Tortoise motor

Peter Prewett said the following on 29/03/2007 07:35:

Well, how often has this failure been reported by users? Whilst it doesn't help you, no company can ever guarantee 100% perfection - I don't imagine Tortoise motors are inspected by AOI, but by humans. Where they can score though is how they deal with a problem, but you haven't given them that opportunity.

I've certainly seen it a few times. Any kind of "sliding contact on PCB" switch should be avoided like the plague for this kind of application.

MBQ

snipped-for-privacy@hotmail.com said the following on 29/03/2007 09:23:

Ah-ha! So maybe there is an issue. Did you make Circuitron aware of the problem? With my Quality Manager's hat on, one of the things that bugs is when customers don't tell you if there's a problem. If the manufacturer knows about it, they should be able to act upon it. If they don't know about it, the chances of it being picked up are a lot less.

Anyone would think I've just had an ISO9000 internal audit session, covering the "customer complaints" part of the procedure :-)

I should have been clearer. I meant to say I've seen reports of it, not the actual problem. I'm not a customer myself. Circuitron should know about the problem. If they don't then that's another reason not to buy from them.

MBQ

snipped-for-privacy@hotmail.com said the following on 29/03/2007 11:29:

OK.

In a sense you're right, but you could also ask how they get to know about the problem if no-one tells them? (OK, they could have a second, higher level of sample inspection, but what if every sample is OK?) When you're doing inspection (as I should be right now!!), you are looking at the job to make sure it's right. After the first 185000 or so, you might not be looking quite so closely at the whole job as you should. If you are aware that a particular bit is a potential problem, then an inspector will still be able to focus on that bit just a bit closer even after 185000 units.

A real-life example - we had a problem with 15way high-density D-type plugs where the insert was back to front. At a "normal" inspection, it's "Yes, it's a plug not a socket, and it must be 15way because it's in the board". The incorrectly oriented insert isn't actually blindingly obvious, until it's pointed out, of course. Now it's "ah - a

15W HD plug - check the insert". To relate that to Circuitron, that could be "yes, there's a copper strip rubbing the board and they are OK" translating to "ah - we've had some wrong - make absolutely sure it's the right way".

It's just a fact of life that humans can miss things.

I'm talking about the design choice of having a copper strip rubbing on a board when I say Circuitron should know about the problem. IMHO it's a bad design choice for something that will be operated repeatedly over the lifetime of a layout. Taking your example of D connectors, I'm sure a lot of people would be surprised at the low number of operations these are often specified for (100 mating/ unmating cycles for the first example I found at Farnell). Unless it's very well designed (and probably expensive!) a sliding contact on PCB will not be much better.

Inspection and reporting for manufacturing faults is a different issue and, yes, I would hope QA is on the ball and that customers point out such issues.

MBQ

On 29/03/2007 15:10, snipped-for-privacy@hotmail.com said,

Well, yes, they ought to know about the bad design without being told - I take your point :-)

Surprising, isn't it, the number of mating cycles on some. People spend hundreds or thousands of pounds on a layout, then buy cheap D connectors to connect it all together. A bit of a weak link, methinks!

You don't have to use these. If they replace solenoid motors there are already three wires from the panel. If you use a 3-way double pole switch, you can use the third wire and the 3rd pole to route power at the frog, using the + and - track feeds.

It's got the advantage that you are using a regular, insustrial strength switch instead of the rubbing contacts and should bnever fail even after a long time.

Which raises another question. What connectors do people use?

On 29/03/2007 17:43, Christopher A.Lee said,

Er, D-types ;-)

I do use BICC-Burndy (aka QM Multipole in RS) connectors for power though, such as from the floor-mounted mains box to the panel.

Reason I asked, was that I haven't used D-plugs because they're not rated for the kind of current model locomotives pull.

I've used Jones plugs in the past but I'm not sure about their long-term reliability.

I want something that will work almost for ever and can be forgotten.

Christopher,

I use EDAC multi-pins.

They are the multi-pin of choice in the professional audio business and are ultra reliable. The contacts are rated to 8.5A and you can go from 20 pins up to 120 pins per connection. All the plugs have a central screw which assists in plugging and unplugging and there are two locating pins on the ends of each plug which can be orientated to give 36 permutations of mating. The pins can be inserted and removed and it is normal practice to join cable to pin then insert the pin in the holder - a lot easier than trying to wire up something like a 120 plug with all the pins fixed. They also come in panel mount and cable mount varieties.

One drawback is that they make a special tool for ejecting the pins from the receptacles and I think that costs more than a set of receptacles and pins :-) You can eject the pins using a small jeweller's screwdriver and a bit of cursing, but the tool does make the job very easy :-)

Jim.

Thanks. These look ideal.

I see these are in the Digi-Key mail order catalog - the closest dealer to me is in Long Island. Which I didn't know counted as up-state New York :-)

Cheers....Chris

Christopher,

If you are going to try them one or two things to watch out for.

You don't always need to fill up all the holes in a receptacle and you might find occasions where you might only put, say, 20 pins in a 38 pin setup. If this happens, balance the positioning of the pins on both sides of centre so that the forces of mating and parting are balanced on each side, or the receptacle starts to twist and lock up.

I use a wooden spring clothes peg to hold the pins when soldering to the cable ends - quickest and best clamp for the job. they also make pins for crimp, wire wrap, etc.,

The most common fault is not pressing a pin home fully so that its locking spring clicks into place. Then, when you bring the plugs together, the pin gets pushed back out and makes a dodgy connection. It's easy to see this fault if you part the plugs and look at the pins on the mating ends and you can see any pin that has been pushed back in its hole. Just push any misplaced pin fully home until it clicks in place.

It's worth getting the pin ejection tool if you think you will be needing to get pins out of the holders. You might want to do this if you want to re-wire the pins, or alter a wiring setup in some way. The tool fits neatly over the pin and has two arms which depress the locking spring and a central pusher rod which ejects the pin. You can get round not using this tool if you insert a small jeweller's screwdriver to depress the spring, then grab the other end of the pin with a pair of pointed nose pliers and pull it out. Sometimes this goes quite well and other times it doesn't :-)

Jim.

Thanks.

I've saved this and your other message, together with the URL and the PDF I linked to.

I haven't worked out all the connection details yet. It's under construction, just a simple oval with a passing station on one side and a fiddle yard on the other.

Paul Boyd wrote: [snip]

I would but they do not have a web site or email address that I could find having searched many times. Unless you know better:-)) Peter

D25's for track power, although having only six blocks means I've been able to double up on the contacts on each joint.

Accessory power is carried using RCA connectors. In one spot where I have to drive a point motor that's just on the wrong side of a baseboard gap, I use a 6.5mm stereo headphone plug/socket.

Finally, a remote control panel that was a recent addition uses HD-D15s, but that was partly so I could run the 5m cord using ribbon cable and an IDC plug to save time/space and make it neater (this only carries LED currents on half the wires, and a logic level signal on the other half)

Regards,

Stuart

Peter Prewett said the following on 29/03/2007 23:33:

Circuitron, Inc.

211 RocBaar Drive Romeoville IL 60446 USA

Phone: (815) 886-9010 Fax: (815) 886-9076

Email: snipped-for-privacy@COMPUSERVE.COM

Christopher A.Lee wrote in news: snipped-for-privacy@4ax.com:

I don't everything is hard wired, where I need to make a connection I put a screw in at the apointed place wrap the wires that need to be joined tightly around it then soak it in solder. Bloodly bombproof!!!

But then my layout takes up the entire inside of a garage and never leaves home.

Back to the original subject!!

I emailed Circuitron and they replied and it may be worth passing on.

Peter

Hi Peter,

I can understand your concern, but there should be no problem. The plating on the circuit board is VERY thin and it takes little motion to wear through it. And it is an unfortunate fact that as the punching die wears between sharpenings, a slight burr can be produced on the contacts. The die is designed to produce "spoon-shaped" fingers, but there is often some deviation from this. And, as you have noticed, the location of this burr will be the first place to show wear on the copper. However, this is a self-limiting process. IF the burr wears through the copper first, it will encounter the fiberglass material of the circuit board substrate. This material is highly abrasive and will grind down the burr. But what will happen first (since the burr is so small) is that a slight groove will be worn into the copper at which time more surface of the contact will start to ride on the surface, thus preventing much deeper wear. I don't think this will be an issue for you. In 19 years of selling TORTOISEs, I have never gotten one back with worn out contacts or circuit boards.

Steve Worack