LG Dehumidifier

[Larry2]

I have a Goldstar DL504EL dehumidifier I bought in June. When I first installed it, it collected well over a liter per hour of water. After some time owning it, I noticed that it would go into what seemed like a defrost cycle. I hadn't noticed the sound before, but I never paid attention either. Now it doesn't collect any moisture at all. I pulled the covers from the unit and checked a few temperatures. The liquid line from the compressor is only 100F. The exit from the evaporator is only 46F. I would have expected much colder. There is a sensor on the suction line after the evaporator that connects to the brain. I don't know if the brain is looking for cold temperatures to start the defrost or high temperatures to end it. It goes into a defrost cycle after the line has been at 46 F for a few minutes. There are no other sensors to trigger this cycle, so I'm curious if someone could explain how a defrost is triggered. When defrost kicks in, a solonoid reroutes gas and switches off the fan. Defrost takes around a minute. And defrost starts with no visible frost. Nothing in there is cold enough for frost to form. I would have thought something near 32 F would have started this action.

I measured amperage draw for the compressor at 2.5 Amps/120 volts for this 50 pint/day dehumidifier. The nameplate says the rating is 5.5 amps. I was thinking of putting gauges on but of course there's no service fittings. The recovery stubs are smashed and bent up so they are not suited for recovery or afixing gauges or service fittings. As far as I'm concerned, this mangled bent and squashed stub doesn't meet the intent of the EPA which requires provisioning a recovery stub. It's only a port usable by the factory one time when they sealed the unit off during manufacturing.

My guess is the unit is low on gas. I'm not sure it will be easy to find or repair what's leaking. I looked for warranty service. The instruction manual said go to http://www.lgservice.com. Once there, I select the USA site and of course that site is as dead as my unit.

Is this par for the course with LG/Goldstar? I think I got one month's use. They must have shipped it really quickly for it to be working when it arrived but not a couple of months later. Or perhaps the compressor is bad internally. Compressor part number is 2520UCAA014

[Brian_UK]

I don't know what your Sale of Goods laws are in the States but would suggest you investigate a replacement from your supplier as your first action.

[Larry2]

Yes, well you know how that goes with eBay. :o One day I'll learn about getting what one pays for.

I've found an email address for LG Electronics in the US and dashed them off an email with the s/n of the unit. I've asked them to swap it out. I'm not expecting much, but who knows? The unit was brand new when it arrived. I'm wondering if it was refurbished once at LG by gassing it up instead of repairing a leak.

[Larry2]

Well, I spent some quality time this afternoon with my LG dehumidifier searching for leaks. I didn't find any but it has to be there somewhere. There is no sign of oil anywhere. I unwrapped all the foam covered connections too and there's no oil escaping anywhere. From what I read here using the search, dye may not help much if the leak does not include oil. What is the most probable place to find leaks? Where shoule one look first? Do new coils ever leak other than at the braze connections?

Here is a photo of a braze connection to the evaporator coil. The top tube should stick out as far as the next one down, but the production brazer who fitted the connection erroded the copper line half way down it's length using excess heat.

http://i9.tinypic.com/2wpl4iv.jpg


Here's another view of that connection. You can see the aluminum fins took some damage too. Maybe this was thought to leak so they poured on the heat for a second try?

http://i9.tinypic.com/2u9o5tz.jpg



The coil tubes are brazed with U caps at one end. The other end has no brazed connections, just solid copper loops. The brazing of the actual coil assemblies looks quite nice. Actually, the unit seems to be nicely designed but suffered in the workmanship.

Each braze of the coil assembly has a smooth fillet with no junk or debris. I assume those connections were mechanized somehow, perhaps with solder preforms. Then there are the interconnections to the coils, connections to the compressor, defrost solonoid, cap tubes and that sort of thing. Some look okay but others look like there is no notion of controlling heat or solder flow.

I assume one would start inspection with these production line connections. I spent an hour or two with soapy water and an acid brush but I couldn't prove any of them leaked in spite of poor appearance. The one pictured above isn't leaking. I looked for quite a while at various components without proving any leaked.

Does the plastic electrical feed through connection on these compressors ever cause leaks? I read on a forum one should stand clear of those because they have been known to blow out like a champaign bottle cork. Is this a place to look or is it unlikely?

There is an inexpensive solonoid for the defrost function. I expect it has to have an internal o-ring gasketing the pintle. Is this a likley source of leaking or are these parts generally pretty satisfactory?

http://i9.tinypic.com/2u7ul4p.jpg

Maybe I'm jumping to conclusions about leaks. Maybe it's not leaking but the fault is with the compressor. I guess the next step is to install fittings so I can get some readings.

Anyway, where do you guys search for leaks first?

[Brian_UK]

You need to be aware that you are searching for a leak probably at 'standing pressure' only. This could be below the level at which the leak, if any, occurs.

As it happens I condemned a compressor last week when it began leaking from the electrical terminal connections, you get something new every day.

The type of oil used now does not always show as an 'oil leak' like the earlier ones did.

[Larry2]

Of course you're right Brian. It was wishful thinking on my part that I would find the leak (assuming it's problem is a leak) through inspection with the unit switched off. Since soap and water were around, I didn't want to power the unit. It probably needs to be operating at least with what charge remains while searching.

In charging this sort of system, would you charge to get a reasonable superheat on the evaporator? I was thinking charge this to 8 degrees of superheat and see how well it works. It used to rain water out of the garage air a few months ago when it worked.

Tomorrow, I get some line taps so I can access they system with my gauges. It will be interesting to see how much gas the unit is short. I think the unit worked a few weeks, at the most.

Of leaks that happen at high pressure but not modest pressure, what are common places to look?