Replacing SMD with through-hole

[RR Bigman]

Self explanatory me thinks. Replacing an smd diode with its through hole equivalent in my delta labs delay. My google fu must be no good. All I can find on the subject is either doing the reverse or adapters, which I can't afford. Anyone have any tricks? Or maybe some links? Desoldering the part isn't an issue. Thanks in advance

[skullservant]

Yep you should be good to go as long as you are careful to consider clearance of the new diode and get the old SMD guy off

[Gone Fission]

Getting the old guy off is usually the easy part. The trick is getting the new part in an alignment that will fit and where you can get an iron in to heat the pad and component lead to get it soldered. (Don't forget the logistics of getting the solder in there.) If the pcb is really packed and the space is next to taller components, this could be dicey. The tiny little solder drops will be all that holds the diode in place, so be careful of precarious things like if a battery is rattling around in the case unsecured and the diode is flying above the board. Still, generally doable if you have the fine control to handle SMD-scale soldering, but the SMD components definitedly give an easier time of going where they're supposed to.

I have no problem with SMD in my pedals but I don't like working on it. It took me years to get a steady hand on thru-hole and I've got other things I'd rather spend my time on than fussing at that small a scale. Power to people who do, though.

[Jero]

Getting the old guy off is usually the easy part. The trick is getting the new part in an alignment that will fit and where you can get an iron in to heat the pad and component lead to get it soldered. (Don't forget the logistics of getting the solder in there.) If the pcb is really packed and the space is next to taller components, this could be dicey. The tiny little solder drops will be all that holds the diode in place, so be careful of precarious things like if a battery is rattling around in the case unsecured and the diode is flying above the board. Still, generally doable if you have the fine control to handle SMD-scale soldering, but the SMD components definitedly give an easier time of going where they're supposed to.

[mysteriousj]

Done this heaps of times, bend and cut one leg to place first then solder it. Next cut the second leg to length bend it in place and quickly solder it. It helps to have tweezers and fine solder too. You might have to have the diode sticking up a bit and the legs doing a full 180degree turn above the pcb if it's a small package.

You also could replace it with the same SMD part? Not sure how it is in the US but some electrical suppliers here have free shipping and no minimum order so you can get the exact diode (or ten of them) delivered for about $1.(dunno how it works out economical for them )

[RR Bigman]

thanks for the advice and kernels of wisdom gents

pulled out the smd part....still planning out the transplant. I also just realized that this is in fact a silicon rectifier diode (1n4001 as far as I can tell), which mucks things up a bit as rectifiers have thick as fuck leads and are considerably larger than my 1n4148s. might have to finagle some off board vero in there or something. This will require more nicotine and a few gallons of tang.

[TweedBassman]

oh man soldering a 4001 onto SMD pads is gonna be tough.

Is it a power input diode, shunted to ground? if so you may be able to get away with a 4148 type instead.

if you wanna go for the 4001, hold the component by the middle (on the numbers) with your needle nose pliers. if you hold the lead, the iron's heat will be sapped out to the pliers and you won't melt the solder or it will take too long.

[RR Bigman]

I got it on there, but it appears that it's not the only issue. The only other thing I can reasonably replace is the pt2399, which is through-hole (do they even make SOIC 2399s?). This is probably my own fault for discontinuing my debugging after finding the faulty diode. Though I'm fairly certain that this is not issue. I'm sort of jonesing for an analog delay right now, maybe I'll throw a generic delay on the breadboard or something. WE HAVE THE TECHNOLOGY...

[eatyourguitar]

why not replace the SMD part with the exact same SMD part? no hot air? get hot air. no paste? buy solder paste. yes they make 2399 in SMD. they make everything in SMD. even analog delay chips are easier to get in SMD now.

on the subject of troubleshooting, use a tone generator and a probe. find out where the signal disappears and %99 of the time you will already know why the pedal does not work. those rectifier diodes are usually idle waiting for you to plug in the wrong polarity power supply. even then, it would damage your power supply before it damages the diode. 9v will NOT burn a 14001. they are built to take a beating. more important though, are you sure that it was a 1N4001? I think it was a 5v zener that supplies 5v power to the pt2399. what you did now is you either have the 4001 in there blocking all the power from getting to the 2399 or you have it forward biased and you fried your working 2399 with 9v.

[RR Bigman]

The board didn't have power anywhere save for the power jack pins, and the diode was coming directly from the power jack so I assumed (I know, it's bad to assume) it was the culprit. Plus my cheapo meter was saying it was an open circuit when I checked the the diode drop on it. I also deduced, whether correctly or not, that since the diode was connected to the positive pin of the power jack that it was at l least rated above 5volts (another dangerous assumption) figured it was polarity protection or something. Checking the markings (m1) from the diode in question on teh googles brought up rectifier diodes. Though I'm notoriously awful at looking up part numbers on the internet, so there again could be some bad mojo. The 2399 is through hole btw. Also, rereading my response, I come off a little condescending....so sorry about that. I'm not I promise, just running through my reasoning. I'm not sure it would even be worth the expense getting smd equipment to fix a 40 dollar pedal though. I have all the parts to build an echo base and/or a clari(not) which have both been on my build list for at least a year at this point. Thanks for the info though

[eatyourguitar]

ahh yes you have a rectifier diode in series with power jack V+... and you are %100 correct in your assumptions. I was assuming that it was a backwards rectifier diode sitting between the power rails but I was wrong. what do you mean your meter showed the diode drop as open circuit? diode drop is measured in volts. you can kinda guess that you have an open circuit if you know how to interpret the voltage drop. the meter would need to be set to measure resistance across the diode with the power off to really test the diode for being failed and open.

if the diode is open and bad you have two roads you can go down in the "what if".
1) the diode is defective and failed over time
2) the diode failed as a result of something sucking wayyyyy too much current.

I would also be careful not to fall into traps of bad PCB traces and bad solder joints. if you have the failed component removed and it is still intact, test it all by itself with a 470R in series to a 9v test supply. you can measure the voltage drop across your resistor, diode, resistor + diode. the rectifier should read 0.7v. if you get 0v it is failed closed, if you get 9v it is failed open. if you get 5v, it is a zener. also, reverse the direction of the diode (or battery) and do the test again.

back to the pedal, if your sure that what you removed is a rectifier diode in series to the power V+ and the circuit V+, you can just jumper it with a wire to get the pedal working again. thats what I would do.