Wavebox Driver Failure

[Steve Atkins]

Came home last might to find a small wave and a periodic humming noise but no alarm. After some wavebox component/connection swapping the problem was isolated to the 7240.27 and a replacement is already on the way.

While the pump still works (on the other wavebox driver) the interior was extremely hot, but still touchable, when I first opened it up. Is this likely to have damaged the pump?

Inside the driver, one of the main transistors along the side has burn marks around it and a hole blown in its side. Is this failure generally a symptom of something else, apart from water damage? There is no evidence of water on either the inside or the outside or the unit.

Once before, this wavebox started to alarm and I found a snail perched in the end of the impeller magnet. Since it went alright again after I removed the snail, I assumed that the snail had caused the impeller to become unbalanced and this was causing the alarm. I was, however, not sure how a small snail could have managed to get on the end of something that spun up ever second or so in the first place. I now wonder if the snail might just have got onto an impeller magnet that already wasn’t spinning.

Steve

 

[rvitko]

The pump was hot or the driver?

The transistors tend to go first and they can go somewhat violently, he usual cause is a load peak and insufficient contact with the heat sink.

Could be but I would more likely suspect the snail entered and the prop stopped, you can stop it with your finger so it doesn't take a whole lot. This could have then caused the load peak as the driver tried to restart the pump.

 

[Steve Atkins]

The pump was hot. After I took out the impellor I put my finger down the hole to see it there was an obstruction and noticed the heat. I then did it to the other pump and the one with the failed driver was significantly hotter than the working one. The impellor magnet was also hot. Despite this, when I connected the hot pump up to the good driver, it seemed to go OK. Just wondering if that heat build up, presumably caused by the failed driver, could have done damage to the pump.

The snail event was several weeks ago and the snail was only about 5mm long. If this did cause a load peak then it has taken a while for the transistor to fail because of it. Perhaps a snail got on it again but had left by the time I got there.

I might apply some fresh heat paste between the transistors and the heatsink.

Steve

 

[rvitko]

Does the pump still run hot though? If it still runs hot I would check with a watt meter and verify the wattage demand of the pump, if it is higher than about 60w it is bad.

I would not recommend modifying the attachment of the transistors. Loosening and reattaching the heatsink is more likely to cause a problem then to solve one. I think the driver must have had some sort of defect but without some readings from a watt meter with your pump it is hard to say if the pump is damaged and the culprit or if the driver was just bad.

 

[Steve Atkins]

The pump has not been running over the weekend but the new driver should arrive today. When it does, I will see if the pump runs hot and check the wattage.

Steve

 

[Steve Atkins]

Looks like it might be the pump.

It is drawing 72W and the other only 48W.

Bad news is that since few waveboxes or 6200s get sold here there are no spare 6200 pumps and I will have to wait for one to arrive from Germany. In the mean time I have left it running with the top off the driver. And am about to aim a fan across it.

Steve

 

[rvitko]

72W is not so high I would believe the motor is defective. Usually a bad motor gives 150+W readings. 72W is only 20% off of the rated 60W and could be accounted for by a bad or dirty drive unit (calcium fouling the cooling hole in the center of the magent or upper bearing sticking to the magnet) or a higher voltage. If NZ is 240V this would definitely account for it if the transformer is for 230V and it would explain the fried driver. The driver can handle peaks of 100W about but should not be pushed past 80W for long periods of time, this causes it to overheat and fail. Perhaps the other pump is a 6100? This would use about 45W but the 6200 should use 60W, definitely not less than 55W. The 6200 prop is 45mm, the 6100 prop is 42mm. Even if the motor and transformer is a 6200, the 42mm prop will make it essentially a 6100.

 

[Steve Atkins]

It is 240 V down here. I measured the Amps on the low (24V) side of the transformer. One got to 1.9 - 2.0A (2.0A x 24V = 48W) and the other 3.0A (3.0A x 24V = 72W).

It seems unlikely the other pump is a 6100 unless someone accidentally assempled one into a wavebox in the factory and managed to put a 6200 front on it as well. I guess I should check. Can the impellors be swapped between 6200s and 6100s?

I shall give both waveboxes a thorough clean out and then re-measure them.

The one driver with the higher wattage pump was certainly running hotter than the other one but you could still touch the heatsink without getting burnt. With the fan blowing through their interiors they are both running quite cold today.

Steve

 

[rvitko]

The output should be 20V. The problem I believe is the higher input voltage is leading to a higher output voltage.

The prop assemblies are interchangeable.

 

[Steve Atkins]

I didn't check the output voltage as I assumed the 24V mentioned on the Tunze web site (or more precisely in the online stream 'Instructions for Use', x6100_8888, 10/2002 - 01/2007) to be correct. That changes things a bit. More measuring.

Steve

 

[rvitko]

24V is just the absolute maximum allowed. The transformer itself should say 20V 80VA.

 

[Steve Atkins]

Without any cleaning so far, just a check and some measuring, we have.

Wavebox 1:
• Transformer output voltage = 22.1
• Max Amps = 1.9
• Pump body = 6260.10
• Apparent wattage = 42.9

Wavebox 2:
• Transformer output voltage = 22.8
• Max Amps = 3.1
• Pump body = 6260.10
• Apparent wattage = 70.7

Next is to clean them and then work out some way to see if the lower wattage pump is producing the same flow as the higher wattage one.

Steve

 

[rvitko]

I would double check wavebox 1, it sounds like it has a 42mm prop, at the end of the prop it is written in small numerals. They should both draw between 55 and 65w. Wavebox 2 pump just sounds dirty or worn.

 

[Steve Atkins]

Worn seems unlikely as they were both installed new in mid January this year. I’ll check the props and give them a clean tonight.

 

[Steve Atkins]

Cleaned both thoroughly, although neither proved to be very dirty (cleaned when opening up after driver failed) and swapped prop assemblies. Wavebox 2 dropped from 3.1 Amps to 2.8 Amps (63.8 W). Was still quite hot inside while other one was just warm in comparison.

Amps on Wavebox 1 did not change.

Both had same prop, with 48 and K on the end.

Steve

 

[rvitko]

My mistake, they should be 42mm and 48mm respectively. I am not sure what the cause is in this case. The operation of wavebox 2 though seems normal now, that is about the wattage I would expect. Is the driver of wavebox one turned down? Have you tried swapping drivers?

 

[Steve Atkins]

Swapping drivers is my next trick.

It is not the wattage that concerns me so much as how hot the inside of the pump gets and how hot the driver heatsink gets when I am not running a fan over it.

From many year in the electronics assemply industry I feel we are dangerously close to fry-up territory. I think a new pump may have already been ordered and may just run the driver with a fan blowing over its innards until then.

Steve

 

[Steve Atkins]

The pump finally answered the question of what was wrong itself, by failing. When I got home last night it was not opperating at all. I think this was due to temperature as it started again sometime later after I had finished checking the the driver was still OK. I have never actually compared the outputs of the two pumps prior to last night but after the restart the high wattage pump was found to be only putting out about 1/4 (at most) of the flow of the other one.

I have shut it down now. Interestingly, I am still getting about 3/4 of the wave height that I was before with both pumps going, which now makes me wonder if the failed pump has been in trouble right from the beginning.

A new 6200 pump is on the way.

Steve

 

[rvitko]

Hmm, that is odd, I have seen pumps like you describe, usually half the coil is dead, from an overload or defect. They usually use a much higher wattage so that makes it obvious that a problem exists. Usually double or more of the rated wattage.

 

[Steve Atkins]

A new 6200 pump, complete with driving electronics, finally arrived for the failed Wavebox. It operates at pretty much the same wattage (38W) as the one that did not fail. Actually very slightly less, probably because it is cleaner.

This wattage assumes that the clamp meter I used round the grey wire from the transformer is accurate.

Despite a corner overflow I am now getting a wave between 2.3 and 3.0cm with the bigger wave when the 6100s are on the low power part of their cycle.

Steve