ฝัง
- เผยแพร่เมื่อ 22 พ.ย. 2019
- 10 Free PCBs on your first order at: www.pcbway.com (delivery extra)
GT Sonic Ultrasonic Cleaner: bit.ly/2l89aPA
Yellow Solder Flux: bit.ly/2Yhd9rg
Best BST-863 1200W Hot Air Station: bit.ly/2XRCkAQ
Best BST-328 Solder Paste: bit.ly/37yl7kY
Patreon: / sdgelectronics
Mustool MDS8207 Multimeter: bit.ly/2kKHr7A
FY6900 Generator: bit.ly/2yZI40n
Differential Probe: bit.ly/31xNTOF
Micsig on AliExpress: micsig.aliexpress.com/store/1...
T12 Soldering Station: www.banggood.com/custlink/3KK...
AC Version: www.banggood.com/custlink/3GK...
T12 Iron Tips: www.banggood.com/custlink/3DG...
Daniu Tip Cleaner: bit.ly/31VHwWM
KSGER T12-BCM3 Tip: bit.ly/2xgYXCQ
Amscope Microscope Head (US): amzn.to/2YmZWNx
Amscope Trinocular Microscope (US): amzn.to/2YdBE8k
Amscope Trinocular Microscope (UK): amzn.to/2V8w9Gh
Swift SS41-WF20 Binocular Microscope US: amzn.to/2VLWEpb
Swift SS41-WF20 Binocular Microscope UK: amzn.to/2GYGYFV
Visit my website: www.sdgelectronics.co.uk/ #ElectronicsCreators - วิทยาศาสตร์และเทคโนโลยี
![[TH] VCT Masters Shanghai Playoffs DAY 3 // FNC vs FUT | EDG vs PRX](http://i.ytimg.com/vi/pw2T5v0ogDI/mqdefault.jpg)
![แข่งแลกของ 1 วัน!! ของใครแพงสุดชนะ!! [Ver.2024]](http://i.ytimg.com/vi/joRT59ilPHg/mqdefault.jpg)
That is Impressive! A great pleasure to work with you, Steve!
It should also be mentioned, that not all SMD LEDs are tolerant of ultrasonic cleaning, like the Cree UHD1110 series. So taking a look at the datasheet beforehand is probably a good idea :-)
“2 hours seems a little excessive” - I once left an iPhone 7 PCB in the bath for 2 days by accident - still worked 😂
I'm more surprised that the ultrasonic cleaner managed to survive 2 days of continuous operation.
Nathan they don’t stay powered on for 2 days, there’s a timer on them, it’s just they get left in the fluid for a long time occasionally. 🙄
Thanks for checking this out! For some reason, the only component I've managed to consistently kill in an ultrasonic bath is the Teensy 3.2 microcontroller board. There's some component on there that dies for around 30% of the units I clean, even when running with a mild PCB-safe detergent for only 4 minutes and drying thoroughly afterwards. Of course, the boards wouldn't die immediately--they would quit working a few days after being cleaned, usually after being installed in some hard-to reach place. I never got around to swapping components around to see what was dead, but I wrote off the idea of ultrasonic cleaning Teensy's altogether and have since switched to more robust microcontrollers for my water-wash pcbs.
Huh, that's interesting. Have you reached out to Paul from PJRC about that? I wonder what which component might be failing. Have you ultrasonicly cleaned other Teensy models besides the Teensy 3.2?
@@SeanCMonahan I haven't! It's been a hot second since I encountered this issue so I've since moved on to other projects. I've had lots of success cleaning other types of boards and components of all kinds in the years since, so I do wonder what might have been causing the issue.
Pro tip: No-clean rosin-flux solder like Kester 44 is difficult to remove, even with an ultrasonic cleaner. It is meant to harden into an amber-like substance that resists moisture absorption. Instead, use a water-based flux, which comes off almost instantly in the ultrasonic cleaner. I use Chip Quik SMDSW, a 63/37 eutectic solder wire. I use ethanol in the ultrasonic bath, and dry it with the air hose. If there is any remaining ethanol lurking under a component, it will evaporate easily.
Just be aware that water-based flux MUST be cleaned off! Even more so than activated rosin, it is not safe to leave on the board.
Great video, good to see this subject in a controlled test.
Btw, in my experience it's the sticky mess which can be left when cleaning which can 'damage' switches & relays and also the problem in drying them completely inside rhe devices afterwards.....rather than the ultrasonic waves themselves causing problems.
Marvellous! Another uncommon test that clarifies the ultrasound damage debate. Well planned, well presented, well scripted. THANKS!
First hand experience: Piezo buzzers (in this case small SMD buzzers) generated voltages from the ultrasonic excitation. The micro on the PCB had some pins fried because of it. This was during a first production run, so in this case the solution was to use a different cleaning method, but the point is it can certainly cause problems.
I don't know if you'll see this ,what other cleaning methods did you end up using?
@@Sergi762 I can't remember, it's possible they just stopped cleaning the boards (no need for it with modern no-clean solder.). That was a long time ago.
Though I can tell you what I use these days on small run prototypes for customers? carbcleaner. Yup - the stuff you spray into a petrol engine carburator. It's the best I've so far found, but it is very potent, so I scrub the panel / pcb down quickly, then rince with water, then spray "CleanGreen" (or Dishwashing soap) on and wash the carb cleaner residue off (and then rince in distilled water if I am going to do a conformal coating). Incredibly quick, and perfect results - even cleans years old rosen-based residue off if I repair old PCBs. Just take note the carb cleaner will eat paint, QC pen markings, etc off, so be careful when you use it. Also highly flamable etc etc.
Very useful Video Steve,I've always wondered about the effects the ultrasonics would have, I have more confidence in using mine now👍👍
Way back in 1972 when I left tech I got a job with a watch maker, as the newbie I got the job of taking clocks apart and cleaning them.On this occasion I put the parts of an alarm clock and a grandfather clock into the ultrasonic cleaner just before I went to lunch, when I got back some of the alarm clock parts were no longer and the grandfather clock gears had holes in them, the cleaning solution used was toluene, the cavitation was what did the damage brass parts.
Ultrasonic cleaning can defiantly cause damage to components... At a previous production I was working on we upgraded to a 300W ultrasonic cleaner from a 100W version to fit more in, from this moment onwards in we found approximately 1 out of every 200 resistors (0402 [1005 Metric] AEC-Q200 1%-0.1%) to have their resistive element chipped by the cleaning process. As soon as went back to the 100W version a month later this problem stopped. In the whole process it was only the ultrasonic cleaner that was changed same batch of PCBs. The cleaning cycle was 60C for 5 minutes.
I can believe that. I just need a higher power cleaner to test that now. Was the change prompted by insufficient cleaning from the 100W model?
@@sdgelectronics The cleaning power was just fine from the 100W but the volume too small for the production rate as we could only fit 6 assemblies in at a time. The company loved to use Farnell so it’s easy enough to remember; the original cleaner was PN: 1308822 and the one we had problems with was PN: 2428947. As it was triple the liquid volume and I wasn’t expecting much of a change in the cleaning. Maybe the cleaning was slightly better with the 300W but I can't really remember as it was a few years ago.
Something to note with this is the PCBs were glued in to a shallow stainless steel enclosures, so potently the enclosure walls were reflecting more of the power on to the PCB. The cleaning was to remove thick flux from feedthroughs before further processing.
If you’re interested I’m pretty sure I have some images of the damaged resistors saved somewhere.
Good job Steve 👍 Think I've only ever had problems with switches after being in the ultrasonic, and even then after they've cooled and dried most have still worked again.
Cheers
John
I haven't had issues with the ultrasonic damaging components so far but I have had issues with flux migrating during the cleaning and turning into a thin film on contacts and that causing issues until manually cleaning.
mems ic's with a hole in it are usually barometric sensors and microphones ADMP401, BMP180, BMP280 etc.
also electrolytic capacitors could be vulnerable to this, water creeps in trough the sealed bottom contaminating the electrolyte
The one thing I am afraid for is destroying bondwires to dies. Thanks for testing, if these components survive I am a lot less afraid for IC's. I use the US only on industrial boards before repair (you do not want to know how dirty that stuff can get.) And for some calibration gear boards but often I do that by hand just to be sure
I wonder if it's the removal of the flux that accounts for the slight change in frequency. Maybe it's slightly conductive, or it changes the capacitance of the PCB.
I think temperature was the main factor for any minor drift.
Thank you for sharing this video , really informative
Another thing to consider is that the liquid could seep into your plastic packages of your ICs. It might short out pins, which in turn could carbonize the substrate and create shorts. Also if the part gets hot, the liquid might boil and blow your packages apart.
Great testing. It looks like the 8MHz oscillator at 9Hz change is about 1ppm which is nothing.
I used PCBWay once. The circuit board was slightly over 100 x 100mm. It ended up costing me €80. The exact same circuit (same Gerber files) from JLCPCB cost me €10. So no way pcb way.
Ive found PCBway to be expensive too, I also use JLC.
This is nice and I appreciate the effort you put into it. That said is an n=1 meaningful here? I mean isn't ultrasonic cleaning the sort of thing we expect to raise the odds of failure not just outright cause it? I have seen it fracture bond wires.
Great info thank you 🏴
Thanks for sharing all this!
Evidently the modern components are quite tolerant to ultrasonic. I still remain unsure about old TO-5 packaged transistors like 2N1613, 2N7011, 2N2905 or even the TO-18 types which also have bonding wires hanging supported only by their welds at the ends. Then another message here also reminded about the kovar leads going through the glass into the interior. Some lead length above the board might produce a detrimental resonance at the ultrasonic frequency?
Come to think of it, I have both several of those old metal can parts, a small ultrasonic cleaner and means of determining the frequency and maybe the resonance frequencies (in air). So maybe I should take the matter into my own hands an perform a few tests...
How did you do with import duties with PCB Way?.. I got some PCBs from them and ended up having to pay nearly their value again in customs duties.
I guess I've been just pure lucky not damaged anything yet ,not saying anymore don't want to tempt fate, well explained Steve ,I like the video.
Hi Steve, recently found your channel very impressive. Anyway, one of your videos implied that you had ordered parts from LCSC. How did you find them? Where they 'Chinesium' or of quality and genuine?
Thanks Bud
LCSC sells genuine parts.
Genuine as far as a could tell
Another great video! As an aside, testing for vibration and shock are frequently lumped together in a common specification, and for crystals, the shock specification is usually much more problematic. So I would be more worried about dropping the board on the floor (yes, I am very clumsy) than putting it through the ultrasonic cleaner! ;-)
Another note. The PCBWAY 5th year badge is very catchy, but it looks way too much like a stylized SS insignia to me. I guess this just isn't a thing in China, but I can't imagine any western company letting this logo out...
Ever since I demonstrated that my ultrasonic cleaner could put pinholes in aluminum foil, I have been worried about damaging things.
For example, I've destroyed print heads with a conventional ultrasonic cleaner. But I did read somewhere to to clean print heads, you have to use frequencies around 80-90 kHz. I don't know where you buy those.
Also the ability of the UC to get into cracks and the smallest of features makes me wonder if it could follow the legs into chips .
Hi, have watched a few of your videos and just subscribed! What is the backend of the air gun you used to blow off the board? Is it a regular compressor or an air can? If the former can you share the details of the unit ? Thanks!
It's a little silent air compressor. I don't think it's for sale any more, but it's almost identical to the Clarke Shhh Air 30/9
@@sdgelectronics Thanks! I'll check it out. cheers.
Another concern more so is damage to components at the components physical connection point (end caps of capacitors / resistors) & soldering joint integrity of improper setup / type / power level of ultrasonic uses.
what liquid do you use for pcb cleaning?
Modern IMUs that I have seen have a high max acceleration (in terms of damage, not measuring). The MPU9250 has a max rating of 10000g for 0.2 ms when unpowered (I have no idea about powered), since it is such a small and light device getting it to 10000g is probably pretty difficult in the first place and the frequency of an ultrasonic cleaner is that high that for 45 kHz the period is 0.02 ms so a tenth of that the rated time too and even within that the max acceleration will be for a much shorter period of time.
It would be interesting to measure the acceleration using an accelerometer whilst it is in the cleaner. A +-16 g one might not be enough but you can quite easily get +- 200 g ones for model rockets and similar applications. You should probably add a conformal coating first though, preferably a hard one so it doesn’t dampen the vibrations.
Great video, thank you.
What about electrolytic capacitors?
I realize this is a 2 year old video. After a run of boards came back with a high percentage of failed IMU's, I conducted an experiment on my bench. I took a board with a working IMU and placed it in distilled water inside my Branson 2200 ultrasonic cleaner for 10 minutes. After drying the board, I retried the firmware and the IMU was now returning 0's for Gyro reads on all 3 axis, and extremely erroneous Accel results on all 3 axis. This is with a TDK Invensense part. So while I'm sure there are some parts that are unaffected, people should probably refer to appnotes from the manufacturer.
Did you check the ambient temperature during the measurements?
Yes, 22.5°C the whole time, but only 0.5°C resolution on that measurement. As mentioned, I think temperature is the only reason the frequency drifted slightly.
Can you explain what the Earth's magentic field has got to do with an accelerometer?
I am a beginner - which software did you use ?
The frequency measurements of the crystals are meaningless unless you control for temperature. The delta that you see could easily be explained by temperature differences.
Yes, I fully expect those small changes to be temperature effects. The room was 22.5°C the whole time, but not controlled to anything better than 0.5°C
How would a power supply hold up? Can you ultrasonic clean those? Trying to think whether any materials are in those that aren't in a typical pcb/mobo etc.
You may have problems with the tape used in some transformers, but other than that there's nothing that would typically not survive.
Those numbers are probably within the component tolerances. You didn't exactly tell us what were the tolerances of the parts used, but I don't think you used anything extremely accurate or high tolerance but something quite common?
To be honest 2 times i fixed some HMI panels (siemens) after washing these in the ultrasonic cleaner didnt works no more,then i analyzed the boards i seen some mosfet and inductors was ko i dont know how it is been possibile but this is what happened!
What about MSL 3 components that absorb moisture from the air?
Yeah, interesting topic. Thanks for that. 👍 from me. 🇬🇧
There are a few extra mistakes in your data :
- for the 32.768MHz XTAL the 1->2 difference is 70mHz not 1Hz
- the "before" number you wrote for the 32.768MHz XTAL is probably wrong as it doesn't have the same number of digits as the rest
- for the 2.048MHz Oscillator the 1->2 difference is 860mHz not 868mHz
Thanks, the perils of late night video recording and not checking answers...
Which solder paste are you using? Yellow Solder Paste is wrong in description, sorry. lol. You miswrote, should be renamed to solder flux
Sorry for the confusion. I've updated the description
@@sdgelectronics Thank you! :D
Not sure why would you test "earth's magnetic field" with accelerometer that's not the part that gets damaged. It is usually the MEMS part and after that earth gravitational field either does not have effect on accelerometer or it does not work at all. I've damaged complete batch of prototype boards for customer and after that I never put boards with accelerometers in ultrasonic. The rest is fine i never noticed any damage on the crystals at all.
🎵Good Vibrations🎵 👍😁
Such low frequency loss is likely to be the ambient temperature rather then effects of cleaning.
Is there any truth in that the fixed freq ultra sonic cleaners might damage some components and that you get much better cleaning results with a sweeping freq ultra sonic cleaner?
Fixed-frequency cleaners don't work as well because there will be nodes where the standing wave is always at a minimum, and antinodes where it is oscillating strongly. Sweeping the frequency moves the nodes and antinodes around the surfaces of whatever you're cleaning.
I doubt the fixed-frequency cleaners would cause more damage, though if the cleaner is less effective, you may end up using a longer cleaning cycle.
With fixed freq cleaners always worth moving the pcb's around, turning them over periodically.....which helps a little.
I notice with this model, the ultrasonic frequency is modulated on top of the 50Hz AC. Although it's not technically a sweeping frequency cleaner, it does help with the issue of standing waves.
how come when I dropped my phone in the bath it didn't work even after I dried it with my hairdryer yet you can put your stuff in that bath and it works?
Late reply... but your phone had a battery connected. You can dunk an entire PC motherboard into water, alchohol, whatever, as long as (1) it's not connected to anything, i.e. no electricity flowing; and (2) you dry it properly afterwards.
When phones still had removable batteries you had a chance saving it if you were quick. Now with non-removable batteries: don't even bother, nothing will fix it after a dunking.
@@stromlo Thank you so much, Deon
is this a sweeping frequency cleaner or one for cleaning jewelry that stays at 40hz?🤔
I believe that model is a 40KHz static frequency
@@EliTheDriftPanda it has to have a sweeping frequency or it will damage what you put into it.
@@jessihawkins9116 that's my general thought too. Most electronic components survive but if the wrong resonance occurs in a crystal it could damage it, MEMS don't like it in general and higher power can damage even basic SMT components. It seems to be an "it depends" situation but sweep would help a lot, as well as physically moving items inside of the bath
@@EliTheDriftPanda look into what Louis Rossman says about these ultrasonic cleaners. He’s tried several and found that only the sweeping frequency units are suitable for electronics cleaning. There is a big price difference between the two also.
Elmasonic has P Series ultrasonic cleaners that can sweep at 80kHz which is gentler and safer for more sensitive components than the usual 37-40kHz. Curious how that would fare in comparison.
You have my attentio0n: what do you mean that 80 kHz would be safer? Is the cleaning action compromised with the different frequency?
Now i understand why i've lost a compass on my iphone 6 :D ups
Thanks for clarifying this
Yes, I originally planned to test a magnetometer, hence my balls up when talking about the magnetic field. That's probably more likely to see some damage since it is a lot more delicate.
@@sdgelectronics true, after your video i searched online about this subject and indeed i might have destroyed the chip.
Thanks mate for your videos, keep up the great work
I've got a component here keeps waving it's fist and shouting "these ultrasonic cleaners, their houses are filthy.... and they've got stupid high-pitched squeaky voices.... why don't they go back to where they came from?". I said to it, "You're so intolerant to ultrasonic cleaning!" I'll get me coat.
Anyone have experience trying to ultrasonic clean an ESP32 or similar?
@ba0sh1 That's very good to know, thank you! Can you tell me what liquids were used and if you used any type of alcohol to rinse afterwards to make sure there were no water residue left? Or is the module entirely watertight?
@ba0sh1 Thank you
I had no issues with an RN421 and SAMW25 which likely have similar components.
People wash $2000 video cards in ultrasonic after repair is done.
Because it's not harming anything. Even the cristal.
The trouble is we cannot measure failures easily. Same with ESD - can you correlate occasional PC crashes or pixel glitches? These issues rarely manifest as a total failure of the device
Is putting electronics in water ever a good idea?
only if its powered on
MEMS may be destroyed
There is no such thing as free PCB, the cost was putting your local PCB fab out of business.
Face -> Palm. Learn how to: test. Learn how to: draw schematics. Truth may hurt, but this video is bloody awful and I make no apologies for saying that.
What is your criticism of the schematics? What is your criticism of the tests? What is so awful, in your opinion?
Oh good, another shill for PCBway. Moving on...