Real Bull Controller box question

[sacentre]

There are dozens of off-the-shelf panel mounted RPM displays available on line. I suppose anyone of them would do as long as there were circuit details available - like what encoder to use etc. Many of them come with magnetic (hall effect?) sensors all for around 5 or 6 dollars.

The suggestions in your second and third paras are a bit over my head, I'm afraid. I've never owned a signal generator or a scope for that matter so it will just have to be a cardboard disk on the end of a drill - more my level. I have assembled kits using 555 timers in the past. As you say, one of the cheapest ICs around.

Although I've enjoyed mucking about with practical electronics since I was a kid (I'm 74 now), I never got very far with theory. As for the maths involved. forget it! I can follow a simple circuit diagram to build a PCB but if it doesn't work afterwards, I'm lost - unless it's something obvious like a badly soldered joint or a diode the wrong way round.

Many years ago, I used to get a great kick out of editing single-side PCB artwork on a PC (from an existing circuit) using Protel Autotrax and then plotting the pattern out on a Roland XY pIotter directly onto the copper clad board (I never mastered the photo-mask method). The .3mm fibre pens were etch resistant and produced very clean results. All just .1inch stuff, of course. It was hugely satisfying watching the PCB tracks appear on the board before dropping it straight to the etching bath. After etching, I'd drill the holes by hand with a .7mm drill. I made lots of small gadgets that way. Sadly, I had to give that method up because I couldn't get the plotter pens. If I were doing this stuff now, I'd most likely be looking for a CNC router to produce PCBs.

These days, it's just assembly and soldering I really enjoy. Recently, I bought some kits to assemble that used SMD exclusively. I'd only ever assembled through-hole PCBs before this. I had to buy one of these el-cheapo digital microscopes to cope with the .5mm pitch ICs. Great fun learning to solder with a soldering tip like a needle while looking at an LCD screen!

Please excuse the ramble.

 

[sacentre]

Well, Mark, Homebrewed, I'm more confused than ever now.

I ran the display and encoder off my bench power supply (Clip 1) for more than an hour and the fault never showed up once. "Aha!" said I. "Now we're getting somewhere!"

Next I re-connected it to the controller box (Clip 2) and .... nothing. No fault! I switched off and on again, left it running for many minutes. Nothing.

So, I'm back to square one having learned nothing from all this and see no evidence that anything needs replacing. There has to be some other variable I'm just not seeing. As you can make out from the clips, I used a small hand held-drill gripped in my bench vice with the encoder fastened to a strip of steel with doube-sided tape. The disk is just a bit of thin black card with a single slot. I had the controller box connected to a length of mains wire plugged into a socket near my desk. What difference any of that makes to anything I just don't know. I'm pretty sure that once I re-install the encoder on the lathe and reconnect it back to the box, having re-connected the lathe mains cable, I'll see the mysterious flashing 8888 again. I might just have to live with it.

When things get like this , I'm reduced to suspecting that it could be the mains socket it's plugged into so I'll be running an extension to a different room to test that half-baked theory.

 

[homebrewed]

Well, Mark, Homebrewed, I'm more confused than ever now.

I ran the display and encoder off my bench power supply (Clip 1) for more than an hour and the fault never showed up once. "Aha!" said I. "Now we're getting somewhere!"

Next I re-connected it to the controller box (Clip 2) and .... nothing. No fault! I switched off and on again, left it running for many minutes. Nothing.

So, I'm back to square one having learned nothing from all this and see no evidence that anything needs replacing. There has to be some other variable I'm just not seeing. As you can make out from the clips, I used a small hand held-drill gripped in my bench vice with the encoder fastened to a strip of steel with doube-sided tape. The disk is just a bit of thin black card with a single slot. I had the controller box connected to a length of mains wire plugged into a socket near my desk. What difference any of that makes to anything I just don't know. I'm pretty sure that once I re-install the encoder on the lathe and reconnect it back to the box, having re-connected the lathe mains cable, I'll see the mysterious flashing 8888 again. I might just have to live with it.

When things get like this , I'm reduced to suspecting that it could be the mains socket it's plugged into so I'll be running an extension to a different room to test that half-baked theory.

Ugh. The symptoms look like some kind of intermittent, since messing around with the connectors/cables/boards temporarily fixed the problem. Looking at it as a which-is-most-likely scenario, I'd look at the cable going between the display and controller board. Problem is, it could be a broken wire inside the jacket. So nothing visible externally.

Replacing that cable with a new one, or one you make yourself, might fix the problem -- or might not. You won't know until you use your machine awhile. If the problem comes back, at least you've eliminated the cable, then its time to try something else. That's the nature of intermittent failures.

 

[sacentre]

Your advice about possible cable trouble is of course, perfectly sound and I'll keep that in mind as I run the box on the lathe. There's still a question mark over the controller board itself as I'm still seeing the "hunting" problem I reported earlier. Replacing motor brushes made no difference. That might turn out to be the problem all along. As you said, it will really suck if I've got another faulty one.

On the subject of weird intermittents, you might find this interesting. I hope I'm not upsetting a Moderator with this OT stuff but I'll try to keep it brief.

When I worked as IT manager for a company in Singapore, we ran into a problem with our Dell Exchange server that suddenly refused to boot. It was in a 19" rack along with several others all of which were working normally. Me and a technician went through all the normal substitution routines checking connections, mains cable, re-seating and then substituting RAM modules, CPU, HD - everything. In the end we called our local Dell support tech who brought a replacement motherboard (what else was left?) which we installed. This was after he'd tried all the things we'd already done. Nothing worked. He even suspected the non-Dell monitor and keyboard that were plugged into it (I have come across a faulty PS2 keyboard that stopped a PC from booting) and then the UPS so we swapped these out. Nothing we did made any difference. It simply wouldn't boot, no POST nothing.

Even the Dell guy said he'd never seen a problem like it before. Everyone I've told this story to says we must have missed something. It had to be something we overlooked. I would never have believed it myself had there not been three of us there at the time. There cannot but be a logical cause and effect, right? It's empirical science, not magic, elves, gremlins, Feng Shui or anything else? So we all believed and I still do. Yet, there was nothing anyone could think of that we didn't try and still the damn thing wouldn't boot. Until ....

We decided to take the server to another room down the corridor along with monitor and KB, where we had a big table and more space to work so we could strip it down and progressively re-establish it. Naturally, the first thing we did was plug it into the wall and .... You guessed it! It booted normally! Back again to the server rack. Nothing. Remember that the rack housed 3 other servers all working OK from the same mains socket strip and the same UPS. Back once more to the other room and it booted fine.

We never got to the bottom of it having worked the problem to exhaustion well into the small hours. What was it about that one server that prevented it booting in that particular room? In the end, we took it to my office on another floor, plugged it into the wall and connected the LAN where it worked without fail until it was replaced under our Dell replacement programme. There are more things in heaven and earth etc, etc..

Thanks for reading and apologies again for a not so brief ramble.
Trevor

 

[markba633csi]

Weird story! Thankfully those don't happen that often.
The 4-pin IC that Homebrewed mentioned on the power board is, I think, the bridge rectifier for the readout power supply
so it seems likely the opto interrupter feeds directly to the readout chip.

You could use an LED and approx. 2K resistor to make a signal probe, see if the opto is going dead when fault occurs

Japan has a section of town called "Akihabara" a market place for all things electronic- maybe there is a place in your city like that?

Remember that these products spend a long time on ships in salty air before delivery- your problem may simply be some slight corrosion on a connector or IC socket- by moving things around you "fixed" it. A poor solder joint could also be the culprit

 

[markba633csi]

Simple logic probe:
By the way, you never told us what happens if the unit is working properly and you disconnect the opto cable. Do you get 8888?

 

[sacentre]

Weird story! Thankfully those don't happen that often.
The 4-pin IC that Homebrewed mentioned on the power board is, I think, the bridge rectifier for the readout power supply
so it seems likely the opto interrupter feeds directly to the readout chip.

You could use an LED and approx. 2K resistor to make a signal probe, see if the opto is going dead when fault occurs

Japan has a section of town called "Akihabara" a market place for all things electronic- maybe there is a place in your city like that?

Remember that these products spend a long time on ships in salty air before delivery- your problem may simply be some slight corrosion on a connector or IC socket- by moving things around you "fixed" it. A poor solder joint could also be the culprit

 

[sacentre]

And not the first time I've see totally inexplicable problems with PCs. I'm not thinking this controller display issue is anything like that though.

Thanks for the logic probe suggestion. We have a place here called Sim Lim Tower which has 3 floors of wall-to-wall electronics parts shops.

I've now run the lathe several times without the fault occurring and I'll continue to test it each day while the box is off the lathe. You mentioned corrosion and I've seen plenty of problems caused by the humidity here which is 70-90% all year round. It's just possible one of the headers might not have been making a good connection but there's certainly nothing obvious. Anyway, I'll continue to monitor.

 

[sacentre]

Mark, Homebrewed
I'm sure you must be pretty bored with this thread by now but I thought I'd let you know the state of play. As I expected, the problem reappeared after I'd screwed the box back on the lathe and started using it. Exactly the same behavior. This suggests now the trouble lies with the filter board and the 5v DC supply to the display/encoder combo. I'll take the box off again and have a look a that regulator IC or perhaps just replace it and see what happens. I'll wait till after Christmas and then make a trip down to the electronics shops to see if I can find one. Assuming the maker hasn't scratched the ID of it (I never bothered to look when I had it on the bench). Of course, this wouldn't explain why the problem never appeared when I had it running on the bench but then again, that was with no motor connected.

Thanks again for all your help and I wish you both an enjoyable and relaxing Christmas.
Trevor

 

[sacentre]

Mark, Homebrewed

Problem with RPM Display fixed.
I felt I owed it to you to let you know the outcome of this issue after you both contributed so much of your time helping me with it.

Following your suggestion, I replaced the BD107 bridge rectifier and the 78L05 voltage regulator and there was no further problem with the display during several hours of use today. After running the module on batteries without trouble prior to this, it seemed this was the only thing left to try. It seems likely that it was after all, a thermal intermittent fault with one or both of those components. I would have replaced one at a time and re-tested but I got lazy. Anyway, all's well now.

Thank you both once again for all your help.

Trevor