Repairing an Apple Disk ][ drive with multiple faults
ฝัง
- เผยแพร่เมื่อ 8 ก.ค. 2024
- Welcome to Mid-Week Mini Mail Call #38! This time with have a repair of a broken disk drive sent in by a viewer.
0:00 Intro
0:37 Unboxing
2:10 On the bench
22:08 Final testing
25:44 Conclusion
Sony CX-065B Speed Controller
Apple A2M0003
-- Video Links
Apple Disk II
en.wikipedia.org/wiki/Disk_II
Apex II Diagnostics
macgui.com/downloads/?file_id...
Shugart SA-400 Manual
deramp.com/downloads/floppy_d...
Commodore 1541 Service manual
personalpages.tds.net/~rcarlse...
Frank's TH-cam Channel (IZ8DWF)
/ @iz8dwf
-- Tools
Deoxit D5:
amzn.to/2VvOKy1
store.caig.com/s.nl/it.A/id.16...
Jonard Tools EX-2 Chip Extractor:
amzn.to/2VazxDS
www.jonard.com/Products/EX-2-...
Wiha Chip Lifter:
amzn.to/3a9ftWw
www.wihatools.com/precision-c...
O-Ring Pick Set: (I use these to lift chips off boards)
amzn.to/3a9x54J
Elenco Electronics LP-560 Logic Probe:
amzn.to/2VrT5lW
Hakko FR301 Desoldering Iron:
amzn.to/2ye6xC0
Rigol DS1054Z Four Channel Oscilloscope:
www.rigolna.com/products/digi...
Head Worn Magnifying Goggles / Dual Lens Flip-In Head Magnifier:
amzn.to/3adRbuy
TL866II Plus Chip Tester and EPROM programmer: (The MiniPro)
amzn.to/2wG4tlP
www.aliexpress.com/item/33000...
TS100 Soldering Iron:
amzn.to/2K36dJ5
www.ebay.com/itm/TS100-65W-MI...
EEVBlog 121GW Multimeter:
www.eevblog.com/product/121gw/
DSLogic Basic Logic Analyzer:
amzn.to/2RDSDQw
www.ebay.com/itm/USB-Logic-DS...
Magnetic Screw Holder:
amzn.to/3b8LOhG
www.harborfreight.com/4-inch-...
Universal ZIP sockets: (clones, used on my ZIF-64 test machine)
www.ebay.com/itm/14-16-18-20-...
RetroTink 2X Upconverter: (to hook up something like a C64 to HDMI)
www.retrotink.com/
Plato (Clone) Side Cutters: (order five)
www.ebay.com/itm/1-2-5-10PCS-...
Heat Sinks:
www.aliexpress.com/item/32537...
Little squeezy bottles: (available elsewhere too)
amzn.to/3b8LOOI
--- Links
My GitHub repository:
github.com/misterblack1?tab=r...
Commodore Computer Club / Vancouver, WA - Portland, OR - PDX Commodore Users Group
www.commodorecomputerclub.com/
--- Instructional videos
My video on damage-free chip removal:
• How to remove chips wi...
--- Music
Intro music and other tracks by:
Nathan Divino
@itsnathandivino - วิทยาศาสตร์และเทคโนโลยี
Several people suggested a bad pot on the speed controller. I had checked the pot and it was fine -- it has a range of 0 ohms to 50k ohms. With this circuit, the pot only has a very small range of adjustment around the target motor speed, so removing it almost certainly would only result in the speed being slightly out, not wildly out and uncontrolled.
Did you by chance check the input vs output of the MOSFET? I think those things fail "on" sometimes, right? You should be able to test that by itself without the rest of the board powered.
Yeah I did -- even comparing how things look to the working drive (just didn't show any of that on the video for time.) I wanted to see what the output looked like on a working drive first so I knew what to look for.
The two things that I immediately suspected that i didn't see get checked in the video are in the first comment thread... Always good stuff from this community
actually no, if you removed the pot, then the on chip oscillator would not run.
I was wondering if it was a dry joint or bad connector. (?)
Fun fact: this Sony IC was used on some of their turntables (record players) that also used 300 RPM DC motors.
Ah neat fun fact!
I was about to say Americans and others easily recognize both terms, "turntables" and "record players", but... I'm old. Maybe younger people don't?
@@squirlmy Well, it's the same for me but I'm almost 40. I have younger coworkers who think that a turntable is a lazy Susan
Why so fast?
@@phonotical 300 RPM is not that fast actually, most old school belt drive turntables had synchronous AC motors that spun at 1500-1800 RPM
That "upright package" is called a SIP 😀
This is the greatest video over made. It shows troubleshooting 101, presented in a comfortable pace, using clear and sensible explanations. Using techniques and tools found in our everyday garage, including that Scope. I usually skip through the videos to get specific information, but I watched this entirely. I'm the proud owner of an Original Rainbow cabled Disk ][ on my Apple 2+ and it sill works 100% original, thanks to valuable informative videos like this. The schematics, the conversations with other Enthusiasts, allowed Adrian to complete this repair efficiently. Adrians knowledge of computers and electronics is excellent.
"Median diff alignment is -0.8" At the bottom of the screen: "Normal values are 1 for median diff."
I bought over 30 of these drives from a barn find. Nasty nasty. After clean up, and rust removal and similar repairs, 29 worked. They are work horses!
Adrian, for the bad chips why not getting a cork type board and start making a small mural with them.
I actually know someone who has covered an entire wall in their private museum with nothing but pc boards from TVs, etc.
Better yet, get a coffin-shaped box from a Halloween store, since they're dead parts. :D
Hi Adrian, I am two years late to the party but yes, I also have an ALPS half-height 5 1/4" Floppy Drive that runs flat out, not on an Apple or Commodore but on my Osborne Executive.
I had done some preliminary examination of the wiring to the backplane of the drive, and was preparing to use a breadboard to jumper the SHUGART Connector from the ribbon cable to try and find the problem, however you just saved me a heap of time and effort with this video (and your method is far simpler by going directly to the speed controller board).
Also, you provided the speed controller schematic and an obtainable replacement drive to source parts from, and I can tell you I have been seeking assistance to this problem on two Osborne Computer Groups but no one has been able to help me, so a thousand thumbs up and my eternal gratitude!
I have an Apple Disk II Floppy Drive with a dodgy 74LS125N chip (fried on the weekend by accident, when plugged in incorrectly) and when I pulled the cover off the drive to check the number on the chip (I was watching this video in the background) you mentioned the ALPS Drive and my ears pricked up. Then when you mentioned the drive running flat out and looked at the speed controller board, I realised that it is identical to the board that the Osborne Executive half-height ALPS Drive uses.
My Apple Disk II ALPS Speed Controller board is slightly different (component sizes = HUGE caps) and I want to fix that drive with a replacement 74LS125N IC, so I won't cannibalise it, however I did jump on EBAY to grab a Commodore 1541 ALPS drive that should hopefully provide a good donor board for the Osborne drive speed controller.
Thanks again and wishing you every continued success,
Cam
Canberra, Australia
Thanks a lot Adrian for pointing me into a right direction! I had a Disk Drive which was spinning at the same speed as yours! I thought, me speed controller might be dead as well. Still I probed the tachometer signal and found, there was no such thing! Nothing! No signal from the motor.
I removed the tachometer from the Compal LC-177 motor (two tiny screws) and found, that the magnetic ring on the motorshaft was spinning free! I fixated it with some superglue and activator, screw it back to the drive, turned it on and bam - drive speed was back to normal. Quick stroboadjust and the drive was booting again! Without you - me beeing a complete Apple drive hardware noob - I would have never ever found this fault! Thanks a ton again! Cheers!
Satisfaction ... One more drive rescued from junk pile, feel so good.
The Apple ][/+/e case is an still an outstanding industrial design... I miss the days when an 8-bit 48kb Apple ][+ was all I needed in life...
Agreed. Back when most micros (as they were called) came in utilitarian sheet metal cases, the Apple was designed with an elegant, off-white "futuristic typewriter" case that blended into any office environment. That was Steve Jobs' contribution, as was the use of a switching PSU, instead of the clunky linear ones that were still widespread. The latter was sourced from ASTEC rather than from WOZ. :^)
And of course you can't help but love the original "rainbow" logo instead of the insipid one they use now (ditto their current insipid product design).
The ULN2003 is a 7 Darlington transistor package with internal back emf diodes, they are usually reliable
And still available.
It warms the heart to know that another Apple ][ drive has been saved!
And that the parts from another vintage drive actually found a use.
Yet another video with perfect timing!
Just recently I got a UniDisk with the same issue (runaway speed!) and put it aside for some later troubleshooting. Like you, I made a quick effort to find schematics for the speed controller but I failed!! Many thanks to you documenting your repair in this great video which should make short work of fixing mine!!!
Thanks for the great content!!
I still love the VHS quality intro. Really sets the mood to nostalgia.
Love your channel and positive attitude Adrian - it has inspired an old Software Dev to get closer to the hardware!
Sorry that this drive was a big pain. I'm glad that you got this working. I didn't even look at it before I sent it over, as I don't have anything older than 1986 or so.
Hey Adrian, when will you do a tour in your basement and home. And a Q & A would also be Nice.
Man, I loved these disk drives.
So robust and industrial looking.
Better go pickup one from ebay!
Nobody troubleshoots in such an engaging way for a novice like me as Adrian’s digital basement!
This guy repairs old radios and related electronics but it's a similar style. th-cam.com/channels/U9SoQxJewrWb_3GxeteQPA.html
I use "Disk Speed" - Locksmith Ver. 6.0 on my Apple 5.25" drives. Includes calibration graphics.
Disk speed gives you a more precise alignment.
Wow. I forgot all about the useful utility available to pirating school kids back in the day...
Back then I was too young to realize what pirating was all about. So if a friend said "I have a copy" that meant make a copy!
Wow. Haven’t seen one of these for, oh, 35 years... brings back a lot of happy memories... amazing to think just how brilliant Woz was ...
Woz's brilliance was for the elegant efficiency and economy, in terms of total transistors/gates/logic for a given function, of everything he designed. No one did it better. But ironically, Apple charged a premium price for their computers. This was probably Steve Jobs' idea. Woz had originally set out to design a simple computer that any hobbyist could build for themselves, which ended up being the Apple 1. I don't know rhis for a fact, but I bet that on his own, Woz would have wanted the Apple II to have been more affordable in order to have been within reach for more people. Fortunately, the Apple II was so widely adopted by school districts that it got extremely wide exposure anyway.
I am 14 and I love your channel please keep up with the great work :)!
Im 16 and love this stuff, good to know Im not the only kid who likes old tech
I'm in my mid 20's now but started developing an interest in this stuff when I was 12. Good to know teens can still find this stuff interesting!
I'm just about old enough to have been around when this stuff was still being used… I'm so glad that today's younger generation (god that makes me sound old!) still have an interest :)
To me, it sounded like it just wanted to be a vacuum cleaner, but never quite made it... :P
I remember working with one of these drives in the mid 80s when in the middle of booting up one night I heard a pop from inside the drive and it stopped reading floppies. Turns out the Analog to Digital converter chip blew up. Thankfully the college I was attending at the time had replacement chips. I bought one for $8 and it fixed the drive.
Glad you could get this working. I am amazed by your diagnostic abilities.
I love it when the oscilloscope and schematics come out.
Great bit of troubleshooting thanks Adrian!
Brilliant... Wish you well and stay safe also
You are Doctors without Borders for the Electronics World!
Very friendly and with a clear and humorous explanation... I liked it
Perfectly timed video! I have a drive that is behaving the exact same way. Thanks for the troubleshooting tips of what to try.
You make it look so easy, another great repair video!
wish I could recognize board components as good as I can recognize cars. pretty smart dude
Great video. I'm procuring an Apple IIe and this brought me back to my hardware days
Adrian, I thoroughly enjoy watching, and hopefully learning from, your troubleshooting methodology. Two thumbs up!
Remember PC. For diodes, the arrow Points to the Cathode. That's how I remember.
while this is sort of wrong the arrow falls and goes to the ground ( negative )
And for my boat friends - there is no port left in the bottle
Such a cool little CRT! I ❤️ it
Thanks for ur video Pal. I repaired my old Drive for my French Apple IIe thanks to u. Last task for me is to Build up a DB9>Scart(peritel) cable
Another awesome video Adrian, it really makes my day. Thank you 😊
Awesome video, thanks from Mexico
Great video thanks Adrian.
If the mosfets or whatever switching transistor has failed, they often fail short circuit. If it's chopping 12V into some square wave pattern to regulate speed that would explain full speed operation. You could measure voltage output on the bad controller board. If it is the same as VREF then either it isn't switching because of a short, or isn't receiving a switching signal at the gate. Should be relatively easy to trace with a scope. Sometimes the gate driver is through smaller transistors to protect the IC's. One of those could be bad too. Find the PWM output pin on the Sony IC. Is it switching in relation to the RPM input? No=bad sony. Yes=find the next transistor in the chain. One of them must be bad.
Great job
Back in 1987 I once plugged the ribbon in backwards on a Disk ][. The Apple Repair Manual said replace the 74LS125, so I did and it worked!
I would suggest checking the speed control pot. Pots do fail.
That’s what I was yelling at my screen the whole time! 😆
Just found you, watched your sad c64 resurrection just now. I'm from NE Portland! I love your unconventional methods and great video style!
Nice job. Loved watching this. I've fixed scores of those drives and still have chips. Wozniak was an excellent TTL designer. Read his book sometime, it's a good read.
another great save!!!
love these videos ! I remember in 4th grade computer class using these on the Apple IIe ( word munchers ). Looks alot like my Tandy coco2 drive 3029.
Change out the stepper motor.
I used to build these at Shugart.
After 38 years I don’t remember how to fix them but I assembled hundreds of them.
When you mentioned the motor control PCB was from a Commodore drive, I thought someone in the past had hacked in a Commodore part to try to fix it. And that's exactly what you wound up doing.
Nice electronic sleuthing. Very interesting video and always great to see an Apple product get a new life.
Look at that '70s work of beauty!
Thank you, Adrian, for your excellent video showing the detailed repair of an
Apple Disk ][ floppy drive. I really enjoyed some Apple ][ content for a change. In my early twenties, 1982, my father, brother, and myself bought
Apple ][ computers for use in the family business. My major in college at the time, was computer science, and I really enjoyed programming my own computer, though it was more of a pass time. My school work had to be done on the school's mainframe. I spent a few years on the
Apple ][ and always loved it. I would REALLY enjoy seeing more videos about the
Apple ][ internals, ROMS, repairs and converted SD drives acting like hard drives. Maybe a kit on how to build one of these drives?? Really a great channel! Thank you!!
Adrian, thank you again for a great repair. I appreciate you and other TH-camrs keeping these classic systems alive. I also really appreciate that you are right here in the NW!
Good job man, u´re skillz are one with the matrix now... And as always, great video, great content, great editing. Job well done!
I have a 3.5" floppy drive in my old PC and the drive didn't work after I took it out and put it back some time ago. Last week I didn't a thorough cleaning of the machine and I figured the IDE cable was plugged in both reversed and shifted 1 row to the side, I thought it was dead but nope, I just put the cable back as I should have and still works perfectly fine. They probably built in some protection into these modern floppy drives.
Your videos keep inspiring me to do repair and maintenance jobs on old hardware. In the past week I did some repair on a 5.25" drive and did basic maintenance on it plus on another one that didn't need repair. Both work beautifully, I just checked the prices on these on ebay and they are crazy high!
Yeah IDE and floppy drives on PCs only have low current signals going through the cable, so plugging them in wrong doesn't cause an issue -- versus this disk drive where it's sending 12v and 5v right from the power supply to power the motor. When you send 12v into things that don't expect it, bad things happen.
If you want to see some impressive code look at the read sector code, the 6502 running at just 1mhz had to read the data directly from the head and process it without the help from a disc controller, the timing had to be spot on.
I remember rewriting the code back in the 80's and stripped out everything to use the drive like FRAM with no directory structure so it had greater capacity to be used as a database.
I think it's called a step-down gear. The electric motor spindle rotates several times to one rotation of the step-down gear. Some turntables have a belt drive while some have sprocket teeth gearing. I think all the names of stuff gets muddled when it doesn't describe the function. Great show. Thanks.
I've been through this exact problem on all of the floppy drives but one that I have in my collection for my TRS-80 Model 1 (I have 6) Two of them are Shugart SA-400 mechanisms and the other four are Tandon TM-100 mechanisms. All but one of them had drive speed board failures in the lone IC on the board and a couple also had bad transistors. After the first two that had failures I just replaced all three components out of course (futureproofing if you will) and they worked as they should after. So these problems are not limited to the ALPS mechanism only. They are common on both Shugart and Tandon mechanisms as well, and as shown, also not just common to Apple Disk II drives. I have yet to have issues with the controller boards on either mechanism on the TR-80, just the drive controller boards. This may be due to the fact the drives can't really be connected up wrong like it's possible to do on the Apple drive. Hope my experiences help with your troubleshooting of the defective drive motor board.
I’ve got 4 of these. Must try them out at some point. Bought with 2 Apple IIe’s for £70.
Congrats for this great repair. Nice job! I wish I were as good as you.👍👍👍
Nice and *slidy* video Adrian. Thanks :)
Hard to really appreciate just how small the Apple II disk dives were until you see one in person. On the down side, it was a complete slave to the Apple II computer. But on the upside, it was wonderfully small and elegant.
I didn't see this in a quick check of the comments, but I think the 60-50 Hz "timing wheel" is called a strobe disc or stroboscope disc.
Yep. And you can even use your 50/60Hz monitor to check/adjust the speed with it.
I don't know if anyone has said this, but the reason that the screws rusted is likely due to the two different metals in contact, essentially making the screws an anode.
Good repair one more time 👍
Old computer chronicles and such I've always heard them pronounce Shugart as shoe-gart. First time I've ever heard anyone pronounce it shug-art. Good video and I've worked on those motors long ago with that tach.
That’a for the great vid 👍 I envy you for all you ram, I have ordered 4x 2mb 30 pin simms from the USA for macs and they worked great in my gvp for my Amiga a500, have a rev 6a board with 1mb chip ram and now 8mb fast ram 😊 got there in the end 😊
Her Adrian! greetings from Argentina!!! I learned alot from your great great work and passion. Keep doing amazing things!!
BTW , crazy idea, If u have so many dead parts, chips, etc. then make a table/chair/workbench and resin epoxy.
11:02 - It's amusing to see all the 'waste space' in that original 'full-height' design!
There's partly a reason for that. Woz came up with a brilliant, minimalistic design for that analogue board mounted above the drive. It was smaller and used far fewer components than Shugart's. The same goes for that controller we see being plugged into the Apple's slot (usually #6) -- it's very sparse and tiny by the standards of the day. That said, the risk of accidentally misaligning the cable connector pins and destroying the TTL chips on the analogue board was a serious design flaw.
Incidentally, the cost of manufacturing these drives was equally minimal given the low component count, but their retail price was ridiculous. Apple and many other manufacturers made more profit from floppy drives than the actual computers they connected to!
the suspect ic in the sip package is readily available as is the drop in substitute p/n NTE1300. also used ones could be found in many cassette recorders and vcr machines.
modded my disk II’s back in the early 80’s with write protect disable switches, got tired of punching holes.
Haven’t seen one of these drives since high school geometry. My teacher had a few Apple 2e machines in a corner and wanted to run geometry software, but had the wrong versions. I spent the last half of that class writing small programs for her. I completely forgot about that.
Hi Adrian;
It might be worth checking that big green power transistor for shorts. If the switching signal is present at its base terminal but the motor is still running maxed out then most likely the transistor has a collector-to-emitter short. It appears to be a standard part that should be easy to source a replacement. Good luck :)
I got lots of those 74LS125
100k subs getting close! Look at that!
The ULN2003 is readily available and are cheap too. I'm using one in a control board I'm building to control 2 relays and a servo motor from the signals of a ATMEL micro-controller . They can take 3.3-5v TTL signals and can bring them up to 25v 500ma on each channel. You can drive them in parallel to increase the current. Never knew they were in a Apple II floppy drive from 1981.
You should try the original 74LS125 to see if it just needed reseated. If so, you can use your favorite DeOxit!
There were 2 drives. On worked an Apple drives one worked on Alps.
I still around 8 Shugie drives in a box in my girlfriends storage including a huge Winchester drive.
well done great vid
The pattern on the round sticker is called a Stroboscope
very fitting that the Commodore became the organ donor to the Apple
yay, new vid! :D
I remember those drives. We had an Apple IIe when I was 16 or 17.. There was lots of room inside. I installed a switch that I glued (with epoxy) so that you could reach it by sticking your finger just inside the door and to the left. It had 3 positions it was wired to the write-protect switch so that in one position it write protected your disk, in another position it write enabled the switch and the 3rd position did whatever the switch normally did based on the notch on the disk.
The bottom "vents" weren't actually vents either - the box was mostly water-tight. And if the can fell just right to spill into the open door could hold about half a can of Dr. Pepper. Don't ask me how I know this.
I have the working guts to one of those somewhere in my retro gear. Interesting to know the speed controller is shared with other companies drives.
Not just the speed controller - the entire drive mechanism. Shugarts were the mech inside IBM 5150's, later ones used half height MFM drives. The drive in the IBM XT is the exact same drive as the C1571 (which is a standard PC/MFM drive with a Commodore card).
There's only so many ways you can make a magnetic disk drive in a cost-effective manner anyway.
@@Stoney3K I once had a Tandy Color Computer Floppy disk controller and enclosure and put in two half height PC drives without issue. I had two more floppy drives than I had original floppy disks for the system though. :)
Fun fact: the DuoDisk and the Unidisk 5.25 both had a hole on the bottom of the case where the speed adjustment screw was 😉 not sure about the //c secondary drive, which was basically the Unidisk 5.25 without a drive 2 pass-through port...
Hi Adrian! thanks for the video!
I have an idea for the dead parts bin, it's probably overkill but it may be a good piece of wall art for the basement: you could design a PCB with the pads for the chips, add a silkscreen with your logo, chips contour and names and at the bottom of the board the title "DEAD PARTS" with the date you started to fill the bin to the date you emptied it.
And for the next board, you could add the exact episode where you replaced the chips.
That would be a nice memorial to all these chips who died doing their part (or mistakenly getting a 12V boost :D)
The motor on off is working as is the tach into pin 1.
The only parts remaining are IC , R1 , R2 , R10 ,VR1, C2 , C3 , C4, C5 & C6 .
Or the PCB has damaged print or faulty joints .
The apple 2/2+/2e drives would read a peanut butter sandwich if you could fit it in there! ^-^
Surprised i didn't hear the "clacka clacka clacka" sound of the heads bouncing as they are run to track 0. Some of the better drives (aftermarket mostly) had a sensor that would stop the tracking motor from doing that, but none of the Apple brand drives did, as far as i know.
The entire apple 2 series was a study in doing a lot with very little hardware resources.
I still have a 1541 for which I did head alignment one night. Everything went ok, it worked ok. The next day I turned it on and had the same issue. Came to the same conclusion that the Sony IC must have gone bad.
It's the heat from the transformer. Back in the day, I had to leave my drive on all the time to keep it hot otherwise it wouldn't read my disks. That chip just does disk speed, not head alighnment.
@@UberAlphaSirus It cannot be farther away from the transformer in a 1541. I also kept it disassembled for the night. So, I really have no idea why it went broke.
Well done! Never knew the drive shared some parts with the commodore. Also surprising: no deoxid was used in this episode! Doesn't happen too often! 😄
Well, they are both based on Alps mechanisms in this case. It depends on the individual drive, though. Earlier Apple drives were based on Shugart mechanisms, while later 1541s were based on Newtronics (Mitsumi) mechanisms. If you had an older Apple drive, you might be able to get some parts from a Commodore 2031 or 4040 (drives for the PET) instead, since they were based on the very same Shugart mechanism (SA390).
By the way, the Shugart mechanism is where the Apple and Commodore 5.25" floppy disk formats got their 35 tracks from--this was all the tracks it could access. The later Apple drives and the 1541 (all variants) can access 40 tracks (and usually one or two more), but the formats had been set by the earlier limitation (because of the PET drives in the case of Commodore). These formats could have had 20 kB more capacity (not 100% sure about Apple, but definitely Commodore because if its speed zones and type of GCR encoding).
Always enjoyable, even when C= parts are used to restore Apple stuff. C= fanboi inside me, cried a little. :D
Usually when the disk][ is incorrectly connected, you'll know it as there tends to be sooty marks where the 74LS125 was. Whoever connected it incorrectly will also know it as it tends to go BANG as soon as power is applied.
All my disk][ drives but one are "red label" drives with ALPS mechanisms. The remaining one (a "brown label" drive) is an actual USA-made SA390.
nice multimeter :)
A great video. It seems I am getting no Tach Signal on mine ;-)
Closing in on 100K subscribers. Almost graphite.
The markings are called a stroboscope wheel
I would have tried to solder the Sony chip again from the top side (between package and board surface) in case the legs had microscopic cracks in it. Although I don't know if that package type is prone to that.