Jet AFS-500 air filter PCB repair – help please.

[Jack (W)]

Last December my Jet AFS-500 air filter developed a fault causing the motor to run in an underpowered, sickly sort of way. Whilst the remote control commands were acknowledged by a ‘beep’ from the sounder on the PCB, the appropriate LEDs only illuminated dimly and they flickered rapidly (50Hz?).

I spent £58 on a replacement PCB from Axminster, and so far the air filter is working fine. But as the replacement PCB is identical to the one with the fault (i.e. not a new improved version) I thought it might be prudent to repair the faulty one if I can, but I could do with a bit of help please as my knowledge of electronics is rather rudimentary.

The faulty PCB is on my workbench and for safety reasons I’ve no intention of connecting it to the mains at any time other than when I put it back into the air filter to see if I’ve rectified the fault.

I’ve examined the PCB (no visible faults) and I’ve produced a partial circuit diagram showing the components between the mains AC input and the low voltage output to the logic and control section of the circuit (5 volts DC).

To try to determine if the fault lay in the logic and control section I hooked up my bench PSU to points TP1 and TP2 on the circuit diagram, set the output to 5 volts, and tested the PCB for responses to commands from the infra-red remote control and from the PCB mounted buttons.

The responses were perfect. The appropriate LEDs illuminated at full brilliance with no flicker and the buzzer ‘beeped’ in response to all control commands.

This led me to believe that a stable 5 volt DC supply was not being created by the PCB components, and so as a first stab at fixing the fault I replaced C8 – the 1,000uF 16v reservoir capacitor, to see if that was the culprit – but it wasn’t.

My gut feeling was, and is, that the fault lies closer to the mains input, and so I checked those components that I could.

Diode D1 (1N4007) works correctly.
For the time being I’m assuming Zener Diode ZD1 (5v6 1W) also works correctly.
R15 resistance is correct at 200k Ohm.
The combined resistance of R1 and R1A, which are wired in parallel, correctly measures 50 Ohms.
The fuse is intact.
The resistance across resistor VAR measures 5M Ohm, which I take to be correct.
For the time being I’m also assuming yellow capacitor C1 works correctly.
I do not know what component BF1 is, but I guess it might be a filter, and for the time being I’m assuming it works as intended.

All of which leaves the large black rectangular capacitor C2 – marked “105k 400VAC” as my prime suspect at this stage.

Unfortunately I can’t find a 105k 400VAC capacitor in the Farnell / CPC /RS / catalogues – least, not as I recognise it, and so I wondered if anyone here might know where I can get one.

I’d also be grateful for any comments, suggestions or advice you may be able to offer.

Many thanks,
Jack

 

[Racers]

105K is a 1uF capactior.
I think your diagrame is wrong a fuse across the supply can't be right.

Pete

 

[Jack (W)]

Pete, thank you for your reply.

I’ve used Photoshop to superimpose the fuse and component VAR on top of a view of the tracks, so I hope you can see that my circuit diagram is correct.

Component VAR is a CNR07D471k Metal Oxide Varistor, and I believe it serves as a safety device in the circuit.

As I understand it, under normal operating conditions the Varistor presents a very high impedance, effectively isolating the fuse from Mains line AC 1.
However, should the mains voltage for some reason become excessively high, the Varistor impedance will fall dramatically and place the fuse across Mains lines AC1 and AC2, and as a consequence the fuse should blow. Once that happens I imagine the Varistor will return to a high impedance state.

Jack

 

[Racers]

O/k
Well change C2 and see if it works, connecting mains to it and checking the voltage across the zener would be my next move, dangerous I know, do you have a isolalation transformer?

Pete

 

[Woody2Shoes]

I suppose the fuse makes some kind of sense if AC2 is the phase(live) and AC1 is the neutral.

Triacs are not indestructible IME - they don't like any kind of inductive component in the load. I'd try and test them - here's some clues: http://www.circuitstoday.com/how-to-test-a-triac

Cheers, W2S

 

[Jack (W)]

Pete,
Unfortunately I don't have an isolation transformer, and I can't find a suitable 1uF 450VAC capacitor (or higher rated) in the UK at a sensible price. An EPCOS B32674D8105K capacitor from Farnell (only found by searching for that manufacturer's code) costs £7.55 plus £15.95 handling charge due to it being "US Stock" - plus VAT.

I can buy the same EPCOS B32674D8105K capacitor from DigiKey or Mouser in the USA for £1.62, but the shipping cost is £12.


W2S,
Thanks for your timely guide to checking triacs. I've now done that and they're fine.


Circuit Diagram Correction: Component D1 is a 1N4007 diode, not a 1N4001 as indicated on the diagram.

Jack

 

[Skeety]

Hi,

Here is a link to an EPCOS one, 1uF rated at 630V delivered for just over a fiver:

http://www.ebay.co.uk/itm/EPCOS-B32674D ... 2a593a4200

Hope that's of use, good luck with fixing it.

Jon.

 

[Jack (W)]

Jon,
That's really great - thank you.
Jack

 

[Skeety]

No worries,

It's been many years since I have done PCB repairs but your logic is sound. Something is pulling the 5v low, most likely a cap.

However, due to the board being in a unit that suffers vibration and dust I would inspect every soldered joint under a magnifying glass. To be honest, for the size of the board I would work through it, de solder and re-solder every joint. Looks like it's been flow soldered originally and I have seen faults induced by these joints when used in environments with heat\vibration\dust. Even if it doesn't fix it you know then that it's not a resistive joint causing issues and something else to rule out

Cheers,

Jon.

 

[gnelson]

Hi Jack,

Did the EPCOS capacitor fix the problem? I have the same model, and it has developed the same fault. It is only about 18 months old, and can only have done a few hundred hours. The motor all looks fine, so I am suspecting the PCB. It would be good to know if the capacitor is the likely suspect- my soldering skills can strectch to that!

Thanks,

Glyn.

 

[Jack (W)]

Hi Glyn,

Welcome to the forum. Sorry to hear that your Jet air filter has developed the same fault.

A highly qualified “Yes” in answer to your question - in that the black capacitor C2 was undoubtedly the faulty component on my PCB, for replacing it with the EPCOS B32674D6105K capacitor restored the PCB to proper working order.
I determined this by electrically connecting the repaired PCB to the fan unit and operating the remote control buttons over a period of ten minutes or so, though I did not run it long enough to check the 2/4/8 hour time-out periods.

Now I’ve refrained from posting a claim that I’ve fixed the fault because there are a few aspects about this ‘fix’ that bother me.

I don’t understand the circuit diagram sufficiently well to know if capacitor C2 is effectively across the Live and Neutral mains input, but I think it almost certainly is.

I note that the faulty capacitor was rated at 400 volts AC, whereas the EPCOS capacitor is rated at 630 volts (OK in itself) but DC. If I’ve picked up the gist of the data sheet correctly (a little late in the day perhaps) then it seems the EPCOS capacitor has been designed for use in DC circuits, not AC.

So although the EPCOS capacitor clearly worked in this circuit during my limited testing, I do not know if it will be prone to failure, nor do I know what the consequences of its failure might be.

Really wish I could confidently say: “This is what you need…” but I’m sorry, as yet I can’t .

Jack

 

[RogerS]

I too had a problem with this PCB on my Jet. It's been replaced but every so often, even though switched off, the motor will buzz a bit and sort of start to rotate but then stop. Given both of our experiences I reckon this is a duff designed PCB.

 

[stuartpaul]

Me too!

Just installed a replacement PCB provided (FOC) by Axminster for a 14 month old device.

It has sorted the problem but with less than 70 ish hours of operation I'm sort of wondering how long before this ones goes as it looks identical?

The nice man from Axminster said the sales to faults ratio wasn't big enough to warrant concern on their part but nice to know they are as always happy to support.

 

[Racers]

Hi Glyn,



I note that the faulty capacitor was rated at 400 volts AC, whereas the EPCOS capacitor is rated at 630 volts (OK in itself) but DC. If I’ve picked up the gist of the data sheet correctly (a little late in the day perhaps) then it seems the EPCOS capacitor has been designed for use in DC circuits, not AC.

Jack

All capacitors work on AC signals so don't worry about the DC rating, as long as the voltage across the capacitor isn't over the maximum then its going to be o/k.


Pete

 

[gnelson]

Thanks for the info. I spent a good bit of time reading about capacitors and how they are rated. Yes, using Ac is fine, it is just whether the manufacturer has tested them for AC or not. The conversion from DC to AC is a bit complex, partly depending on the material type of the capacitor. I found a polypropylene example on RS (part 441-9701) which said it was 630 VDC and 370vAC (incorrectly labelled as both AC on the sales page, but explained correctly in the pdf), which means that it would be a bit low. I've found and ordered part no. 882-9313 from RS, which I think should do the job fine (750V DC). It was the largest VDC rating that will fit in the space on the PCB and not too deep to prevent me refitting the board in the housing.

I'll see how it goes.

Glyn

 

[RogerS]

.....
I don’t understand the circuit diagram sufficiently well to know if capacitor C2 is effectively across the Live and Neutral mains input, but I think it almost certainly is.

....

Jack

You are correct and I think that by replacing it you did fix the problem. Capacitors have an impedance (think of it as a resistance to AC) and that depends on the value of the capacitor and the AC frequency. The impedance of a 1uF cap. at 50Hz is about 3k and so in conjunction with the other circuitry sort of acts as a potential divider with the 240v mains being split across the capacitor and the rest of the circuit that derives the rough DC (D1) and then the more accurate DC by the zener (ZD1). So if C2 had gone open circuit then that 3k impedance is gone and in its place is that 200k resistor ...which leaves precious little voltage for generating your 5v.

Hope that makes sense.

 

[Jack (W)]

My grateful thanks to everyone for helping me resolve the problem with the Jet PCB, and for putting my mind to rest over the suitability of the replacement capacitor.

Jack

 

[gnelson]

I'd like to echo Jack's thanks to everyone. My replacement capacitors arrived from RS today. A quick spot of soldering and my air filter is working again! If it is a fault with the PCB design, RS only sold the capacitors in packs of 5, so I have another 4 to go through before havign to buy more. Definitely cheaper than a replacement PCB from JET, and I'd rather not throw good money after bad to Jet, given the number of people with the same fault on this model. For info for anyone else, I bought Capacitor PP B32674 1uF 750Vdc, Stock no.: 8829313 from RS supplies.

Much happier getting back to my carpentry with the filter running again! Cheers,

Glyn

 

[chrisrm]

Add another to this most helpful thread; She Who Must Be Obeyed's AFS-500 filter is back to running properly again, thanks to this bit of work. I found the legs on the RS capacitors too short to go through the PCB, so chopping the originals off close to the dead cap and then soldering to the legs worked a treat. =D>

Many thanks,

Chris M

 

[Jack (W)]

Hi Chris,

Welcome to the forum, and thank you for taking the trouble to add your experiences to this thread. I’m really glad that the help provided by members here has enabled you to repair your air filter too.

Jack