Chip Harness

Adder_Noir

Hi,

My shiny new op-amp arrived today for my hydrophone project. Great. Only problem is it's difficult to actually see it, it's so small. The pins are barely a millimeter apart and obviously I can't solder anything that small. So is it possible to get harnesses which increase the size of the legs and the spread distance between them, which the chip can plug into?

Thanks for any help offered

Adder_Noir

Found out it's called a DIL socket. I've ordered a 0.3in spaced one whatever that means. Hopefully it will be ok. Please could we not lock this topic just yet incase I need some more help on it thanks.

waht

There is not much you can do, these chips were designed to be handled by specialized machines that solder chips like these 50 a minute or more.

Under homebrew conditions, you would have to polish up a very fine and sharp metal tip you could attach to a soldering iron. The best way to hold it in place is using a small straw under a vacuum pump. It's going to be a very tedious, and you would also need a giant magnifying glass.

Can't you use a simple LM741 opamp from radioshack?

berkeman

You may be able to use a "surf board" (not sure where that term comes from) to adapt from your fine pitch SMT part out to 0.1" through-hole pins. The 8100-SMT4 adapter board is something that I use for adapting medium-pitch SMT parts out to 0.1" TH:

http://twinind.com/catalog.php?id=8

Other vendors make other adapter boardlets as well. waht is correct that you will need to upgrade at least the tip of your iron (buy a sharp tip, though, filing or polishing an iron tip generally will not work). You can consider getting a better soldering iron setup as well, if you can spend the money on it. The Metcal soldering irons are a standard around here in our labs. They use RF heating of the tip to improve temperature control and speed up your soldering.

berkeman

Here's another adapter boardlet supplier that a co-worker just sent me in response to my asking him about it...

http://www.schmartboard.com/

Check out the 10 y/o kid soldering fine pitch stuff...... Yikes!

waht

Wow, that's impressive, the scmart boards have special groves such that the chip will fit in like a jigsaw puzzle piece, and the groves are high such that and solder can fill them up, and still maintain an insulation between the adjacent pin, this is all under 0.4 mm. When i was a kid, I used to solder those huge germanium transistors my dad had, but I guess every generation gets more advanced at every age,sweet.

Adder_Noir

Oh my word this place really is a hive of useful information! Makes me feel a bit better after I broke my mobile phone today by leaving it on top of the car roof. I guess I must have gone 3-4 miles before hearing a loud knock on the roof and seeing it fragment in the rear view mirror as it was introduced to the tarmac.

Thanks again mate I'll cruise on over there now. Oh by the way got myself a nice new more intermediate book too. 'Practical Electronics For Inventors' a much easier read for someone like me and yet still thorough at 900+ pages.

Thanks again

Averagesupernova

I was soldering in .05 inch spacing smt devices 15 years ago doing rework with nothing more than a soldering iron and a heat gun. Make sure you have some solder built up on the circuit board pads. Use an appropriate sized iron to do this. Only so much solder will stick to the pads, the rest will stick to the iron. Lay the smt device one top of the pads lined up about how it needs to be. Heat the whole device and pads with a heat gun with a nozzle that is no bigger than necessary from directly above. It's important that the heat gun gets hot enough fast enough so it doesn't do a 'slow burn'. Once the solder flows, tap the board close to the device and the surface tension of the solder in combination with the slight mechanical shock from your tapping will pull the device exactly where it needs to be. Remove the heat gun and let it set. I've reworked hundreds if not thousands of boards this way. It takes a bit of common sense and a little experience. Also, I removed the old part from the board in a similar manner with an appropriate sized pair of tweezers. SOICs, PLCCs, didn't matter which, it worked fine.
-
Edit: After rereading the thread I realize you want to adapt an SOIC to DIP. Oh well, maybe my info will help someone some day.

Adder_Noir

Very nice advice, I'm sure it will assist someone, someday (i.e. me!)

chroot

Is it possible you can just buy the same part in a more user-friendly package?

- Warren

Adder_Noir

I ordered the LM741 instead

berkeman

Keep in mind that the 741 has significant limitations in terms of input common mode range and output swing (see H&H, for example). This is not an issue if you are running it off of +/-12V rails and dealing just with +/-5V signals, but if you are thinking of trying to run it off of a single 5V rail or something, you will want to use a more modern CMOS opamp that can drive its output closer to the rails, and tolerate input voltages that come closer to the rails.

Double-click Rail-to-Rail Opamps at this National Semiconductor flash web page to see some examples (or click on the other opamp categories to see the selection guides):

http://www.national.com/cat/index.cgi?i=11

Adder_Noir

Ah I see, so the 741 doesn't like signal inputs which are close to ground but can handle negative and positive signals quite well. Hmm... a CMOS op-amp could be the order of the day I'll have to do a bit of reading first.

For what it's worth and for the hell of it:

Could I have used a half dc biased emitter follower and beefed up it's output signal with the 741 which would have prevented any near ground issues, or would that just be inviting the same problems, or even is this present paragraph total garbage, please do feel free to point out to me when I'm talking nonsense.

berkeman

No, when the older LM741 opamp is run with split rails, the problems are when the input and output signals get near the rails. When you run the 741 with a single supply (like 12V and GND), you have the same problem, but it just happens that the negative rail is GND. The 741 does not have enough headroom to be run between 5V and GND for most practical purposes. If you want to process a 2-3V signal with an opamp running only off of 5V and GND, then you definetly need a newer CMOS opamp designed for that purpose.

Adder_Noir

Thanks again for the clarification

Adder_Noir

Sorry to bother you folks again and I'm only asking for my own educational reasons I'm not questioning you, but:

Is the issue berkeman stated above in any way related to circuit loading of the signal voltage which is highly undesirable? Is it something to do with the op-amps lower than normal input impedance?

Also I thought I knew but I don't having re-read through it, what does op-amp 'head-room' mean? I'm coming to all this stuff in my private studies but I like discussing it and any advance preparation would really help

berkeman

You shouldn't get any unusually low Zin unless you are above or below the power supply rails of the opamp, in general. At that point, the input protection diodes will start to conduct.

"Headroom" is the voltage margin between the signal itself and the power supply rails. So say you have a signal that varies between 1V and 4V, and you are using a single-supply OpAmp running off of 5V. You would then have 1V of headroom to the upper and lower supply rails. The less headroom you have, the harder it is to design an amplifier circuit with low distortion.