Bosch SGS43CO Dishwasher Problem

 Most washing machines and dishwashers eventually start to go wrong and one is faced with fixing it (or getting someone else to fix it) or ditching it and getting a new one.

As with most things, dishwashers will have stock faults or, in other words, design weaknesses that eventually let them down. We bought our first dishwasher in 1978 and it had lasted over 30 years when we replaced it with one of the same make, Bosch. The new one is a lot different, having a microprocessor and it's also mechanically very cleverly designed. Lots of secret clips and interlocking plastic bits.

I say cleverly designed, but in fact, unless you've dismantled one before it's tricky to get apart. Anyway, I won't get into that.

 

 When our new dishwasher first began to give trouble, the symptom was not doing a decent wash. We diagnosed this to insufficient water, putting it down to low water pressure.

I opened the stop cock a little more. The tap tends to leak if fully opened, hence it's usually turned half off, which is why we suspected this to be the cause of insufficient water in the dishwasher.

Oddly, having upped the water pressure, the dishwasher went completely wrong. This, it turned out, was just a coincidence.

On the last day of April 2015, after turning on the machine for a normal wash, everything sounded more-or-less OK, all the normal noises were present except maybe sloshing water.. lots of gurgling but no actual sloshing and, because there was no water in the machine the dishes came out unwashed. What was happening?

I searched on the Internet for a clue. Now, this particular dishwasher model (SGS43CO) in the world of Bosch dishwasher users is pretty famous, or at least "infamous". Would you believe a dishwasher could burn down your house? Well, it's said, this model burned down several. It went like this... the water heater is driven from a small relay fitted to the control circuit board. Eventually this relay starts to degrade, it's contacts burning and developing a high resistance. The high resistance, passing current to the heater generates lots of heat. The heating of the relay contacts works its way to the solder joining the relay to the circuit board. Over a period of time this heating, followed by cooling when the relay is turned off, causes metal fatigue and the solder joints crack. The joints develop resistance and the current through the degraded solder joints heats the circuit board so that eventually it catches fire and burns. The flames heat the plastic surrounding the circuit board. The plastic catches fire and this spreads to the plastic front of the control panel, and if the dishwasher is located under a worktop, the worktop catches fire and so on until the house is burnt down.

Lots of people turn on their dishwasher when they're going out, and when some people arrived home they discovered their house had burned down. Some of these people discovered the reason and so Bosch had to publish a warning for owners of certain specific models having a serial numbers in a particular range to contact them and have their circuit board changed for a new, redesigned version. I checked our model number then the serial number...oops. I called the Bosch phone number and they sent out a chap with a new circuit board which he fitted so our house didn't burn down.

Back to my problem. My Bosch dishwasher failed to wash dishes so I thought it might be the circuit board. Once bitten, twice shy they say. I checked the circuitry and it all looked pristine. Removing four relays showed a couple had contact resistance of a few ohms so I changed them and refitted the board. No luck... the fault remained. The Internet had lots of examples that sounded exactly like mine. One suggestion was the water inlet pipe... that was fine. Another the water inlet valve. I checked it and found it was also fine. Next a large complicated water inlet manifold. This is a thing about a foot or so high and nine inches wide with lots of pipes and plenty of gunge. I removed the gunge and refitted the manifold... the fault remained. Next another plastic thing with a float and a pair of microswitches. Microswitches are always going wrong, but these were fine...

After much testing and poking I decided to watch what was happening. With the door cover removed and the left side panel removed I turned on the machine and carefully watched what happened. The drain pump ran... then the water valve opened and water flooded in for about one second then stopped. Surely not enough to fill the machine.

At this point I had a brainwave. I was working in the dark. I removed the low energy lamp in the utility room light socket and fitted a proper old fashioned 60 watt bulb. Now things were changed. I got oodles more illumination and I could actually see water levels in the plastic pipes. No wonder those low energy lamps were given away free to old age pensioners. They're rubbish (I mean the low energy lamps). Anyway now I could see what was going on. What was happening was the float chamber immediately filled with water... the microswitch controlling the inlet valve activated and switched off the water inlet valve. But, as I peered at the float chamber, it ever so slowly emptied. This was now many hours into the exercise and my brain was tired so I gave up and had my dinner.

As I sat thinking about the days results I pondered... the circuit board turns on the inlet valve... OK.. that was happening ... water rushed in and immediately shut off the valve... why?

Suddenly it came to me. Water filled the float chamber but surely it should have first filled the machine with water. The float chamber was only there to monitor the water level in the machine and clearly it wasn't because the machine had no water in it. Obviously the pipe from beneath the float chamber... a pipe of around one inch diameter should allow water to pass through it... and it wasn't therefore it must be blocked.

I thought a bit more... Bosch engineers had arranged the microprocessor to only activate the water inlet valve for a limited period. Say 2 minutes? During this period the machine would fill with water and, once filled, the float chamber would contain just the right amount of water and the float would have risen and cut off the current to the water inlet valve. In my machine, after the sudden filling of the float chamber within about one second the float slowly sank, releasing pressure on the microswitch and closing the inlet valve circuit. This took about 4 minutes. By then of course the 2 minute period had expired so no extra water was allowed into the machine. This clearly is a safety measure preventing a serious leak from flooding one's house.

I pulled off the one-inch pipe and it was completely blocked. After poking a drain cleaning device into the pipe, with the cover and filters removed from inside the main chamber, where the other end of the one inch pipe emerged, there was initially lots of resistance then it was free and a nasty brownish yellow smelly gunge escaped.

I refitted the pipe and filters then turned on the machine. It filled with water. After putting a special cleaning cansister inside and turning on the machine for a normal wash...it worked perfectly. Looks like our dishwasher got a reprieve, but only just...

Being wise after the event is easy. Over quite a long period our dishwasher had been acting strangely. My XYL had remarked that there didn't seem to be enough water in the bottom of the thing and this might explain why dishes were not perfectly washed. She'd solved the problem by opening the door after a time which she had calculated to the minute and poured in an extra two pints of water. This worked OK although I had been continually nagged to sort things out. I'd in fact shelved the problem with a mental note about low water pressure because the cold supply tap leaked when fully open so had been partly closed to the point where the leak had stopped. This wasn't the issue however. What was happening was this... the water feed pipe into the machine had slowly become blocked. Water flow had been reduced and only a limited flow of water was taking place to the point where the fill period had expired. At some point the float had cut off the flow, but if the fill period was still open more water wouldn't then flow in. This balancing act had, over a long period, meant that less and less water was admitted to the machine and dishes had progressively washed less well. This process had proceeded until the water level in the machine was below the level of the circulating pump pickup tube. Once this point was reached the circulating pump just made gurgling noises because only air was being sucked in.

At least the exercise which all took place on Friday 1st May 2015 means that I now know precisely how to dismantle our dishwasher and extract its major parts so I should be able to keep it going to match it's predessor's longevity.

Returning to potential fires due to burnt relay contacts.... In my repair business I see loads of examples of burnt circuit boards. Usually the material from which the board is made isn't really combustable and burning just produces charred material. It's not just relays, but sometimes a design failing can result in catastrophic damage. Yesterday I had a small circuit board with a melted integrated circuit (an NE555). This was fitted in a DIL socket which had also melted. What was the cause? Well, the board was supplied with a "nominal" 12 volts AC from a transformer. This was rectified and smoothed and ended up as 19 volts DC. A 5 volt regulator (actually an LM317) was helped to deal with excessive temperature by the addition of a 4.7 ohm ballast resistor. This is a good idea in the circumstances, however, although the resistor was a wirewound thing in a square section ceramic case, it still got very hot. The board material underneath the resistor was charred, but more importantly, the track joining one end of the resistor to the circuitry was open. The wire ends of the resistor had been exceedingly hot and the copper track to which they were soldered had eventually lifted from the board. Continual heating and cooling had then resulted in metal fatigue in the copper track which had caused it to break. The regulator had then produced, not 5 volts output, but virtually the whole DC input of 19 volts. This was in excess of the maximum rating of the NE555 and a second chip, a CD4011. Thankfully there are no other chips involved.

Some circuit boards I've seen have supported a small fire especially when parts are very close together. Generally speaking a fuse will blow and all will be safe, however, in a large appliance, such as a fridge freezer the fuse will be 13 amps because it needs all of this current in order to handle its operation. I must say though that now solid state relays are available there's no excuse for fires to result from burnt relay contacts, especiallly when these relays are switching high currents. Another component sometimes used in a fridge freezer is a device identical to that used in old colour televisions. This is a component which is designed to regulate the surge current at switch-on and then to stabilize the working current. Sometimes these devices, which use a carbon disk clamped between metal plates, can fail and I've seen a few catch fire.

This happens when the carbon disk disintegrates resulting in the thing getting extremely hot, melting its solder connections and resulting in dry joints. Arcing can then occur and hence a fire might result. Thankfully this is rare and a fuse will usually blow before anything serious happens. That is unless the fuse has been replaced by a wad of silver paper or a nail...

Before I close this topic, I'll mention that old problem introduced by the Government and the EU. Continental mains is 220 volts and UK mains is 240 volts, give or take. Three phase equipment is often marked 380 volts corresponding to its use on the continent. UK mains can be 440 volts. Within a lift controller there's a smoothed, rectified DC voltage of whatever is the peak value of the RMS input. Quite a large difference if you work out the numbers.

Now, take a block of flats. The mains voltage delivered to the 13 amp sockets should be 230 volts according to Government "aspirations" because years ago the Government and the EU decided to split our differences and declare the mains across the EU should be 230 volts. The voltage within a UK block of flats is therefore supposed to be 230 volts. Not really because, first of all it might be way over 250 volts because the local electrical plant in use was installed before "harmonization" and still be within permitted UK tolerances. It could be even higher though. Let me explain... power stations provide 3 phase mains to areas of population and do this in a really clever way to save on copper. 3 phases= 3 wires... it should be 4 wires (an extra wire for neutral) but one can usually eliminate the 4th wire if consumers all use, on average, the same amount of electricity. Maybe they do, in which case their mains voltages will be 240 volts. However, the end result also depends on the work of electrical contractors. These chaps need to ensure that roughly the same number of consumers is wired to each of the three phases supplied to a particular area, and no consumer extracts more than his nominal share of the available current.

If the 3 phase mains is balanced all is well. If the three phase mains is unbalanced then one phase may be very high and the other two very low and this will be reflected into the single phase mains supply. One set of consumers will need to replace their light bulbs frequently whilst others complain about dim lights... but at least they last for ages. I recall a lift controller in a London block of flats was forever failing. I worked out that this was probably due to poor three phase connectivity, possibly too many consumers wired into one leg of the 3-phases or one particular consumer perhaps nurturing an Indian Hemp farm.

Lots of electrical equipments include a protection device which will cause a fuse to blow if the nominal mains voltage is exceeded. Many protection devices were determined by the designer to fit in with their local mains spec and if that is say Germany it would be 220 volts. I've seen loads of protection devices fail in the UK because their rating is marginal at 240 volts (sometimes going on fire) and when the associated fuse is big this can be serious. The fuse in a lift controller can be a whopping 90 amps because it's protecting a consumption of tens of kilowatts. Given an unbalanced input of say 460-390-390 you'll have a problem. Returning to the subject of light bulbs. New, low energy, high efficiency lamps use switching power supplies buried in the plug part of the lamp. In a cheap light bulb the reliability of these switching power supplies can be poor and even result in a plastic casing running at well over 100 degrees centigrade. Some that I've recently replaced were charred and this is not good because, sooner or later, the wrong plastic will be used which will combust and set a house on fire... (click to see this)

 

 

Return to Reception