would have thought that a surface grinder would be more sensitive. I'm using a 3-phase motor, as I have a 3-phase supply. My grinder would have originally been supplied with a single motor driving both wheel and table feed (all mechanical) but had been converted at some time in the past to two single phase motors, one for each. The wheel drive motor smoked soon after I got the grinder, so I threw the whole lot away and reverted to a single 3-phase motor driving everything.

Regards,

Andrew

That's nothing, sometimes I lose the entire workshop; I've even had to ask for a new desk at work as the old one seems to have disappeared under papers, PCBs and boxes of parts.

Regards,

Andrew (Mr. Untidy)

Don't think you need anything fancy. In the linked thread EN3B is recommended:

http://www.tractiontalkforum.com/showthread.php?t=20177&highlight=studs

Regards,

Andrew

That might have been partly true in the first half of the 20th century, but it certainly wasn't the case in the second half. Both my uncles went to university in the 1960s, and they weren't rich; my grandfather was a building site labourer. However, I would agree that it's probably more the case now, thanks to tuition fees; that's socialism for you.

I can't answer the OP directly as I read electronics, and I expect the situation has changed since the 1970s. Personally I'd look for a course that deals with the fundamentals, rather than the latest whizzy stuff. The whizzy stuff gets outdated but the fundamentals don't change. Look at the Russell group of universities. The object of the exercise is to get the best class of degree from the best university possible. That's what's going to get you interviews, once you're in the interview it's up to you, but if you never get the interview you can't shine! It may be a culture shock but I think it's good to go to university away from home.

I wouldn't be frightened by Oxbridge, there are some mind bogglingly clever people there, but the vast majority are not. Personally I think the Cambridge Tripos is a bit too mathematical; I'm glad I didn't do it as an undergraduate. It was stressful enough having to work through it in order to supervise the undergrads!

I suspect that chartered status may be more relevant in mechanical engineering, but I've never found it to be a handicap not having it. I've never, ever, been asked about it in an interview, nor asked why I wasn't a member of the relevant professional body. That may be because I've always worked for fairly small companies, where it's ability and attitude that count.

Regards,

Andrew

PS: To save people having to read my profile I did my first degree at Liverpool, and my PhD at King's College, Cambridge. I'm just damn glad I went through the system when I did, not only were there no tuition fees I actually got paid to go. I did a thick sandwich course with MOD, and got a salary while I was an undergraduate, not the usual LEA grant. I've been very lucky.

I concur, at least county and country would be useful - Andrew

Ian: Correct, I had second thoughts about the words and related link before posting and cut down the detail. After posting I had second, second thoughts and deleted the link. While details of the product are freely available on the manufacturers' website; it occurred to me late in the day that they may not be too happy about it being stated that it wasn't designed in their home country.

I've used two methods for hand soldering fine pitch leaded ICs. Fine pitch means 0.65mm and below; wider than that and it's easy to solder individual legs. In both cases I start by tacking the IC in place by soldering opposite corner legs. At this stage I don't worry about shorts between legs. I used to then flood the IC with a liquid flux. Then I'd get a big blob of solder over 3 or 4 legs. If you get the blob molten you can then pull it along the row of pins; each leg gets soldered and surface tension pulls the blob along without shorting adjacent pins. That's all very messy though. So I now just solder the whole row of pins, shorts and all, and then remove the excess solder with 'solder wick'. All the above applies to tin/lead solder, all bets are off with lead free solder.

Regards,

Andrew

Hi Neil,

An accelerometer from Analog Devices? I used the ADXL321 a few years ago in a small unit designed to monitor vibration as a means of preventing 'Whitefinger'; nerve damage in the fingers caused by using vibrating machinery. It was a small ceramic package with all the pads underneath including one in the middle, a real PITA to solder by hand; sound familiar?

It seems to be an increasing trend to have a central pad underneath surface mount ICs. They're good from a mechanical point of view, and really help with thermal management and EMC on switchmode power supplies. But they're difficult to solder to by manual means. Unfortunately a lot of the pads now double as GND connections and must be connected.

Over the years I've developed my own techniques for soldering fine pitch SM ICs; what method do you use?

Regards,

Andrew

Edited By Andrew Johnston on 07/04/2012 12:11:08

Not any more; legs are so yesterday! I've just finished designing a fairly complex board and only one IC had a lead pitch of 1.27mm, and none at 2.54mm. The majority of ICs didn't have any legs, being either BGAs (ball grid array) or LCCs (leadless chip carriers). The only big ICs that had legs were the Ethernet MAC/PHY and Ethernet switch at, I think, 0.65mm, which is of course 25 thou. All the other ICs with legs were 0.5mm. The use of BGAs is pretty much universal now, and not just for the larger ICs.

BGAs are used for several reasons. For large numbers of connections the packages are much smaller than equivalent leaded components. For example the processor we used had 423 connections, and that's a fairly low number. Even at 0.5mm lead spacing that's a big component. A large leaded package is more expensive to make and requires stamped out lead frames. Also, a large leaded package has significant time delays and parasitics associated with leads in the corners of the package.

If you think components are an odd mix of units, then PCBs are worse. Thickness and size can be metric or imperial, or both, depending on the final customer. Via holes are normally metric, but track and gap design rules are often imperial. To cap it all the copper on each layer is specified by weight. On the PCB mentioned above all layers used 0.5oz copper. That indicates the weight of copper used to cover 1 square foot of PCB, in our case the copper is 17.5 microns thick.

Regards,

Andrew

PS: Sorry about the off topic 'lecture'; it's been a hard few weeks!

Nice theory, but sadly disproved by the facts. I'd class myself as a computer kid with DROs, but nevertheless it's "bleedin' obvious" to me.

Andrew

Jon,

No problem, it was a tongue in cheek comment.

As it happens the stock I turned started off just over an inch in diameter, so after a depth of cut of 0.1" followed by 0.2" it ended up about 0.5" diameter. Total length of cut was about 1", again by chance. Seemed to work fine, although if I was turning to a specific diameter I'd have taken a third lighter cut before measuring for the final 'to size' cut.

Regards,

Andrew

Jon,

No need to be insulting; menial indeed!

I assume you actually meant trivial, as in small? Depends how you define trivial. To ensure I'm not talking rubbish I've just tried a couple of quick tests on 1" diameter 6082. Running at 1200rpm (bit slow, but it is a very quiet Sunday morning here in rural Cambridgeshire) and a feed of 5 thou/rev I used a depth of cut of 0.1" and 0.2". No problems; the swarf came off in beautiful ribbons with no sign of build up on the insert. The insert wasn't new, it just happened to be in the lathe already as the last turning I did was on 6063 and delrin.

Regards,

Andrew

Here's my take on machining aluminium. It can be summed up like this:

1050 - pure aluminium, absolutely horrible, worse than copper, might as well try chewing gum

All the aluminium alloys I've machined (2014, 5083, 6063, 6082, 7075) were straightforward

For specific operations this is what I do:

Turning: I turn dry, with a special light alloy insert from Greenwood Tools. I run the lathe as fast as it, or the workpiece will allow, with a least a 4 thou/rev feedrate. I rarely get problems with built-up-edges on the insert. HSS also works well, provided the tool has a decent rake angle, is sharp and is smooth (I use diamond hones after grinding). The main issue is making sure the swarf doesn't end up as a birds nest.

Manual Drilling: No problem with HSS drills, I use WD40 to stop clogging of the drill flutes

Manual Milling: I don't do much of this, but when needed I do it dry. I have had a few issues with HSS cutters getting swarf build up in the flutes, so for preference I use carbide cutters

CNC Milling: Flood coolant (Biokool14 from Hallet Oils), mainly to get rid of chips with both HSS rippa and uncoated carbide mills. I run cutters fast, for example a 6mm carbide slot drill will run at 4000rpm

Best Regards,

Andrew

PS: Bazyle - I know I've quoted various specification numbers, but then again I very rarely buy metal from ME suppliers, because you haven't got a clue what you're getting. I prefer to use commercial metal stockists.

I bored the crankwebs for my hit 'n' miss engine on a lathe in a similar way to that described by JasonB; except for the following variations. I used the 4 jaw chuck rather than the faceplate, as it's difficult to clamp small parts to my faceplate. The crucial difference is that I made the two crankwebs overlength and bolted them together before boring. That way you only have two holes to line up, and even if you're a bit out, the holes are guaranteed to be in line.

My crankwebs were a similar size; I loctited the crankshaft together on a simple jig and then drilled, reamed and loctited 3/16" silver steel pins in each joint. I didn't bother making the crankshaft in one piece and cutting away the centre afterwards.

Best Regards,

Andrew

But surely the things we make are inherently 3D, so why would we want to design them in 2D?

Over the years I've used pencil and paper, proper drawing boards, several flavours of 2D, and of 3D, CAD. Personally I wouldn't use anything other than 3D CAD now. I find it more intuitive and quicker than 2D; it also allows me to assemble components to check for fits and alignments before cutting metal. And for multi-dimensional curves and free form surfaces it's not easy to accurately represent them in 2D form.

Regards,

Andrew

Quack, quack: that's me ducking!

Only reference I can find is for petcocks on old British motorcycles; like this:

http://www.steadfastcycles.com/cart/index.php?main_page=product_info&cPath=171_176&products_id=3917

Be about right; use an obscure thread nobody else uses, so the punter has to come to you for spares.

Regards,

Andrew

They look like half round drills, used for drilling holes in brass. See for instance:

http://www.drill-service.co.uk/Product.asp?Parent=020320040000&Tool=388

Regards,

Andrew

Sid: Thanks for explaining that my comment was slightly tongue-in-cheek.

Harold: No need to apologise, no harm done.

While I do have a CNC mill, and I find it very useful, it is just another tool in the box. After all a previous post in this thread shows some parts that were made entirely on manual machines. I certainly do not expect everybody to have CNC equipment; people may not want one, need one, have the money, be able to use the computer, or for a myriad of other reasons. Each to his own.

As for the shape of the hobby in the future who knows, but I suspect that it's changing. I know quite a few people in my area who have workshops in some form. Some are building model engines, many are not. Only one person belongs to the local model club and none of them post on this forum.

Regards,

Andrew

PS: I've blown a raspberry at the Raspberry; every time I looked at the Farnell website today I got a message saying did I want to register an interest in the Raspberry Pi before I could look up what I needed.

However it's cheap enough, so I might buy one later just to see how it's made. Looks like they're using package-on-package for the processor and SDRAM, pretty nifty. It would be interesting to know what design rules they had to use for the PCB.

I don't think so! If the job involved repetition I wouldn't be p1ssing about on a manual mill, stops or no stops. I'd fire up the CNC mill.

Regards,

Andrew

I don't have stops on any of my mills; and I can't say I've ever missed them. Both manual mills do have adjustable limit trips for the power feeds, but that's another story. The cavities in this picture were machined manually without stops:

I do have a DRO though, which makes it a bit easier.

Regards,

Andrew

Not so, gears of different pressure angles will not mesh together. It's all down to the geometry of the tooth form, it has nothing to do with load or speed.

Regards,

Andrew