Fingernail polish and RF front ends, do they mix?

[cdev]

Can I use hard, clear fingernail polish as a conformal coating for RF filters, LNAs, and so on, specifically when I am coating the input side?

Any residue or grease, solder flux, etc impacts performance, I can tell, although this is hard for me to measure. Should I use something else there?

(thats not super expensive)

[richnormand]

I remember the subject from an SMTA presentation a while back.

https://www.smta.org/chapters/files/UMW_Conformal_Coatings_Workshop_-_Final.pdf
https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=31393
https://www.microwaves101.com/encyclopedias/rf-printed-wiring-board-hints

At the end of the day it all depends on the situation  (like corrosion, non-hermetic case) but they do have different dielectric coefficients and loss factors.
Might be important on a surface waveguide on a pcb and less for a small compact component.

Do you have access to a good VNA at frequency of interest to actually measure the component or pcb guide geometry with before and after measurement.
Of course this will not help much for potential aging effects.

I posted the MG Chemicals stuff I use in your other post about soldering but it seems to work well in the 2.4G region for a thin coating.

[coppercone2]

I recently invested in various glues and adhesives from loctite and 3m, the good ones, and even the expired products kick ass.

I am so tired of using random consumer chemical shit. It's worth investing here, I got a mini fridge to keep the chemicals in also. Get the dispencer, bunch of pipettes, the mixing nozzles, little cups, masking tape, sand paper kit, lots of alcohol..

I am SO tired of dealing with being cheap with glue. Paint you can get away with, but glue/coating.. you need to pay the piper here.

I have a big coils that is peeling epoxy because I was going crazy thinking about buying actual adhesive then trying home made thinned stuff. No more of that please. I just stopped working on a big project because the fucking coil I carefully made and installed started to peel (eevblog advise was right but I chose to ignore it).

I am starting to think of the shop as a kitchen, eventually you get some bullshit meals if you keep trying to use everything in the refrigerator. It is not the siege of Leningrad. There is some middle ground between waste and just being too resourceful.

I manage to do wonders with scrap metal, at the expense of my time, but you need to be careful using flap disks and sand paper and paint strippers to stay profitable. Glue/chemical products are a disaster though... you can't really do much to improve quality being resourceful. Also, with metal, the fucking tools begin to wear out of you deal with too much garbage. My oscillatory sander started to make noise, I think I put too much rust through it (using the grinder became inconvenient after a while, too messy etc, especially working with thin plates) So much nicer to have magnetic fixture and a dust extractor. A way to minimize use of rust remover is to put thin stuff in ziplock bags and fold them over with tape.

[coppercone2]

I remember the subject from an SMTA presentation a while back.

https://www.smta.org/chapters/files/UMW_Conformal_Coatings_Workshop_-_Final.pdf
https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=31393
https://www.microwaves101.com/encyclopedias/rf-printed-wiring-board-hints

At the end of the day it all depends on the situation  (like corrosion, non-hermetic case) but they do have different dielectric coefficients and loss factors.
Might be important on a surface waveguide on a pcb and less for a small compact component.

Do you have access to a good VNA at frequency of interest to actually measure the component or pcb guide geometry with before and after measurement.
Of course this will not help much for potential aging effects.

I posted the MG Chemicals stuff I use in your other post about soldering but it seems to work well in the 2.4G region for a thin coating.

Using a VNA to determine if you should steal your wifes old nail polish is too much dude!

[joeqsmith]

Can I use hard, clear fingernail polish as a conformal coating for RF filters, LNAs, and so on, specifically when I am coating the input side?

Any residue or grease, solder flux, etc impacts performance, I can tell, although this is hard for me to measure. Should I use something else there?

(thats not super expensive)

      I admit, I am now on my second bottle of Revlon 771 clear in maybe 30 years.   The stuff keeps very well assuming you tighten the cap.  It comes with a nice little applicator brush.   I use it to coat lettering.  Rub on types as well as markers.   If I have a lot to do, I will use a spray can of enamel. 

Normally if I am coating components, it's with dope.  It's been years since I worked on designs that required conformal coating.  Dow Chemical and Humiseal are the two brands that come to mind. 

I would say a simple question, you will get a simple answer.  We had that person posting about swearing by using vaseline as dielectric grease in  their meter.  Others swear by spit although that seems to be an Australia thing.  Maybe read the following or do a little Googling. 
https://www.microwavejournal.com/blogs/1-rog-blog/post/26221-finding-the-best-blend-of-pcb-solder-mask 

Shown coating an 0603 10pF capacitor with two coats while running at 280MHz.  Allowing 12 minutes of dry time between coats. 
   

[joeqsmith]

Working on other things so it's no big deal to let this Nano collect data. 

PCB that I attempted to spatter a fair amount of flux on.   This time sweeping from 10MHz to 280MHz (vertical is in pF).  I started the collection, cleaned the board but it looks like it caused to much disturbance to see anything.  2pF isn't much.  I then applied three coats allowing for a longer dry time.  Each additional coat had less of an effect.   It appears to return to the baseline once it cures (at least in this range). 

[joeqsmith]

I've been working on the software for the old HP anyway so here is a second test board.   This time I slowed the collection rate down (it's a lot of data to post process)  and let it cure for much longer.   I swept it from 10M to 2G.   Shown is data for 10M-300M and at 1.9G to 2G.   

Using a VNA to determine if you should steal your wifes old nail polish is too much dude!

Of course, if using a few thousands of dollar VNA to look at a $12 bottle of nail polish wasn't silly enough,  the next step would be to have a look with the electrometer and see what we get for resistance.   

[coppercone2]

its not just silly, its going to be inconsistent, they have no standards to adhere to other then looks good and is non toxic for nail polish, who knows when the sales director in the company feels like letting a bad shipment through that does not stick so well for a fashion company (oh they change the nail color every day now according to new marketing data, who cares if we make it a little weaker in the summer months when consumers are dating more). People make money by making inconsistent products and carefully managing supply lines based on distribution region, unlike professional/semi professional products. (that disclaimer about using product outside of use makes them money).

" I know, we made a crappy batch, but you know I talk to the people in the xyz fashion store chain in the xyz region, they have consistant customers, we can get rid of it there safely).... we don't expect the backlash to be that bad there " (so long they think they can make money and not risk reputation too much). Also, a good reason why to buy from a distributor if you are running a business, rather then buying direct from another company, because their sales guy might 'get to know you'.  (this is why some people don't like giving company tours to sub assembly vendors). Then at least, you are protected by the whole distribution region.. sorry to hear if you are in a poorer country. One example is European potato chips (poorer countries get worse quality chips like Lays.. expect fragmentation when you open a bag in eastern Europe.. (no, its not shipping problems lol, its practically racism). Brand name is not a brand name with these guys.

Supplier asking you too many questions about something non suspicious? Its a sign they are binning their products.

IMO dangerous because someone might get the idea that your measurements will be enforced by the company and it could lead to failures. Well dangerous in the sense that you might end up with a bunch of bullshit that has problems down the road, and for cheap companies. At least structural glue has to not fall apart, so its safer for consumers.. but cosmetics......

i don't know if there will ever be a reliable industry watchdog for nail polish RF impedance. this is why i don't like it being used for trimmers, because someone on a assembly line doing something important might think they can get away with it.

Another example of misusing parts for electronics: I decided to make a bus bar to hold an electrode by filling a copper pipe with solder. I put what seemed like too much solder into the thing to fill it, while it was carefully heated by a torch after the pipe was cleaned. I look on the floor and there is drippings of solder. And it turned out the side of the pipe disintegrated and was leaking solder (careful to all the crazy people using plumbing for water cooled induction heater coils, I had a hunch it was a bad idea and this was for a 6VAC application). I cut it carefully and did nothing wrong to the pipe, fluxed it with proper flux, etc.. (everything correct, gentle flame). Looked perfectly fine during visual inspection.

[Bud]

I'd be wortied about temperature variations on the PCB and coated parts. Depending on what your circuit is temperature swing may be large. The coating may crack and let moisture in and all that.

[coppercone2]

I'd be wortied about temperature variations on the PCB and coated parts. Depending on what your circuit is temperature swing may be large. The coating may crack and let moisture in and all that.

that too, nail polish temperature is damn well stabilized by humans.

[joeqsmith]

I would hope no one would take my nail polish testing for anything beyond what little entertainment value it has.  Then again, I would have never considered spit and vaseline for use in a handheld DMM but here we are.  So there's always a few.   I have a hard time considering that a company would ever use it.   At $12/bottle, even dope is less expensive.   For that matter where would you buy your gallon containers of Revlon?!    

As for you comment on inconsistency,  I would like to add that the tests themselves are not going to be consistent.  We don't have a documented procedure that I am following,  I am certainly not controlling the thickness that I apply, time between coats, etc.     

I'd be wortied about temperature variations on the PCB and coated parts. Depending on what your circuit is temperature swing may be large. The coating may crack and let moisture in and all that.

Good point.  I have some things that I have lettered many years ago and have not seen and visual signs of it clouding up.    Certainly the clear coats used in automotive would be much less expensive, and they get tested year around in the elements.... 


**** Back to the testing ****
To measure the resistance, I have placed a virgin test board in my small fixture.  The sheets of Teflon are required to minimize leakage.  Once installed it was cleaned with ProClean.   After it dried, the electrometer railed out.   Next I applied a few drops of 771 to the board and sealed the box.   

My electrometer is very old, like most of my home equipment.   It doesn't have a digital interface but it does have an analog output that can be read with a volt meter.   The voltage is inverted and scaled for the range selected.  In the attached graph, the electrometer is reading 120Gohms.    It is sampling once a second so you are looking at 33 minutes of dry time.   

[coppercone2]

buy a 15$ tube of deoxit for the 300 $ multimeter 

now I need to deal with peoples spit in contacts. thats right out of fucking Alien. When you start factoring in cleaning efforts into ebay purchases its so much less appealing. Perhaps chew has anti oxidizing properties too.

Thank god OCD is prevalent in electronics people.

[joeqsmith]

less appealing. Perhaps chew has anti oxidizing properties too.

About the only cleaners I use with electronics are isopropyl alcohol and ProClean.  I keep some dielectric grease on hand.  About the only coating I use is dope. 

Vaseline, not so good....
https://youtu.be/e_YzwO62feQ?t=2184

[joeqsmith]

My electrometer can only read up to 200Gohm but the output amplifier will rail around 2.7V.   Its non-linear but we can at least see that after 2 hours of drying, the resistance is back to the limits of what this meter can read.    To measure higher values,  I would measure the leakage current with a separate HV power supply (note the SHV connector mounted to the box). 

[coppercone2]

I suppose I can try, I have a 1200V low noise supply and a electrometer (1TOhm) next to each other, but no nail polish products

[cdev]

I remember the subject from an SMTA presentation a while back.

https://www.smta.org/chapters/files/UMW_Conformal_Coatings_Workshop_-_Final.pdf
https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=31393
https://www.microwaves101.com/encyclopedias/rf-printed-wiring-board-hints

At the end of the day it all depends on the situation  (like corrosion, non-hermetic case) but they do have different dielectric coefficients and loss factors.
Might be important on a surface waveguide on a pcb and less for a small compact component.

Do you have access to a good VNA at frequency of interest to actually measure the component or pcb guide geometry with before and after measurement.
Of course this will not help much for potential aging effects.

I posted the MG Chemicals stuff I use in your other post about soldering but it seems to work well in the 2.4G region for a thin coating.

I think I will probably get one of the NanoVNAs - it just seems to bring so much to the table, I'd be an idiot not to get one. Such a great learning aid. I already have a zillion things I think I could do with it.   

Thank you for these PDFs, This is a good incentive to really learn how this is done using a real electronics workflow, using tools like Kicad, LT Spice, QUCS, etc.

[RoGeorge]

If I were you I'll wait for NanoVNA 2, ask OwO when it will be ready for sale.