• The forum software that supports hummy.tv has been upgraded to XenForo 2.1!

    This upgrade brings a number of improvements including the ability to bookmark posts to come back to later. Please bear with us as we continue to tweak things and open a new thread for any questions, issues or suggestions in Site/Forum Issues.

Desoldering the DC power socket

batteryman

New Member
Hi, I'm trying to replace the DC power socket on my HDR2000T - it will make very intermittent contact when tried with the board out of the case, testing with a continuity meter, but not at all in use.

I've tried to desolder the lugs on the socket but the solder doesn't seem to want to melt. Are Humax using some different type of alloy compared to normal? I haven't run into this problem before, even with lead-free solder.
 
OP
B

batteryman

New Member
Thanks for the reply, will try to post a photo tomorrow. It's a fairly standard pcb mount socket with three lugs soldered through the board to the underside. It's mounted on the back RH corner of the main pcb, viewed fom above. The solder doesn't seem to melt for my iron, even with a blob of molten solder on its tip.
 
OP
B

batteryman

New Member
Multicore solder with flux, the iron is 15 watts I think and has tackled bigger jobs than this. If it comes to the crunch, I'll carefully cut the old socket off, leave the shortened lugs in the board and use a chassis mount socket with short flying leads soldered to them.
 

Black Hole

May contain traces of nut
That sounds about the same as I use for most jobs - the standard 15W Antex - but I also have a 25W and a couple of regulated/adjustable irons in the armoury (including one from Lidl/Aldi I've not tried out yet).

All I can think of is that your iron simply isn't powerful enough to heat the larger thermal sink above lead-free melting temperature (which is higher than tin-lead solder, making technique even more critical to avoid overheating components or circuit boards). A bog-standard soldering iron is only regulated by the wattage of its heater element versus the heat loss by radiation, convection, and conduction, so the temperature of the tip drops when placed in contact with the workpiece, and won't recover fully because of the extra thermal load (radiation and convection from the workpiece added to that of the iron itself). If the extra load is enough, the temperature may never get there... and keeping the input of heat going for too long risks damaging the workpiece. Ever tried soldering a wire outside in a breeze?

For other readers:

The soldering iron bit (or tip) is also important: a narrow point designed for fine work has a high thermal contact resistance so restricts the flow of heat to the workpiece. My preferred bit for general use has a bevelled "chisel" tip, which can be used on the flat to provide a good contact (or with its point for finer work).
Flux is important, because (when heated) it takes away surface oxidation to ensure new solder can merge with the base metal or old solder. If the solder you are using does not have flux included within it ("Multicore"), it is necessary to add flux separately otherwise the soldered joint will be unreliable (unless the surfaces are very clean indeed). Soldering works by creating a very thin layer at the surface of the base metal where the solder is alloyed with the base metal. Any contamination will interfere with this, and it is best to clean surfaces as much as possible first, even when using flux. The base metal is usually copper, sometimes brass - but if it looks silvery there is probably copper underneath with a solder coating (or at least a solderable coating).

Assuming the above is the case, and assuming you don't have an alternative iron, there are a couple of things you might try:

1. Add tin-lead* solder to the existing joint, which will gradually reduce the melting point of the existing solder by contamination. Add some, remove it (you have a solder sucker?!), add some more, remove it... Each pass should remove a little of the lead-free solder until you can desolder the component.

2. Reduce heat loss by warming the area around the workpiece, eg with a hairdryer (and keep it running so the surrounding air is also warm).

* Lead solder is only legal in very specific situations - including rework of an existing assembly originally made with lead solder (which lead-free solder won't work on). However: the regulations are only really enforcible for products sold commercially; lead-free solder is much trickier for DIY, and one has to use soldering tools that have never been contaminated with lead solder; and I made sure I have a lifetime supply of tin-lead multicore before it became difficult to get!
In my view, taking lead out of solder to please some environmentalists worried about lead contamination of waste tips, instead of controlling the disposal of electronics assemblies, is like throwing the baby out with the bath water.
 
Last edited:
OP
B

batteryman

New Member
Just soldered a job on some Black and Decker garden shears and it was thermally bigger than the Humax job. I'm wondering if there's a coating on the joints stopping them wetting and heating up.
 

Black Hole

May contain traces of nut
A conformal coating you mean? I think you would see it, and it's not likely Humax would spend money on coating when target environment is sheltered domestic.
 
Just soldered a job on some Black and Decker garden shears and it was thermally bigger than the Humax job. I'm wondering if there's a coating on the joints stopping them wetting and heating up.
I suspect it is more a case of the internal earth plane inside the PCB acting as a heat sink. My experience is that a 15w iron will struggle to heat that sufficiently to melt the solder even if the PCB is getting too hot to handle.

Another thing I have come across, this time attempting to replace a USB connector in a Kindle, is that the through hole components such as connectors are fitted with a higher temperature solder so it doesn't melt and have components fall off when the PCB is sent through the reflow machine to solder all the surface mount components.
 
OP
B

batteryman

New Member
You must have read my mind; I had a mental picture of the socket with its solder already done, sitting in an oven to solder the surface mount components. I think the solution may be to cut off the plastic from the original socket, shorten the lugs and solder leads on going to a panel mount socket. No rush, I got another HDR2000T on Ebay last week for a tenner and it's working fine. Thanks for the replies.
 

Ezra Pound

Well-Known Member
I think the solution may be to cut off the plastic from the original socket, shorten the lugs and solder leads on going to a panel
This is a much safer solution than hacking out a section of the PCB, If it is multi-layered you could be doing all sorts of damage to tracks you can't see
EDIT
I may have mis-read post #5, I would keep the lugs (from post #10) as long as possible, this will make solering with a small iron easier as they will get hotter quicker
 
Last edited:
OP
B

batteryman

New Member
May be able to join the lugs from the panel socket directly to the old lugs if I'm careful cutting them. I prefer to do as little as possible on modern pcbs, they're just a bit too delicate.
 
OP
B

batteryman

New Member
Just thought I'd come back to say I recovered my HDR2000T. In the end I removed the pcb and cut away the plastic on the old power socket leaving just the metal tabs. I could now see which tab corresponded to positive DC in, Negative DC in and switched. I cut the metal tabs to leave enough stub to solder to and tinned the stubs then put the board back in the casing. I got hold of a 2.5mm switched chassis socket - I needed to drill the back panel 8mm above the old socket hole to clear the pcb. I shortened the lugs on the socket, fastened it in and soldered leads from the pcb stubs to the corresponding socket lugs. Finally, I cut the DC plug off the power supply and soldered on a 2.5mm DC plug. Plugged in the supply and switched on - hey presto! got the blue LED on the front panel. A final check with TV and aerial connected confirmed all was well.

Thanks for all the helpful comments.
 
Top