(Don't just Google Motorhead or you might get a completely different answer!) I would give them a ring - nothing to lose. There is a good Dyson repairer near me and I am sure there must be many more. Maybe they could cannibalise another machine?
I did go through the fault routine on the Dyson website and having established it wasn't a stuck brush did ring them as the web page suggested. The only response I got was Dyson no longer support the machine though I could be entitled to a discount on a new model.
My father was a fan of Dyson so he bought my mother the vacuum cleaner, he died over 10 years ago so it is old.
I inherited a Dyson washing machine from him and it was nothing but trouble, after the second control panel failed I binned it. I swore not to buy another Dyson product, though I have had some decent upright vacuum cleaners of theirs as gifts, which have been passed on to family members.
So rather than waste this one I aim to try and repair the motorhead function, though it still works as a normal vacuum cleaner.
I would still check with a local third party Dyson repairer (if you have one) to see if he/she could point you in the right direction. They might have a spare lying at the back for all you know.
He was in an interview about a year or so back where he admitted that the washing machine was a bit too complicated and there were too many bits to fail if the contra rotating drums got stuck or several other special bits went south. It was also not very useful as it never fitted in a standard sized hole in kitchen units.
On the brush issue, what exactly does the pcb do? If an led glows then I'd have supposed that the fault was an over current safety shut down mode. as a stuck brush would look like tat and the circuit would disconnect the motor and light the led. I've seen similar things on other devices like blenders etc. Brian
I don't know, it seems to have a 1 microFarad capacitor across the motor terminals , 2 inductances, 2 diodes and various resistors.
I can post a picture but what picture hosting sites are available now?
Yes but I think the fault is in the supply before it gets to the PCB.
I suspect there is a power supply in the handle near the switch but the trouble with Dysons is getting them apart without breaking anything. They are often assembled with plastic clip lugs in blind holes and if they won't pull apart something else gives or the lug breaks off.
I'll search the forum Mike suggested.
I've already stripped down to the main motor and no sign of any active electronics there.
Thus neatly proving that the motor is working ok. :-)
Actually, there's every chance you've hit the nail right on the head! Most likely (assuming there's no fancy speed controller included), the "Power Supply" is nothing more than a bridge rectifier (a module or 4 discreet 1N4007 or similar diodes).
From your description, my inference is that the PCB in question is in the vacuum cleaner itself rather than in the motor head. If you repeat your test with your moving coil multimeter using a suitable AC voltage range (say 5 to 6 hundred volt scale) and take readings with the test probes applied each way around, you should see only a very low reverse reading when applied the wrong way round and an ac voltage reading close to *double* the expected mains voltage when the rectifier is working as a fullwave rectifier[1]. If the rectifier has an open circuit diode as your voltage reading suggests, it will be acting as a halfwave rectifier and the meter will then read just slightly less than the AC voltage feeding the rectifier[2]. If this turns out to be the case, 1N4007 diodes or 1A mains bridge rectifier modules are readily available and cheap. All you then need is a soldering iron, some multicore solder, a cheap desoldering pump or small flat bladed screwdriver, long nose pliers and the skill to use them to remove the faulty component(s) and solder in the replacements. Alternatively if you lack such tools, find someone who does have such kit and the necessary skills. :-)
[1] This assumes that the rectified output voltage is not subjected to a smoothing circuit such as a 400v rated capacitor placed across its DC output terminals - a pretty safe bet in this case since the 230v DC motor will work perfectly fine with an unsmoothed rectified supply without incurring the needless expense of an extra failure prone component which itself places additional stress on the rectifier.
[2] If one of the rectifier diodes has failed open circuit, you'll only see the expected mains voltage reading. The reason for this behaviour with the classic moving coil multimeter is due to the AC voltage ranges relying on half wave rectification of the test voltage which halves the average current, necessitating an approximate halving of the multiplier resistances used for each ac voltage range (approximate because the meter reads the average voltage but is calibrated to show the RMS value which is about 10% higher, requiring a similar further 10% reduction in multiplier resistance values to compensate.
A typical 20Kohms per DC volt multimeter will have a 10K ohm per AC volt sensitivity (the moving coil meter circuit is either desensitised slightly on its DC volts and amps range to make the half ohms per volt sensitivity figure for the AC ranges true or else the given AC ohms per volt figure is simply a rounded up approximation.
When you connect an old fashioned analogue multimeter to a dc test voltage using one of its AC voltage ranges (say a 12v SLA measured on the
30 or 50 volt AC range) the reading will slightly double that which was read on the DC volts range. If you test a halfwave rectified DC supply, the AC volts reading will be only slightly less than the AC voltage feeding the rectifier diode. Testing with a fullwave rectified DC voltage will double this reading for pretty much the same reason that measuring
12vdc on a 30vAC range gives a reading of just over 25 volt.
No the PCB is actually in the head, it seems to be supplied with DC via a 4 position switch in the handle which I have not been able to open up to investigate. The first position is all off, the second is vacuum only , the third is vacuum and motor head brush which gives a red light on the motor head PCB and the fourth is supposed to just rotate the head slowly. I took one input terminal off the PCB (and broke the push on crimp in doing so) and tested the voltage in positions 3 and 4. Three powers up the main vacuum motor and supplies only a few tens of VDC, 4 gives about 90VDC, this suggests a problem in the switch.
As you say it reads over 200V on an AC scale with terminals one way and zero the other.
Looking at the forum Mike suggested the is a common problem with this switch but no power supply components are mentioned.
There's a good chance that the switch feeds the power to a bridge rectifier in positions 3 & 4. Using a 2 pole 4 way switch will allow one of the brush head motor wires to be transferred from a DC output terminal (+ve or -ve) on the rectifier to one of its ac input terminals (~), effectively turning the circuit into a half-wave rectifier cutting the unsmoothed averaged DC voltage in half[1]. It's a cheap and crude (but effective) form of DC motor speed control.
The problem could be a switch or rectifier fault or just simply a broken or disconnected wire between the switch and the brush head motor circuit. You'll have to gain access to the switch wiring if you want to pursue this fault finding exercise any further.
[1] This slow speed circuit might include a dropper resistor if half speed isn't quite slow enough on its own account. It's also worth bearing in mind that another possible way to feed that dc motor with half voltage is to switch the rectifier's ac input terminal from full mains voltage to half mains voltage available at the join between the two field windings on the main universal motor.
Indeed, if the desired motor speeds are full and quarter speed, the two methods can be combined on the fourth position if a three pole 4 way switch is used. I can't think of a way of doing this with just a 2 pole 4 way switch off the top of my head but if there's a way, you can bet your life that Dyson would have thought of it. :-)
Looking at the picture, I can't see any signs of a triac (nor any place on that PCB where one might lie hidden from view unless it's mounted on the reverse side).
The circled area marked "RV1" looks like this Variable Resistor component may have been replaced by a couple of resistors located at the switch (it's hard to tell, but it looks like the mounting holes are wired back to the switch).
The two electrolytic caps at the opposite end of the board look suspiciously like they may be a couple of 200vdc rated caps wired in series to provide a smoothed 350v HT supply for a switching converter (perhaps there *are* additional surface mount components on the underside after all but the PCB's appearance suggests otherwise).
Apart from the 6 rectifier diodes and a small glass diode (probably a zenner), I can see what looks like a 15 ohm (green bodied) resistor with what looks like a smaller 22K ohm resistor alongside. The 15 ohm resistor is probably a current inrush limiter for the 350v HT rectifier circuit and the 22K looks like it feeds current to that glass diode (zenner?).
The remaining components could both be small capacitors. The black one certainly is but the red "ceramic disk" might be a spike clipping varistor (although, ime, these are usually coloured blue).
Unfortunately, without a set of decent photos to show both sides of the PCB *clearly* (the wires should have been moved aside or a better angle of shot chosen to make it clear as to what and how many wires are connected to the board), it's impossible to work out exactly how this functions as a DC motor speed controller and what possible failure mode could have led to the symptoms and measurements you've taken with your moving coil multimeter. We can take educated guesses but unless someone familiar with the innards of a DC05 recognises that board, an educated guess is the best you're going to get.
If you can't track down a service manual with circuit diagrams for that model of vacuum cleaner, the wiring will have to traced (both the PCB circuit and the rest of the wiring including switches and motors etc) to recreate a circuit diagram in order to continue the quest to effect a repair.
Unless you can't actually see the switch connections, it might be worth posting a few more photos (uncluttered views both sides of the board and a photo or two of the switch wiring). I'm guessing you'll find a couple of resistors wired directly onto the switch contacts once you gain access (if you have access to one of those cheap snake (bore-scope) cameras, you might be able get enough of a view to determine whether or not this is the case).
is a good starting point of reference which, at a glance, looks quite useful. :-)
However, further googling got me no nearer, as promised in the link above, to tracking down a service manual or circuit diagram so it looks like you'll have to trace the circuitry yourself to create a full circuit diagram along with an accurate parts list. Once you have this, the only likely proprietary bits will be the switch and the motor. The motor is apparently readily available (take heed of the advice).
I'm not so sure about the switch - you can google the various spare parts traders' sites yourself if the switch proves to be the fault. The fact that the 230vdc motor still runs slowly on a few dozen volts suggests the problem isn't there (BICBW).
The black rectangle with a heat sink in the side above C2, what is it?
RV1 is the red blob
Yes I thought it was a zener too.
I will get some better photos up and see if I can trace the circuit and also the other PCB in the motorhead but as I only want the brush to work would there be any harm in abandoning the Dyson PCB and feeding it with a simple 400V rated bridge rectifier off a separate switch?
I can't see any evidence of a heatsink from that angle. However, it's possible that what I thought was a small 100nF capacitor might well be a plastic power transistor or triac with the heatsink tab chopped off flush to the encapsulation.
Ok, now you've given me another clue, I can see how that would fit. The red blob, which I have to assume is a varistor in spite of it not being blue, looks like it has been flipped up from its intended laid down flat mounting position as implied by the silk screen print.
Assuming that the "100nF capacitor" is actually a triac, I doubt you'll get away with just a simple bridge rectifier. However, if you're going to experiment, that 400v rating should be for the ac input voltage based on the assumption that the output will be connected to a simple reservoir smoothing capacitor ripple filter.
The rule of thumb in this case is to rate the diode PIV at 3 times the maximum rms input voltage since each diode element in the rectifier bridge will be subjected to twice the peak inverse voltage (1.414 times the rms ac voltage) due to the peak voltage stored by the smoothing capacitor. Assuming a perfect sine wave form on a 240v ac supply, each diode will be subjected to a PIV of 2.828 * 240 volts, just under 680 volts! With an rms voltage at its upper 265v limit, the PIV will be a whisker under 750 volts.
However, the Public Supply Utility voltage is far from a perfect sine wave, typically looking like a slightly flat topped sine wave which will reduce the theoretically calculated peak voltages by a few percent.
Even so, it's always best to use an even higher voltage rating, in this case 800v PIV, to allow for spike voltages and sub station voltage controller faults that allow the PSU to go above the 265v rms limit (unless, of course you deliberately choose a 700 PIV rated diode to protect the expensive HT switching transistors against such over-volting events in an act of self sacrifice to blast the safety fuse to smithereens - diodes and fuses are a damn sight cheaper and easier to replace than HT switching transistors!).
The gratitude is appreciated. :-)
If you do decide to post any more pictures, try and compose the shots so as to maximise clarity. Provide two or more angles of view, bending any wires aside if necessary and pay attention to the lighting which can make a huge difference to the clarity of the images.
Digital photography allows anyone to take as many photos as they fancy of such a "Still Life" subject as this. You can easily fire off half a dozen or more at various angles and lighting conditions from which to pick out the best two or three to post up on a web site for us all to peruse and admire.
Oh, and just one final point. As far as I can make out, there's no signs of obvious component damage on that little PCB but, of course triacs and diodes can develop faults without any obvious signs of failure. However, when we're dealing with circuits carrying mains voltages, any such internal failures tend to result in physical damage, either co-lateral or self inflicted since the most likely failure mode tends not to be a fail safe open circuit state but rather a dramatic pyrotechnical short circuiting event.
If as you surmise, there's a triac on the circuit board, it must be controlled from the switch using a couple of resistors to control the triggering delay. Such resistors are typically high value resistors of a low wattage rating. If they've been soldered straight onto the switch contacts, it's possible they may have been damaged by the action of the switch sending shockwaves into the contact tags. If the resistors have been soldered to those contacts via very short leads, they may well have failed as a result of the accumulated cycles of use over the years. IOW, don't ignore such components which must exist somewhere external to that PCB.
I looked at the circuit board by the switch again and realised it is likely to be a speed control for the brush in the Zorb (rotating no vacuum) position, so irrelevant to the problem.
formatting link
Is a picture of what I now think is the power supply to the motor in the head. The power comes in from the let and out to the motor at the bottom. The LED is cropped but at the top. It seems to have an integrated full wave rectifier, some inductances for smoothing and capacitors plus an unidentified black blob
When I apply 230V directly to the power in the LED glows red and the motor does not run.
I have also checked the switch and I think some contacts have burned out but that doesn't account for the fact that the motor won't run when 230V is applied to the board directly.
If the black blob you're referring to is what looks like a disk mounted on edge to the board, that's most likely a VDR to trim any transient voltage spikes.
I see the rectifier module (W08M) of which two of its diode elements form a second fullwave rectifier in conjunction with the two back to back discrete diodes on the left which act as an additional positive output terminal feeding one of the two 33K ohm 1W resistors used as a DC ballast to feed that red LED with ca 5mA of current.
The white oblong component at the bottom is obviously a high voltage capacitor across the motor terminals and this, in conjunction with the two inductors forms an RFI filter to suppress interference produced by the motor's brush gear.
The 2.7Mohm 1/2W resistor (RHS - possibly 2.0Mohm if that dark purple band is actually black) is effectively wired across the 400vdc rated smoothing cap at the top to act as a safety discharge resistor in the event of the circuit having been energised with the motor disconnected or open circuit just prior to being handled.
In which case, it looks like that rectifier module (W08M) must have gone faulty[1]. The inductors are the only components in series with the motor supply and they both look to be in robust full health (they should each show a short circuit reading when tested on the resistance measuring scale of a DMM or MM (mains disconnected!). I'm assuming that the motor is still in full working order as your OP suggested (turning slowly with a few tens of volts applied - hopefully, slowly and *steadily*)
That picture would have been more useful if you had created a mirror image (post processed in the GIMP or Photoshop) or had simply used a mirror to take the photo in the first place!
[1] The only fault I can think of that would still allow the LED to light up (the difference between full and half brightness can be difficult to distinguish by memory alone) would require that both positive diode elements to go open circuit in the bridge rectifier module (W08M).
It says bTh 100 on it, a search suggests it may be a varistor.
I have checked the forward voltages and the integrated full wave rectifier is okay. I thought the other two diodes just rectified the supply for a voltage divider that triggered the LED, I think the LED has two colours, red (at present) and green when working normally.
Yes I see this.
I see the resistor as brown black green gold, 1 M Ohm but I see what you say about a discharge resistor
It turns steadily with 12V and actually runs up fine with 230V ac but I think the above circuit is sensing something wrong with it and somehow inhibiting its supply.
I think this is not the case and that the LED in fact contains red and green elements but not a clue what switches them.
I'm going boss eyed trying to draw the circuit which I can upload if it would help in further diagnosis?
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.