By Steve Bush 29th September 2021
I bought a small old and rather battered milling machine.
It is a Sieg X1, painted yellow and sold as a Clarke CMD10 in this case.
These are basic machines built to a price and tend to be rather roughly finished in the places that don’t show (and some that do).
Much as it was apparently loved by its previous owner, Day 1 was spent un-seizing and lubricating parts – particularly the bed, and getting the leadscrews to move reasonably (it would take a lot of careful fettling to get it to move really nicely – some folk have the knowledge, skill and time, and have done it).
Drawbar cover missing – hence swarf inside the gear casing
Day2 was the big switch-on, and it ran immediately, as soon as I plugged it in – Problem 1: no proper ‘no-volt release’ – dangerous.
Problem 2 was that the speed controller did not control the speed – it just ran at one constant speed.
Deep breath – the controllers in Sieg machines are notoriously fragile and expensive (£185…).
Replacement drawbar cover designed and 3d printed, swarf vacuumed out
Web searching revealed some circuit diagrams that kind folk had reverse-engineered and posted – not of my board, but related boards in other Sieg machines.
Apart from early thyristor versions, they are all mosfet choppers – the dc motor being supplied with rectified mains (no link capacitor) chopped by a mosfet.
When they break, it is almost always the mosfet – most people blame too much current through the mosfet, but it is an IRFP450 in my 150W machine which can handle 14A cold and 8A hot, plus pulses up to 56A – and the circuit has a current over-load trip built in.
Far more likely, if feel, that its 500V rating is not enough to handle transients on top of the 380V mains peak – at least one other person the the web thinks the same (as it happens, the mosfet in this mill was working).
Update: Read Luke Hear’s detailed comment below – it identifies poor current trip design and over-heating as the likely culprit
Taking a deep breath, I started (what I thought was) fault finding – made more tricky because the whole control circuit is at mains live potential – so: battery-powered scope, set-up, stand back, turn on and view – then turn off and repeat.
So that I did not have to probe the live board, I soldered insulated wires to test points. And that was my undoing, as, in swapping the scope probe from one test point to another, a wire pinged back and touched the metal chassis – ‘pop’ went one of the opto-couplers as its package disintegrated (and out when the lights as the RCD tripped).
Fault-finding had turned into fault-adding… Heaven knows what else was ruined as the 240V-driven surge went through all those LM324s on the controller board.
I dejectedly disconnected everything and left the workshop for a lie down.
The latter uses a pair of opto-isolators to bring the PWM waveform to the mosfet gate (right).
Much simplified circuit to show the gate driver. For full circuit (thanks to Steve Kurt comment below for pointing out this omission) two people have reverse-engineered versions: John Swift (and here) and John Gerling (sorry can’t find original source).
The leds are wired back-to-back, so that one or the other is on when they are driven with a bi-polar waveform.
The output transistors are wired as a totem pole drive for the gate (from a local 18V supply), giving the drive reasonably fast turn-on and turn-off – none of the long turn-off tail that a single opto isolator would give.
This is the version in the FC150BJ driver board, intended for 150W 220V dc motors. 250W and 350W versions duplicate both the mosfet and the two gate resistors.
The control circuit is tied to the un-filtered rectified 230V rail so that motor current and voltage can be sensed using resistors – Heaven knows why it is tied via +12V and not 0V, nor why a bi-polar supply is required.
It was one of the opto-coupler input ends that I earthed with such disaster.
Hi everybody, what an interesting article! Here’s my two penn’orth… I had what seems to be a very similar electronic speed controller in for repair a couple of years ago. The output was two MOSFETs in parallel, and tests showed that one was broken. The controller was from a model-makers’ lathe, it had a speed control and a selection of lights. It was quite old so I put the failure down to old age. I replaced the offending MOSFET and returned it, having tested it briefly as a dimmer for a mains filament bulb. A few weeks later, back it came with two duff MOSFETs this time. It turned out that the owner had been using it to drill holes of a size and in a material that were not recommended by the manufacturer. There was an overcurrent protection circuit on the board which should have stopped it prior to the destruction of the MOSFET but obviously it hadn’t. So I replaced both MOSFETs, connected the ‘scope to it and switched on… On the screen was clearly displayed a full-wave rectified and chopped output waveform. The speed control seemed to vary the mark:space ratio of the chopper, which was not synchronised to the mains frequency or indeed anything at all. Then I had an inkling, especially when a faint buzzing noise started up from the bowels of the circuitry. The overcurrent system was doomed to failure – it looked as if this part of the circuit had remained unaltered when the original speed control design went from straightforward voltage control to the chopper system. The current pulses were too short for the dropout relay to, er, drop out, it was the relay chattering that caused the buzzing noises. The overheating MOSFETs were not helped by having to switch on and off lots of times and by being mounted on a much-too-small heatsink. A look at the spec and further tests with a thermometer showed they were indeed getting too hot. The ventilation system was a few holes in the box, which was so tightly packed that hardly any air could travel through it. Running it with the lid off made quite a difference, and when I temporarily fitted a small fan loosely held by gravity the problem went away. There was no room for a fan to be permanently fitted inside the box, so I drilled a fan-shaped array of holes ready to take an external fan (with fingerguard of course!) that would run from the 12 volt supply inside the box. An alternative was to fit a bi-metallic strip temperature-sensing switch, a solution that the owner preferred despite my remarks about the motor stopping without warning in mid-drill. I screwed the switch to the heatsink with a bit of heatsink compound between the two, the switch is about the size and shape of a Viagra tablet (so I’m told). Everything worked a treat, and the switch only took about a minute to reset. I keep the filament bulb under lock and key, mainly because LED bulbs operate at all sorts of odd frequencies. As such, they are no good for checking the speed of (gramophone) record decks.
Morning Luke Hear Nice of you to take the time to write that – I enjoyed reading it, and have added a note to the article just in case anyone has a similar problem. Your circuit sounds much like the one I blew up through clumsiness – it is odd that they have these flaws – almost as though it was a design modified by someone who didn’t quite have enough knowledge, or was too rushed. The cheap-ish controller board I replaced mine with was better designed – the circuitry seemed to have everything it needed, including speed regulation, but only using four op-amps. The only draw-back was the hard-to-filter thyristor switch (and the stop switch wiring I removed because it led live mains out of the enclosure….) The (laboriously!! reverse-engineered replacement is here: https://www.electronicsweekly.com/blogs/engineer-in-wonderland/reverse-engineering-dc-51-motor-speed-controller-2021-10/
a follow-up on the subject of high voltage differential probes… I found the Dave Jones video on reverse engineering the Micsig DP10007 HV Diff Probe, which pointed to a EEVBlog forum, which mentioned a Circuit Cellar article about designing a high voltage differential probe: https://circuitcellar.com/research-design-hub/high-voltage-differential-probe/ It has a schematic suitable for fab’ing yourself. Seems like it might be an interesting project, especially if you need one. A search for the Micsig probe shows that Amazon has them for around $300 (at least right now and here in the middle of the USA).
https://www.screwfix.com/p/carroll-meynell-1500va-intermittent-isolation-transformer-230v-230v/752HV?tc=MB2&ds_kid=92700055281954493&ds_rl=1249401&gclid=EAIaIQobChMI66Tbreit8wIVBbDtCh0HgADhEAQYAyABEgL8GvD_BwE&gclsrc=aw.ds
They save your life & equipment.
When I was (very briefly) fixing SMPSUs for slot machines, I used a 300W security light as a ballast in the live line since I singularly hate unexpected bangs & flashes.
And much cheaper than changing endless fuses.
If only you had mentioned isolation transformers to me last week ? I shall put one on my christmas list. BTW, I needed a 100W incandescent bulb as a dummy load – and had to ask around, eventually I found a neighbour with one in a dusty cupboard. It never occurred to me that I would ever need such things again when I binned all of mine in favour of led lighting. ps – never though of using a light bulb in series as test protection in a mains circuit, even though I have used them during car wiring adventures – thanks for the idea (now I need to find three more 100W incandescent bulbs…..)
I noticed that Wilko had a supply of 118mm halogen lamps the other day, though where you’d get a suitable fitting from is anyone’s guess.
Yesterday I pressed the 500W halogen lamp into use once more when testing some wiring for interesting faults.
And nothing tripped, so even better.
The other problem is, of course, that a 500W isolation tranformer is quite large, heavy & expensive & a 1kW one is even more so.
I’ve found a few strange faults by connecting a bulb to where a fuse ought to be too. Its a technique that is very useful at times, especially on cars. Those little blade-type fuses only need a bit of plastic to be removed, a length of figure-8 soldered to the exposed metal, and a bulb at the other end placed where it can be easily seen.
Morning Luke Hear That is a very nice idea. I tend to put a headlamp bulb in series with new vehicle circuits before powering them up for the first time, but it never occurred to me to make up a dummy fuse.
I’m sure I’m missing some detail, but my first thought was to add some big transorbs at the rectified mains power. That should get rid of any brief, high voltage spikes. The next thought was “why isn’t there any snubber across the mosfet?”. Maybe the turn-on and turn-off are slow enough that it’s not an issue, but when switching something as inductive as a motor, shouldn’t there be something to control the energy stored in the inductance? Maybe just a flyback diode around the motor? This seems like something that has been solved a long time ago in other designs.
Good to hear from you Mr Kurt Indeed, I missed off a lot of detail to show the neat isolated gate drive. There is indeed a snubber across the mosfet (RCD type) and a fast diode across the motor. Two people have reverse-engineered versions of the circuit very well – one is John Gerling (http://timsmachines.com/wp-content/uploads/2018/05/044991-ServiceManual.pdf – sorry can’t find original source) and the other is John Swift (https://www.model-engineer.co.uk/forums/postings.asp?th=38809 and https://www.madmodder.net/index.php?topic=3634.0). Tranzorbs sound a good idea, but for the 230V version, a 650V mosfet also sounds like a good idea too. Thanks for pointing out potential confusion in the article – I will add these links to it to clarify things.
I took a look at the first link.. model engineer… and the schematics are more detailed. The schematics are also drawn in a more confusing manner… so I thank you for drawing your schematic with the input on the left and the output on the right (as schematics should be drawn)! ? Probing high voltages is a risky affair. I recall working on a TV projector with high voltages and needing an extender for the ground lead on the scope. It happened to touch a high voltage trace, with the usual result. I think this was the first (and only?) time I’ve seen a trace actually peel off a board and curl up! It does make me think that a high voltage isolation probe gadget might be a good investment.
Good Morning Mr Kurt Sorry to hear you had a similar experience – and you are the most careful of people. I did indeed look up high-voltage differential probes – but v expensive, even pre-owned on ebay. I have temporarily bought an external motor controller (A ‘DC-51’ for £27 – disappointed to see it is SCR-based, not mosfet) to at least allow me to try the mechanics. Not sure if I should scrap the original controller now (which has the benefit of voltage feedback and over-current protection) of spend the time fixing it.
I appreciate your confidence in my exercise of caution and care.. but there is a reason that I’ve reached this state! I think there is a saying “wisdom comes from experience, and experience comes from doing stupid things”.. or something like that. ? The differential voltage probes aren’t cheap.. probably a few hundred dollars, at least. Dave Jones mentioned an off-brand model that is very similar to a LeCroy (IIRC) and cheaper, but still not “cheap”. He did reverse engineer it, for the most part, and it is really not that exotic. I think you could throw together an equivalent for not a lot of money. It might make a good blog subject too. ?
Ps, Thanks for the schematic direction complement (btw, as a result, I am incapable or representing bi-directional power converters ? I did like the way the Model Engineer guy helpfully split out the mains input side (think this separate circuit might be in a different post in the same list).
Your email address will not be published. Required fields are marked *
Download the Elektra Awards 2022 media pack and book your sponsorship package to be part of the most prestigious awards event celebrating its 20th anniversary this year!
Have your say in choosing the most promising UK university research project reported in the last year, from Bristol or ICL to Surrey or Southampton, and robotic fingers to flexible and stretchy supercapacitors...
Get our news, blogs and comments straight to your inbox! Sign up for the Electronics Weekly newsletters: Mannerisms, Gadget Master and the Daily and Weekly roundups.
Read our special supplement celebrating 60 years of Electronics Weekly and looking ahead to the future of the industry.
Read the Electronics Weekly @ 60 supplement »
Read the first ever Electronics Weekly online: 7th September 1960. We've scanned the very first edition so you can enjoy it.
Read the very first edition »
Electronics Weekly teams up with RS Grass Roots to highlight the brightest young electronic engineers in the UK today.
Read our special supplement celebrating 60 years of Electronics Weekly and looking ahead to the future of the industry.
Read the Electronics Weekly @ 60 supplement »
Read the first ever Electronics Weekly online: 7th September 1960. We've scanned the very first edition so you can enjoy it.
Read the very first edition »
Tune into this Xilinx interview: Responding to platform-based embedded design
Tune into this podcast to hear from Chetan Khona (Director Industrial, Vision, Healthcare & Sciences at Xilinx) about how Xilinx and the semiconductor industry is responding to customer demands.
By using this website you are consenting to the use of cookies. Electronics Weekly is owned by Metropolis International Group Limited, a member of the Metropolis Group; you can view our privacy and cookies policy here.