Linksys EA8300 serial connection: soldering pins?

Hello guys,

I've been looking for a way to have a serial connection for my Linksys EA8300 routers, as I am planning to build custom firmwares for it (as for my Linksys WRT3200ACM, which luckily has pins for a serial connection already exposed), so in case things go south, I'd like to have a fallback/troubleshoot plan.
Opening up the EA8300 was easy enough, however there seems to be some "material" in the pin holes, which prevents me from sticking pins through them and soldering them from the back.
Please see the following image (taken from https://openwrt.org/_detail/media/linksys/ea8300_pcb_top.jpg?id=toh%3Alinksys%3Alinksys_ea8300 and zoomed in)

I don't think it's soldering tin, as I have been unable to desolder/free up the pin holes, but I might be wrong, as I am not an expert in soldering.

Finally, to my question:
How have you guys realized serial connections to the EA8300? Have you just soldered some wires onto the pin holes or did you drill through the holes and afterwards soldered some pins on it - or maybe just soldered the pins onto the holes?

.. or am I just too stupid to remove the soldering tin from the holes?

Every help is highly appreciated - thanks a lot beforehand!

2 Likes

It looks like unleaded wave solder bubbles in that picture. The outer 2 pads look rather dark; oxidized or conformal coating?
I use leaded solder and reheat the 3 joints (not 3.3V pad) and insert berg pins there (red rectangle above) to get a serial connection.

1 Like

it may be soldering tin, but silver solder, which is higher temperature than leaded solder.

the DIR-882/878/2640/2660/1760/1960 come with that nightmare stuff in the holes, and yes removing it has been a chore.

if it is that stuff, then you will need to increase the heat on the iron to remove it.

1 Like

Well, I tried with the max my iron can offer (450 degrees, ERA i-CON nano), although I only tried it for a couple of seconds, as I was afraid, I would damage the board.
Can anybody with a EA8300 and working serial connection maybe share his/her experience on how to solder the pins?

@jeff If I recall correctly, you introduced the support for EA8300 back then, so I would assume you have/had a serial connection.
If so: Could you please share your experience? Thanks a lot!

1 Like

if you're afraid, don't do it. but silicon is quite resilient.

what i will say is that removing that stuff was among the more difficult tasks and you would need to hold it there for a few seconds before it would melt.

it was stubborn. but again don't do anything you're not comfortable with and i encourage you to solicit input of others, as you are doing now.

1 Like

@broly <--This guy solders.

This would be much more difficult - but there are alternatives.

Based on the picture provided it looks like the holes are only half full (half empty)?

This means you could probably just jam header pins straight in (halfway) and then hot glue it in place temporarily for "test purposes". That would be safer and easier than most commonly available clip on test leads and reversible.

Search:

20pcs 1x40 Row 2.54 Breakable 40 Pins Connector Strip Board Module Electronic Part Elektronik Single Male Pin Header For Arduino

1 Like

thanks for the compliment.

never thought of using hot glue to test the signal on those annoying-ass things. that's a good idea.

whatever, i only learned how to solder because of how incredibly difficult they were to deal with.

and imagine using silver solder as your first solder, in addition to having to remove it.

story of my life dawg

2 Likes

Tip: Use a bunch of fresh solder to help flux and melt old solder.

1 Like

Wow. I also really struggled to get the solder bubbles out of the holes. I have so far not succeeded.
Silver Solder has a melting point at 1377F or 743C. Not sure my iron goes that high, but I'm checking.
It would be nice to know what temperature others had to use to deal.

It is only regular solder. The problem is that the ground hole is linked to an internal copper plane in the board which will dissipate heat rapidly from your soldering iron. I use a heat gun to preheat the whole area of the board. You can also use a small drill bit to mechanically remove the solder, but it will still take lots of heat to fully melt the new solder and properly install the pin. Which leads to another workaround of not soldering the pins at all, but just putting them through clear holes (drilled out if necessary) then tilting the pin assembly to the side so they make contact for the one time you need serial.

Most of the time I've been lucky where I have just pushed it out with a pin.

I've been soldering for 40+ years and specialise in rework, so maybe I can chime in about the way I'd approach this.

First, the material in the holes will be lead-free solder, applied during the manufacturing process, as others have said.

Second, lead-free has a melting points higher than leaded solder, which nowadays is normally used only for re-work/repair purposes (there are a few exceptions).

Users often struggle with this kind of situation, but there is an easy option I use all the time:-

Method one - pre-soaking with a hairdryer
preheat the board to raise it above room temperature with a hair dryer. You will find it much easier if you can soak the board in warm air from the hairdryer for a few minutes to get it to around 60-70 degress C. Your normal iron, set to a normal temperature, will have an easier time melting the lead-free solder. This is important because without preheating, you'd be forced to turn the iron heat up so high that you'd damage the board very rapidly. The actual board temperature you need to achive is not critical, but dont go too hot, make sure it is not too hot to handle with your fingertips for example. Just make sure you warm it for a few minutes to enable all layers of copper inside the PCB to rise nicely above room temperature. This can take some time especially around areas that have a lot of ground or power planes (edit: wide expanses of sheet copper, as opposed to narrow traces)

Note: in theory, using such a hairdryer is questionable from a ESD perspective (fast moving dry air creates static discharges), so you can reduce the risk by using a hairdryer on low airflow setting, and not allowing the hairdryer to touch the board or plastic case of the router. If you are very worried, or working on expensive equipment, use proper ESD bonding techniques

ESD edit: the hair dryer is likely to be double-insulated and not easy to ESD bond. Also, on static discharges, proper ESD reduction techniques advocate slow discharge through a high-value resistor not just for human safety, it's to reduce the charge flow to non-damaging levels. So "earthing yourself by touching something metal" isn't really the best thing to do, especially if you have already built up a few thousand volts in your body (not unusual in a domestic environment)
But hey - for a 5euro commodity CPE/ISP router I don't bother. But for a 300 Euro special or commercial work, I bond everything properly.
Always use corded wrist straps, not scam cordless ones.

Method two:
Low-melting point 58% Bismuth solder for initial cleanout.
The other technique I use, when the hairdyer heat soak fails, is to try to remove some of the solder by using bisthmuth paste low-melting point solder to repeatedly wick out the lead-free solder. It is avilable in small pots for BGA rework jobs. You will never reach a true eutectic mix of the two solders, but an ad-hoc mix of Bi-58 with high-melting-point lead-free usually allows some more of the lead-free to be removed with each attempt, and so gradually the hole can be cleared. If you can't get 58% bisthmuth (Bi) solder paste, then you could try to persist wih 63/37% lead/tin solder, which has a very slightly lower meting point (only a few degrees) lower than the more usual 60/40 lead/tin solder but crucially has a wider eutic range so is very slightly easier to work with.
I don't use Wood's Metal,

NOTE: Bismuth solders are NOT usually suitable for permanent use as they can develop some mechanical problems over time where there is any movement or there is no conformal coating or solder mask/laquer. Clean up well with liquid flux (edit: with desolder wick) and use lead/tin solder for the actual soldering later.

HOWEVER: for one-off attachment of TTL-serial you can often temporarily attach the very thin flexible wires (GND, TX and RX to the board with a blob hot glue nearby then temporarily solder then to the respective pads using Bisthuth solder paste alone. edit: use very thin flexible wires around 32SWG/30AWG or finer, also prebend the ends by nipping with blunt/polished tweezers so they lie correctly with no tension - or warm gently with an nearby warm soldering iron to soften the insulation more into shape) This is temporary and not suitable for lontgerm situations but for example when doing the inital flash of a router to change from OEM to OpenWRT (BT Homehub 5A comes to mind) it is worth a try and is far faster than fitting pins. you are relying on the solder mask to provide some insulation from the ground plane so talke it easy with the iron. Strip a mm or two of the wire ends - that's all. Tin the pads first with a tiny amount of Bismuth paste, then tin the wires too. Then hold the stripped, tinned with end down on the pad
with a flat screwdriver tip and apply the iron for a slit second. Hold the screwdriver very still for the next 10 seconds, then see if you have a good joint.

For expensive or troublesome rework like this I both preheat AND use Bismuth paste.

related tips: use a LARGE iron tip (NOT small or too fine). A large tip has high thermal mass so it does not cool rapidly when it hits the board. This avoids the need to crank up the iron temperature way too high, a common mistake.
It must also be clean and tinned (I use salmiak for tip cleaning, also a genione brass swarf "pad scrub" ball). I mainly use rosin flux in alcohol to make a liquid flux, but clean flux residue carefully with Isopropyl alcohol (or rubbing alcohol) to avoid future problems.

Always have good ventilation, but avoid draughts on your work as rapid cooling leads to bad joints.
Rosin/colophony and salmiac have vapour hazards, can trigger alergies and cause occupational asthma. Lead has serious longterm health effects etc etc

1 Like

Thank you very much for the great insights!

thanks - it was just a late-night brainspew and forgot I'd posted it - I've cleaned up a few typos and edited some extra notes in.

I have a few EA and MR 8300's myself, though all in remote use at present. I like them.

This topic was automatically closed 10 days after the last reply. New replies are no longer allowed.