EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: lithic on November 19, 2016, 11:49:39 pm
-
Like alot of people i have been watching Louis Rossmann's videos and well now i want an ultra-sonic cleaner.
But not one from ebay, and not one that costs $2000.
Can anyone recommend some fluid, since i can't find Branson EC in Australia and a cleaner
I know Super Cheap Auto has one.. but yeah its plastic and i don't want to put $500 parts in a $80 cleaner that does not sweep frequency's
Has anyone had experience with Liquid Glass OZ
http://www.liquidglassoz.com.au/category1_1.htm (http://www.liquidglassoz.com.au/category1_1.htm)
The 120w 2.9L seems to be roughly what i want and at $500 its roughly the price im considering
-
I have a 20 litre Chinese ultrasonic cleaner and the instructions say to only use water or water and a little dishwashing soap. It works very, very well with plain water. I've used mine to clean a generator carburetor and one of my handguns with great results.
I think mine was around $250 US from eBay.
-
I forgot to mention what i was planning to buy an ultrasonic cleaner for, electrical components not mechanical parts.
After a few hours of trying to find a page that isn't all hype and has some numbers attached to it for practical uses i found this (http://www.tmasc.com/qa%20process.htm).
When Louis Rossmann had a video on his ultrasonic cleaners Crest CP500D (http://store.crest-ultrasonics.com/product-p/tt500d.htm) vs Branson CPX5800 (https://www.fishersci.com/shop/products/branson-bransonic-cpx-series-ultrasonic-cleaners-model-cpx5800-capacity-2-5-gal-9-5l-l-x-w-x-d-15-6-x-15-8-x-14-9-in-397-x-401-x-378mm-230-240v/15336138) commenting the Crest performed alot better i decided to crunch some numbers.
What frequency of ultrasonic tank is best?
Today you can get ultrasonic systems with frequencies ranging from 20 to 950 kHz, which one to choose depends on what your cleaning job is, what type of soil is to be removed and how clean your part needs to be. In reality, most systems to day incorporate more than one frequency of ultrasonics. They may use 40 / 70 / 170 for a graduated cleaning process. A general list of frequencies that are most common follows.
20-40 kHz Heavy duty cleaning for things like engine blocks heavy metal, heavy greasy soils.
40-70 kHz General cleaning of machine parts optics etc. very good at removing small particles
70-200 kHz ultra fine gentle cleaning of optics, semiconductor wafers disk drives, etc.
How should I choose the correct power level for an ultrasonic tank?
The average Watts per gallon of ultrasonics should be between 70 to 100 Watts for critical cleaning applications. This is the average rating and can be adjusted dependent on the cleaning application. To calculate the power requirements use the following formula;
L x(in) W(in) x. (H -2") /231*100=Avg. Watts Power.
Branson (Tank Capacity 2.5 gallons @ 160 watts)
Should be between 175 and 250 watts
11.5 x 9.5 x 4 / 231 * 100 = 189.18 average watts needed.
Crest (Tank Capacity 1.5 gallons @ 120 watts)
Should be between 105 and 150 watts
11.75 x 7 x 4 / 231 * 100 = 106.81 average watts needed.
If the numbers i found are legitimate and i have not messed up any of the maths.
The Crest has more power than what it requires but sits at the bottom of the average for 'critical cleaning' and 5kHz closer to the 'ultra fine gentle cleaning' range
While the Branson was underpowered and closer to the mechanical parts range.
I'm not sure if these are the reasons why one performed better than the other, but I am hoping someone has some knowledge on the matter for making a general 'rule of thumb when buying'.
eg,
Min. Wattage = gallons * 70
Min. kHz = 40 with sweep but the higher the better
-
3roomlab, do you know what kHz it runs out and did you end up trying a PCB board?
-
I use one for electronics, and it's filled with purified water and mild cleaning detergent EM-80 if I recall correctly.
-
I have tap water in the tank. Cleaning agents (soap water, perclone d, IPA etc.) are in glass jars which are placed into the water filled tank. This way the tank stays clean and you can swap jars quicly. The ultrasonic wave penetrates the glass jar and cleans the items inside.
The cleaner have to be off when changing jars because the water level is too low without jars.
-
I bought a used US cleaner from Ebay, I think it is what dentists use? looks a lot like the 1.8L one on the liquid glass site, mine is quality made in Europe. It is 35kHz , and has a heater and a rotary type knob. I can fit most of my smaller PCB's in it ok.
I'm not sure if the frequency sweep is a gimmick on the expensive ones, the piezo unit is mechanically resonant at one frequency, but you do get nodes and antinodes within the tank, so possibly slight frequency shifts mix the modes up (like a stirrer in a microwave) . I just move the PCB's around every minute to get uniform cleaning.
Chemtools make PCB wash http://www.chemtools.com.au/product/electrical-and-electronics/electronic-cleaning-chemicals/pcb-flux-remover/ (http://www.chemtools.com.au/product/electrical-and-electronics/electronic-cleaning-chemicals/pcb-flux-remover/) I have a colleague who uses this in his U/S bath for cleaning PCB's, he is happy with it, you can buy from Chemtools direct, or through E14. I have some but rarely use it except as a final wash when stripping silicone rubber conf coating. It takes a lot of effort to rinse it off & makes your hands really slippery (gloves don't help much).
I normally use straight alchohol , either as methylated spirits (Cheap from Bunnings or Coles), (the pyridine denaturant makes your fingers smelly) or as IPA.
So I normally have about 3x 1 litre bottles of metho, so you have 2 "dirty" bottles and one "cleanish" bottle, so you do about 4 minutes ultrasonic in the dirty metho, moving the pcb's every minute, then you stop and give them a good scrub with assorted brushes, splash them in the tank again, swap over to the clean metho, do another 4 minutes (moving every minute), and give a final light brush with a clean paintbrush, splash in the metho again, then flick dry, and quick blow dry with an air-gun. If I can see chalky flux residues at this stage that won't rub off with gentle finger rub, they go back in with some more brushing. Otherwise I get some fresh IPA, and two fast-food containers, and just tip the IPA back and forth between the two tubs, running it over the PCB's. This rinsed IPA goes back in the "clean" metho container, and some cleanish metho goes into the dirty bottles. After about two weeks the gunk settles out, and you decant off the 95% of clear liquid and dump the dregs.
This worked well in the old days for hand made prototypes where I use syringed solder paste and a hot air gun, now I'm using no-clean paste , good stencils and a toaster oven, and don't usually need to immersion clean. I've found that the Chemtools liquid flux on a cotton cloth dissolves the gunky flux from hand soldering connectors, but this leaves a slightly sticky film, that is then removed with IPA on another cotton cloth.
Most cleaning methods will leave a faint chalky residue, you need to polish this off, I use a soft round artists brush (about 12mm across) and wrap masking tape around the bristles leaving 1mm sticking out, give your PCB a good brush with this and it will be shiny sparkly clean!
A combination of liquid flux and low residue production type solder for the THT, combined with a lot skill results in almost no-clean PCB's.
CAUTION all of the no-clean fluxes will seriously get up your nose, you must extract the air outside, filtering won't remove the vapours.
Also note different fluxes dissolve in different solvents- as you have probably found out.
If I had the money to spend, I'd buy a cheap U/S with a longer tank, and get a Dyson air-knife hand dryer as well. Getting the boards properly dry is a challenge. Compressed air used clumsily can generate enough static to kill CMOS and MOSFETs.
Note there are some components that can't go in the U/S bath : large can crystals, relays, accelerometers, gyros, most microphones and speakers, silicon rubber keypads, comb contact LCD's.
-
The human mind has strange ways... The title of this post reminded me of Pinky and the Brain...
- Are you thinking what I am thinking?
- I think so, Brain, but where are we going to find a duck and a hose at this hour?
-
I bought a used US cleaner from Ebay, I think it is what dentists use? looks a lot like the 1.8L one on the liquid glass site, mine is quality made in Europe. It is 35kHz , and has a heater and a rotary type knob. I can fit most of my smaller PCB's in it ok.
I'm not sure if the frequency sweep is a gimmick on the expensive ones, the piezo unit is mechanically resonant at one frequency, but you do get nodes and antinodes within the tank, so possibly slight frequency shifts mix the modes up (like a stirrer in a microwave) . I just move the PCB's around every minute to get uniform cleaning.
Chemtools make PCB wash http://www.chemtools.com.au/product/electrical-and-electronics/electronic-cleaning-chemicals/pcb-flux-remover/ (http://www.chemtools.com.au/product/electrical-and-electronics/electronic-cleaning-chemicals/pcb-flux-remover/) I have a colleague who uses this in his U/S bath for cleaning PCB's, he is happy with it, you can buy from Chemtools direct, or through E14. I have some but rarely use it except as a final wash when stripping silicone rubber conf coating. It takes a lot of effort to rinse it off & makes your hands really slippery (gloves don't help much).
I normally use straight alchohol , either as methylated spirits (Cheap from Bunnings or Coles), (the pyridine denaturant makes your fingers smelly) or as IPA.
So I normally have about 3x 1 litre bottles of metho, so you have 2 "dirty" bottles and one "cleanish" bottle, so you do about 4 minutes ultrasonic in the dirty metho, moving the pcb's every minute, then you stop and give them a good scrub with assorted brushes, splash them in the tank again, swap over to the clean metho, do another 4 minutes (moving every minute), and give a final light brush with a clean paintbrush, splash in the metho again, then flick dry, and quick blow dry with an air-gun. If I can see chalky flux residues at this stage that won't rub off with gentle finger rub, they go back in with some more brushing. Otherwise I get some fresh IPA, and two fast-food containers, and just tip the IPA back and forth between the two tubs, running it over the PCB's. This rinsed IPA goes back in the "clean" metho container, and some cleanish metho goes into the dirty bottles. After about two weeks the gunk settles out, and you decant off the 95% of clear liquid and dump the dregs.
This worked well in the old days for hand made prototypes where I use syringed solder paste and a hot air gun, now I'm using no-clean paste , good stencils and a toaster oven, and don't usually need to immersion clean. I've found that the Chemtools liquid flux on a cotton cloth dissolves the gunky flux from hand soldering connectors, but this leaves a slightly sticky film, that is then removed with IPA on another cotton cloth.
Most cleaning methods will leave a faint chalky residue, you need to polish this off, I use a soft round artists brush (about 12mm across) and wrap masking tape around the bristles leaving 1mm sticking out, give your PCB a good brush with this and it will be shiny sparkly clean!
A combination of liquid flux and low residue production type solder for the THT, combined with a lot skill results in almost no-clean PCB's.
CAUTION all of the no-clean fluxes will seriously get up your nose, you must extract the air outside, filtering won't remove the vapours.
Also note different fluxes dissolve in different solvents- as you have probably found out.
If I had the money to spend, I'd buy a cheap U/S with a longer tank, and get a Dyson air-knife hand dryer as well. Getting the boards properly dry is a challenge. Compressed air used clumsily can generate enough static to kill CMOS and MOSFETs.
Note there are some components that can't go in the U/S bath : large can crystals, relays, accelerometers, gyros, most microphones and speakers, silicon rubber keypads, comb contact LCD's.
I would have thought that using an ultrasonic cleaner would lessen the amount of manual cleaning steps. It still sounds like a lot of work to me. Am I missing something?
-
The U/S cleaner shakes out all the solder balls and solder paste from under the chips, so they don't short out later on.
You can't reach these bits with a brush.
It's a lot less effort than without the U/S, you only need light brushing to expose fresh gunk for the a ultrasonics to get to, as opposed to vigorous scrubbing.
Industry is pretty much all no-clean now. But if you want to conformal coat a PCB, it sticks better if its clean!
Also high impedance analog or CMOS circuits or high voltage circuits can have dust/lint stick to the flux residues and cause leakage issues later when it is humid.