Does anybody know of any companies that do soldering? I would like to get a board with some components soldered.
You have to be a bit more specific.
* turn around time
Your request is far too vague. For instance, are you looking for some hand soldered prototypes, or a production run? How many components? SMT or through hole?
Sorry for all the questions, but I hope you get my drift.
Sorry about the vagueness. Here is what I need:
Complexity: About 20-25 components,One 2 layer board
TAT: about 1 week (max) faster would be better
These are through hole components. I am looking for hand soldered prototypes.
So it sounds like you are looking for casual labor in your vicinity, and supplying the PCB and componentry yourself.
Is that a fair assessment?
(I can't help directly, but hope these questions help you in your quest -good luck)
Most people would get out a soldering iron and assemble that sort of thing themselves. If you aren't happy doing that you could try your local amateur radio club. I'm sure you would find someone there who would do it for you; the club I belong to has several members who would help with that sort of thing. If you can't find anyone locally, post it to me and I'll do it for you, for a small consideration.
Leon Heller G1HSM
The reason I asked was I suck at soldering. But I will give it a try.
Where are you located? There might be Freaks near you that want some extra money (or in this economy any money they can get).
edit: I guess it really doesn't matter where you are. However, there still might be people here who could do the work at home and make some money if you are up for that.
Soldering isn't difficult if you have the right equipment. It needn't be very expensive.
You could have soldered those PCBs in the time you were waiting for replies here.
He mentioned that his soldering technique left something to be desired, so they might not work.
A perfect opportunity to brush up debugging skills then.
I'm sympathetic to being cautious about trashing one's possibly first PCB 8-) If the OP has all the parts and will cover shipping both ways, I'll do them gratis as a shameless promotion of my manufacturing operation. I will even throw in some "junk" bare boards for him to practice/learn soldering on.
That is very generous of you Tom. Yes, you guys are right, this is my first full fledged board, but I will give it a try. Hopefully I don't screw up bad. :)
Tom, I will keep your offer in mind. Thanks.
I am sorry that I have to tell you that any embedded engineer must be able to solder at least a soic 8 package, or a 0805 resistor. If he can't, then how he can work on a prototype board?
I think that soldering of electronic components is easier that you imagine. Also peronaly, I like to work more with SMD than through hole packages, because there are years now that I learned how to use them and I realised that I can solder them faster.
Try it and I believe you will get it very quickly.
IAR Embedded Workbench C/C++ Compiler
I agree Michael. I will get used to breathing the lead fumes now :)
Lead fumes? You're kidding, right?
Lead turns to a gas at 1749 Â°C or 3180 Â°F, according to http://en.wikipedia.org/wiki/Lead. Are you sure your iron is that hot?
Perhaps you meant flux fumes? :)
From a Vishay MSDS:
Hazardous Decomposition or By-products: When heated to soldering temperatures, the solvent in the flux will boil away and carry up droplets of rosin and thermal degradation products such as aliphatic aldehydes, acids and terpenes. No lead or antimony are detected in fumes from soldering below 1000Â°F (537Â°C). Melted solder may liberate carbon monoxide, carbon dioxide, lead oxide fumes.
Yes I was just kidding. :)
But that is some interesting info. Didn't know CO and CO2 were also involved in the mix. Wouldn't lead oxide be just as harmful?
It depends on your age and eyesight! I
have been doing it for too many years and without a stereo microscope I can not solder SO8 packages very well. I have some parts that have 0.025" pitch and it is real easy to bridge those. My office builds my proto boards where they do have a stereo 40x microscope. For production we use an ISO cert. house. They have board washers that cost a lot of money. Analog circuits do not like board leakage.
I agree that thru hole parts should not be that hard to do.
I periodically interview EEs for our company. I always ask them about their soldering skills. Major marks against them if they can't at least solder a fine pitched TQFP or an 0603. I view those as an absolute bare minimum.
I can solder SMD components with leads easily.
I normally puts some solder on the pads, then put flux over them.
Place my AVR TQFP on its pad and align it. Once that
is done, touch few pins, so that avr is held in position.
Then i just touch every pin for it to get attached to the solder and pad.
But now , i am required to solder QFN package, such as the RF230 and ATmega 1284P.
Does anyone have any tips for this?
I got some QFN Li-ion charge IC, which i will experment on, as i do not want to spoil any RF230.
So can someone recommed me how to solder QFN, as you can not see the leads underneath, so aligning can be hard. And how do i solder the GND PAD below.
These QFN are so small
I've drag-soldered QFN chips successfully. It helps if the pads are extended. Positioning it correctly can be difficult, I use a microscope which helps when tacking down the corners.
I use a large via for the pad, and feed solder in while heating it with the tip.
Is drag-soldering when you place some solder on the tip of the iron, and then just move it across all the pads.
Would you recommend using masking tap to hold the IC down.
I got a table magnifier glass, would that help,
How big would you say a large via is?HOw would i feed it. I was thinking to place solder on the pad, with lot of fluxs, then once the via heats up. The solder will start melting. Well i could test how long it takes, before placing a QFN IC.
How much should the pads be extended it, should it by 1mm or so. I am using footprint avaialble in Altitum.
I was told that by extending footprint,can effect how solder attracts the component pads.
Yes, that is drag-soldering. I use a special mini-hoof cartridge with my Metcal system. Plenty of gel flux is important.
I hold the chip down with something pointed while tacking two opposite corners.
I use a 2 mm via for the pad.
Extending pads 1 mm should be plenty.
Extending pads like that only affects reflow soldering.
I only do that for prototypes. Production boards should be designed in the usual way and reflow-soldered.
Well i am hoping once i have tested the board it will be for production, so would you say extending would not be a good idea then.
This is the iron i am using, you can hover over the pic and see the tip it comes with.
Is it fine enough,
By the way, whats the chance of two pins shorting or solder join a pin the GND pad,because it hard to see what happening.
I also have this, with a hot air fitting.
Does that help
That's actually more likely with reflow soldering, if the stencil isn't designed properly. It's a good idea to have QFN chips X-rayed, just like BGAs, until the process is proven.
So if i extend my pads , will it be ok for production?
The Rf230 is QFN32, according to the datasheet each pad length is 0.40mm.
In my PCB software under Atmel PCB libaries there are 3 QFN pads, They have the following pad lengths 0.7,0.8, 0.9.
So for a 0.4 pad , should i stick to 0.9 or should i make my version with 1.4mm+ pad.
0.9 mm should be OK. Print it out, put the chip on it and check that you have enough space for the soldering.
Use the IPC guidelines for production boards. PCB Matrix has a footprint viewer with footprints that follow the IPC guidelines.
Printing might be a problem, because with Alitium, I can not get propotional print out.
But in regards to production, if the stencel is made for my board, then why would there be a problem?
Use the plot function. I can plot to a Windows printer with the software I use.
The stencil has to be designed properly, it's quite critical for the central pad (a grid has to be used) to prevent the chip floating and excess solder causing shorts.
I will give that a go tommorow.
I have a Raven board that has the RF230.
The RF chip size is very small and my solder iron tip is not small enough to touch each pin. Mater of fact the tip is twice the RF 230.
So when i place the chip, its ok to place my tip on more then on pad for it to solder? Is this right.
Now the big question, how do you guarentee that the IC pad is over the PCB pad?
Ensure sure that it is centred on the outer pads. You need to use a very fine tip for soldering the two opposite corners, a larger tip can be used for drag soldering.
Thanks, i am in the process of finalizing this PCB and would like to solder things myself.
When you say center the outer pad, do you mean the pad's on the edges of the IC.
Am i correct in saying the tip should be like needle.
Well what i am planning to do is place my tip on few pins so solder folows to all of them.
Is this the correct way of doing it?
When soldering QTFP, i place flux on the pads, then place some solder and then place the component and touch few pins at a time.
But with QFN being smaller and pads below it, i am just worried that two pins do not connect at the base.
The tip used for drag soldering should not be like a needle. You actually want to use a fairly large tip for good heat transfer. When I was learning to drag solder, I spent a while watching these videos to get a feel for it:
Search on Google Video for "drag soldering" and you'll find lots more.
With QFN, I tack a couple of opposite corners while holding the part with tweezers, double-check the alignment, and then begin drag soldering the edges which have not been tacked. If you create a bridge, just add a bit more flux and drag over it once more.
I am able to center 0.5 mm QFN parts using a method somewhat akin to a "vernier" scale. Basically, don't try to line up the individual pads. Instead, treat the edge like an interference pattern, and bump the chip around until it is as "shiny" as it can get. As long as you're not out by a pad or more, this should do it.
For what it's worth, I use a Weller WES51 soldering station with the ETH tip to drag solder just about anything. It's a mid-size "screwdriver" tip--with a couple of flat edges--which helps keep a "bead" of solder on the tip.
I use a mini-hoof cartidge on my Metcal system - it's a bit like a horse's hoof and holds a small amount of solder. It's great for fine-pitch QFP, but a conventional cartridge works better for QFN parts.
I use a 2 mm via for the pad.
I have just updated my PAD, now in regards to vias i have 9 vias each are 0.6mm, with 0.33 drill hole.
So what should be 2mm, becuase of the hole via is 2mm i can only fit one.
A single 2mm via makes it easy to solder the central pad.
Anyone have QFN reflow soldering tips? I seem to be running pretty bad odds with about 2 out of the 5 prototype boards working correctly. (So tips on re working would be appreciated too).
My current method is putting a thin line of solder paste along the pads, one blob in the middle, then putting the board in the reflow oven.
That can work with QFPs but you will get shorts with QFNs, as I expect you have found. You need to use a stencil. The blob on the central pad will also cause problesm, as the chip will tend to float on it when the paste melts. QFN stencils have a grid pattern over that pad to prevent floating.
Does this problem also apply to soldering with a iron.
Also it sounds like if there is a big blob in the middle it can lift the outside QFN leads off the pads. (http://screamingcircuits.typepad.com/.shared/image.html?/photos/uncategorized/qfn_float.jpg shows what might be happening with my soldering)
One technique we tried was having the QFN float on a huge blob of solder and then pushing it down with some tweezers, ejecting the excess solder out the sides and leaving about the right amount to float into position. This works occasionally.
Tomorrow I am going to try tinning the QFN, then adding some flux and sit it on the board, and then using the reflow oven. (http://www.intersil.com/data/tb/TB389.pdf has a slightly similar method for re working)
That problem doesn't arise using a soldering iron.
That problem doesn't arise using a soldering iron.
i have kept a 2mm pad with 1mm drill hole size.
My understanding is that this is an awesome recipe for shorts from the center pad to the pins. As a bonus, without x-ray you might have a heck of a time diagnosing the short.
Anyway, with the caveat that this may be a bad idea for production work, I've yet to meet a QFN chip that actually required that the bottom pad be soldered at all. The SparkFun tutorials seem to suggest that this holds pretty generally (search for "adding vias" here). As it's already been suggested that if you're doing production work you really need to be properly reflowing QFN parts, perhaps "don't worry about it" would be good advice for hobbyists and prototypes? I'm sure somebody here will correct me if this is a Really Bad Idea.
iF the pad isn't soldered the chip might not be properly grounded. Also, if the board is dropped or mishandled, the chip is likely to come off, together with the pads.
In regards to hand soldering
Each pin gets glazed with solder.
Then i place flux
Place the componet and, just touchs pins so solder flows.
But for the GND pad, what should i do? Do I place solder before or after(through the via). Becuase i have seen many images of blobs of solder, this will raise the IC as its not flat which could cuase the other pins from not connecting.
© 2020 Microchip Technology Inc.