|
||||||||
The 9200 --:-- clock problem. |
![]() |
|
|
Thread Tools | Search this Thread |
|
|
#1 |
|
Forum Member
Join Date: Nov 2007
Posts: 2,669
|
The 9200 --:-- clock problem.
I’ve been interested in this clock problem for a while but as my clock is working fine I didn’t have the opportunity to investigate it. Earlier this year a forum member sent me his failed clock display board and for the last couple of weeks I’ve been investigating the problem. I have found the failure mechanism and repaired his board which has been running in my 9200 for several days doing various test recordings from Standby with complete success. I was going to post my finding early in July after a few weeks of running but earlier this week I had the opportunity to look at a second display board which I have now done and repaired this morning. The failure mechanism on both boards is identical so I am reasonably confident that this is the failure mechanism on the majority of failed boards. I have therefore decided to post now.
One of the circuits on the board is a Real Time Clock (RTC), this is responsible for keeping the time during Standby and also in the event of a power failure. I say power failure but in reality the time will be maintained for over a week, maybe more. One of the components of the RTC circuit is the RTC integrated circuit (chip). This chip contains a very low power oscillator which, with the 32.768kHz crystal, provides the basic time keeping accuracy in Standby. It is this oscillator that fails but not due to any component failure. The input to the oscillator (pin 1 of the chip) is very sensitive and is being affected by random small voltages on the surface of the board. The cause of these voltages is a contaminant on the surface of the protective lacquer covering the board in the area of the RTC. I am guessing the voltages are being caused by a chemical reaction, but the end result is that the oscillator is turned off. The repair is to simply clean the RTC area of the board. Anyone who feels confident to change the board will be able to effect a repair, all you need is cotton buds, a small artist’s paint brush (mine had 2 on it, I assume that’s a size) and a cleaning agent, and of course a fair amount of care. A tutorial on how to change the board can be found on the Hummy.org site. The cleaning agent I used in the lab was ethanol, you probably won’t be able to buy ethanol but surgical spirit looks like being a good alternative. I looked in Boots and their surgical spirit is a mixture of ethanol and methanol so should work fine. Rubbing alcohol contains additives such as perfume so don’t use that. Certainly don’t use acetone based compounds such as cellulose thinners or nail polish remover as they will strip off the protective lacquer and melt any plastic parts they come into contact with. Take reasonable static precautions, most electronic components are quite robust but can be easily destroyed by static discharges. Remove the board and hold it the right way up (display at the top, white cable at the bottom) with the display facing you. The area you need to clean is the left hand end of the board. The 12mm round component labelled CE921 is the super capacitor (think of it as a small rechargeable battery), the black square component with eight legs labelled U923 is the chip and the silver cylindrical component 8mm long labelled Y921 is the crystal. These components together with the very small components above and below the chip form the RTC circuit. If you’ve just taken the board out of the 9200 the super capacitor will be charged and have about 3 or 4 volts across it but this should not be a problem. Start by cleaning the back side of the board with a cotton bud dampened with surgical spirit. Don’t worry about removing any felt tip pen markings, they will smudge immediately the fluid touches them. Clean an area from the left hand end of the board to about 30mm in, pay particular attention to the area around the connections to the super capacitor and crystal. Repeat this several times with new cotton buds until the cotton bud stays clean, but remember you are only removing contamination not trying to remove the protective lacquer. Also note that the contamination may be invisible, as it was on the first board I repaired, but my lab equipment picked up the surface voltages. Now clean the component side of the board being very careful not to cause mechanical damage to the components, although they are quite tough. Hold the board with the left hand end down so any excess fluid can run off rather than into the rest of the board. Again clean from the end of the board to about 30mm in paying particular attention to the area between the chip, super capacitor and crystal. Use the brush dipped in fluid to help clean around and under the super capacitor and crystal. The components above the chip don’t really need any attention but don’t worry about fluid getting on them, just dry the brush on tissue and brush the fluid away. Repeat the cleaning process several times. I have the advantage of being able to detect when the stray voltages have been eliminated but cleaning 3 or 4 times should suffice. Dab off any excess fluid from awkward areas with a tissue and lightly blow dry. The success of this repair will depend on how thoroughly you do the cleaning. I used exactly the above procedure on the second board and I could see with my lab equipment that the oscillations were getting stronger with each cleaning process. I will be pleased to hear your success/failure stories but obviously you do this procedure at your own risk. I would like to thank Martin Liddle for providing the guinea pig board and Coulrophobe for providing the second board. |
|
|
|
|
Please sign in or register to remove this advertisement.
|
|
|
#2 |
|
Forum Member
Join Date: Apr 2006
Posts: 1,197
|
Thank you for your efforts; I have saved this post in case I need it.
You suggest ethanol as a cleaning agent. Would vodka be all right, or would the water content cause problems? What about domestic methylated spirits, or would the mineral content rule it out? I assume that ispropyl alcohol would be all right: this is obtainable from Maplin, and is also the constituent of tape head and other cleaners as a rule, which many people may have. |
|
|
|
|
|
#3 |
|
Forum Member
Join Date: May 2005
Location: wisbech, cambs / norfolk
Posts: 3,834
|
Quote:
...I could see with my lab equipment that the oscillations were getting stronger with each cleaning process...
Thanks Big-Les ![]() Enjoy your holiday, very well deserved.
|
|
|
|
|
|
#4 |
|
Forum Member
Join Date: Nov 2007
Posts: 2,669
|
Quote:
Thank you for your efforts; I have saved this post in case I need it.
You suggest ethanol as a cleaning agent. Would vodka be all right, or would the water content cause problems? What about domestic methylated spirits, or would the mineral content rule it out? I assume that ispropyl alcohol would be all right: this is obtainable from Maplin, and is also the constituent of tape head and other cleaners as a rule, which many people may have. Domestic meths has the blue dye in it so not sure what that would do. Industrial meths would probably be all right but not sure if that would be generally available. Isopropyl alcohol would be ok I'm sure and I would have used it if the ethanol bottle had been empty. We use ethanol to clean optical equipment because it leaves a very clean finish and also I find it best for general cleaning. |
|
|
|
|
|
#5 |
|
Forum Member
Join Date: Mar 2000
Location: Cambridgeshire
Posts: 670
|
Thank you very much.
I am entertaining visitors from Australia over the weekend (and I think that they require something more entertaining than watching me cleaning a PCB) so will try this next week and report back. Peter |
|
|
|
|
|
#6 |
|
Forum Member
Join Date: Nov 2007
Posts: 2,669
|
Quote:
Thank you very much.
I am entertaining visitors from Australia over the weekend (and I think that they require something more entertaining than watching me cleaning a PCB) so will try this next week and report back. Peter |
|
|
|
|
|
#7 |
|
Forum Member
Join Date: Nov 2007
Posts: 2,669
|
Quote:
Absolutely amazing.
Thanks Big-Les ![]() Enjoy your holiday, very well deserved. ![]() |
|
|
|
|
|
#8 |
|
Forum Member
Join Date: Apr 2006
Posts: 1,197
|
Quote:
I wouldn't use vodka because I'm not sure what else is in it.
Domestic meths has the blue dye in it so not sure what that would do. Industrial meths would probably be all right but not sure if that would be generally available. Isopropyl alcohol would be ok I'm sure and I would have used it if the ethanol bottle had been empty. We use ethanol to clean optical equipment because it leaves a very clean finish and also I find it best for general cleaning. ![]() More seriously, though, do you think a spray of switch cleaning fluid such as Servisol might be worth trying, and a lot less likely to cause damage than messing with paint brushes and cotton buds? Would it be better to use a spray cleaner that doesn't contain any lubricant? |
|
|
|
|
|
#9 |
|
Forum Member
Join Date: Nov 2007
Posts: 2,669
|
Quote:
How about a squirt of WD40?
![]() More seriously, though, do you think a spray of switch cleaning fluid such as Servisol might be worth trying, and a lot less likely to cause damage than messing with paint brushes and cotton buds? Would it be better to use a spray cleaner that doesn't contain any lubricant? Cotton buds and a brush will not cause damage unless you're really ham fisted, I purposely was not very careful. Keep the metal bristle holder of the brush away from the board. Don't even mention WD40.
|
|
|
|
|
|
#10 |
|
Forum Member
Join Date: Mar 2000
Location: Cambridgeshire
Posts: 670
|
I haven't been able to download the dismantling instructions from hummy.org, but have had a quick look inside. Do I need to remove the hard drive to access the connector for the clock board?
Enjoy your holiday, and when you return, please can you find out which bit needs cleaning to stop the HD catalogue corrupting.
|
|
|
|
|
|
#11 |
|
Forum Member
Join Date: Nov 2007
Posts: 2,669
|
Quote:
I haven't been able to download the dismantling instructions from hummy.org, but have had a quick look inside. Do I need to remove the hard drive to access the connector for the clock board?
Enjoy your holiday, and when you return, please can you find out which bit needs cleaning to stop the HD catalogue corrupting. ![]() |
|
|
|
|
|
#12 |
|
Forum Member
Join Date: Oct 2005
Posts: 112
|
excellent work
the description you have given is very good but is there any chance of any pictures of the process ? |
|
|
|
|
|
#13 |
|
Forum Member
Join Date: Nov 2007
Posts: 2,669
|
Quote:
excellent work
the description you have given is very good but is there any chance of any pictures of the process ? |
|
|
|
|
|
#14 |
|
Forum Member
Join Date: Sep 2003
Location: East Coast of Lincolnshire
Posts: 318
|
I personally would add to the excellent work of Big-Les by suggesting that the condition of CE921 is thoroughly checked. On the three boards I have examined this 'super-capacitor' has leaked. Could it be that this has contributed to some of the contamination?
I might also suggest that the 32.768khz crystal is removed from the board, its legs (gently) cleaned with fine sandpaper before being refitted. In at least one board, the very gentlest of pulls on this saw it come straight out of the board as the electrolyte from the super-capacitor had permeated the solder joints and the corrosion appeared to have broken down the solder-wire bond completely. The super-capacitor provides the backup power source for the clock in the event of a power cut, it could theoretically be removed or replaced with a standard capacitor of the right dimensions/voltage rating without affecting the normal operation of the standby clock other than how long the clock is maintained during a power cut. As I have not yet been able to get any of the three boards I've looked at working, the above are simply suggestions. I am expecting to have a complete machine donated to me in the near future and I will be able to experiment further. Now Les has identified the cause and cure of the problem I hope I will have some success! Many thanks Big-Les. Last edited by mkstevo : 13-06-2010 at 12:17. Reason: Forgot to say thanks... |
|
|
|
|
|
#15 |
|
Forum Member
Join Date: Nov 2007
Posts: 2,669
|
Quote:
I personally would add to the excellent work of Big-Les by suggesting that the condition of CE921 is thoroughly checked. On the three boards I have examined this 'super-capacitor' has leaked. Could it be that this has contributed to some of the contamination?
I might also suggest that the 32.768khz crystal is removed from the board, its legs (gently) cleaned with fine sandpaper before being refitted. In at least one board, the very gentlest of pulls on this saw it come straight out of the board as the electrolyte from the super-capacitor had permeated the solder joints and the corrosion appeared to have broken down the solder-wire bond completely. The super-capacitor provides the backup power source for the clock in the event of a power cut, it could theoretically be removed or replaced with a standard capacitor of the right dimensions/voltage rating without affecting the normal operation of the standby clock other than how long the clock is maintained during a power cut. As I have not yet been able to get any of the three boards I've looked at working, the above are simply suggestions. I am expecting to have a complete machine donated to me in the near future and I will be able to experiment further. Now Les has identified the cause and cure of the problem I hope I will have some success! Many thanks Big-Les. |
|
|
|
|
|
#16 |
|
Forum Member
Join Date: Dec 2006
Posts: 468
|
Great research Les Thanks
Can I also suggest using Wynns contact cleaner which I use for many various electronic circuits I have used it successfully on things as diverse as old radios to hi tech engine management systems and even used it to repair a friends mobility scooter for which he had been quoted several hundred pounds for a new main board so it should work successfully with the Humax |
|
|
|
|
|
#17 |
|
Forum Member
Join Date: Nov 2007
Posts: 2,669
|
Quote:
Great research Les Thanks
Can I also suggest using Wynns contact cleaner which I use for many various electronic circuits I have used it successfully on things as diverse as old radios to hi tech engine management systems and even used it to repair a friends mobility scooter for which he had been quoted several hundred pounds for a new main board so it should work successfully with the Humax Alcohol removes excess flux and this contaminant very effectively so I can only recommend that. The majority of people are more likely to have surgical spirit (main constituent ethanol) from the chemist than they are switch cleaner. Thanks for your suggestion and if people want to try switch cleaner then that's up to them. |
|
|
|
|
|
#18 |
|
Forum Member
Join Date: Nov 2007
Posts: 2,669
|
Quote:
Great research Les Thanks
Can I also suggest using Wynns contact cleaner which I use for many various electronic circuits I have used it successfully on things as diverse as old radios to hi tech engine management systems and even used it to repair a friends mobility scooter for which he had been quoted several hundred pounds for a new main board so it should work successfully with the Humax Cleans and degreases, leaving no residue. Wynn's Electrical Contact Cleaner is an alcohol based electrical component and contact cleaner. It is formulated to: Remove dust, grease, oxide deposits, tobacco residues, oil and other surface contaminants. It is more expensive than surgical spirit though. |
|
|
|
|
|
#19 |
|
Forum Member
Join Date: Jun 2010
Location: Glasgow
Posts: 2
|
Carried out Big-les' instructions to the letter and it has worked like a dream. Used surgical spirits. The operation is straightforward and can be carried out be those with modest technical skills. Going beyond this and removing components seems to me a step to far at the moment and certainly ups the tech skills needed.
Big thanks to Big-les |
|
|
|
|
|
#20 |
|
Forum Member
Join Date: Nov 2007
Posts: 2,669
|
Quote:
Carried out Big-les' instructions to the letter and it has worked like a dream. Used surgical spirits. The operation is straightforward and can be carried out be those with modest technical skills. Going beyond this and removing components seems to me a step to far at the moment and certainly ups the tech skills needed.
Big thanks to Big-les
|
|
|
|
|
|
#21 |
|
Forum Member
Join Date: Jun 2010
Location: Glasgow
Posts: 2
|
Carried out Big-les' instructions to the letter and it has worked like a dream. Used surgical spirits. The operation is straightforward and can be carried out be those with modest technical skills. Going beyond this and removing components seems to me a step too far at the moment and certainly ups the tech skills needed.
Big thanks to Big-les |
|
|
|
|
|
#22 |
|
Forum Member
Join Date: Feb 2006
Posts: 3,118
|
Quote:
it's been like waiting for exam results for me.
|
|
|
|
|
|
#23 |
|
Forum Member
Join Date: Nov 2007
Posts: 2,669
|
Quote:
I think you passed with distinction this time.
|
|
|
|
|
|
#24 |
|
Forum Member
Join Date: Jul 2003
Posts: 178
|
Big-Les - Congrats on your investigation!
FYI - the thermal shock from an aerosol cleaner is extremely unlikely to damage a solder joint - unless the joint is defective to start with! It is commonplace within the electronics hardware industry to use specialist freezer spray to help with fault analysis and temperature sensitivity analysis. This cools far more swiftly and to much lower temperatures than any cleaner aerosol would ever do and is absolutely safe on standard electronic components and solder connections. (Use on a "hot" PCB assembly and you will get a deltaC of around 100 degrees in 1-2secs - that is absolutely safe on solder joints and would be an order of magnitude higher than you could get with an aerosol cleaner alone). Also, the "flux" residue you mentioned is, these days, water soluble. There was an industry wide transition in the electronics industry in the late 1980s/early 1990s to aqueous based cleaning on environmental grounds. Note that I am NOT suggesting that the contamination you have identified is necessarily flux residue, but that IF IT IS then this could probably be cleaned just with water - and some rubbing/patience as it can be quite stubborn when old! (Also, remember to THOROUGHLY dry the assembly after cleaning - e.g. pat dry and leave on a radiator for a few hours). FWIW, I think the earler proposal to use Isopropyl (Tape Head) cleaner is probably a very good one. This is effectively a combination of propanol and water, which should be capable of cleaning most anything on the board. Finally, you suggested that some sort of chemical reaction may be generating a voltage on a sensitive pin. I *suspect* that it may be more simple than that. The sort of residue you describe typically has a high, but potentially significant resistance. It could provide an impedance path of >100Mohms (sometimes much lower) to a nearby pin. The supercap (mentioned earlier) will be supplying a backup voltage to at least one other pin on the RTC, and your residue is probably providing the means by which a leakage current can get to your voltage sensitive pin. (Which probably has an impedance of >100Mohm). Hope that is of some help. Mark |
|
|
|
|
|
#25 |
|
Forum Member
Join Date: Nov 2007
Posts: 2,669
|
Quote:
Big-Les - Congrats on your investigation!
FYI - the thermal shock from an aerosol cleaner is extremely unlikely to damage a solder joint - unless the joint is defective to start with! It is commonplace within the electronics hardware industry to use specialist freezer spray to help with fault analysis and temperature sensitivity analysis. This cools far more swiftly and to much lower temperatures than any cleaner aerosol would ever do and is absolutely safe on standard electronic components and solder connections. (Use on a "hot" PCB assembly and you will get a deltaC of around 100 degrees in 1-2secs - that is absolutely safe on solder joints and would be an order of magnitude higher than you could get with an aerosol cleaner alone). Also, the "flux" residue you mentioned is, these days, water soluble. There was an industry wide transition in the electronics industry in the late 1980s/early 1990s to aqueous based cleaning on environmental grounds. Note that I am NOT suggesting that the contamination you have identified is necessarily flux residue, but that IF IT IS then this could probably be cleaned just with water - and some rubbing/patience as it can be quite stubborn when old! (Also, remember to THOROUGHLY dry the assembly after cleaning - e.g. pat dry and leave on a radiator for a few hours). FWIW, I think the earler proposal to use Isopropyl (Tape Head) cleaner is probably a very good one. This is effectively a combination of propanol and water, which should be capable of cleaning most anything on the board. Finally, you suggested that some sort of chemical reaction may be generating a voltage on a sensitive pin. I *suspect* that it may be more simple than that. The sort of residue you describe typically has a high, but potentially significant resistance. It could provide an impedance path of >100Mohms (sometimes much lower) to a nearby pin. The supercap (mentioned earlier) will be supplying a backup voltage to at least one other pin on the RTC, and your residue is probably providing the means by which a leakage current can get to your voltage sensitive pin. (Which probably has an impedance of >100Mohm). Hope that is of some help. Mark It wouldn't surprise me if the contamination was caused by some 'environmentally friendly' product used during board manufacture. ![]() As I said in an earlier post, if the ethanol bottle had been empty I would have used isopropanol. Surgical spirit available from the chemist is mainly ethanol and there are far more chemists around than Maplin stores. As for your suggestion that the super capacitor is the source of the stray voltages is, I'm afraid, wrong. Have a look at my post on the other thread relating to the clock problem here. If it had been the super capacitor supplying the rogue voltage then the voltage would have been +ve of ground. I simulated the rogue voltage on a bench lash-up circuit and +ve voltages of several volts from a high impedance source had no effect on the oscillator, however just 0.2 to 0.3 volts -ve would turn the oscillator off. Thanks for you comments. |
|
|
|
![]() |
|
All times are GMT. The time now is 16:47.





