Date
1 - 20 of 24
Dark deposit on pins near leaking SMD electrolytic caps
|
Ozan
Hello,
Does anyone know the chemical composition of the dark deposits on IC pins that are close to leaking SMD caps? What is the best way to clean them? Litmus paper shows base (hard to tell because it gets stained by the deposit). I couldn't find a good solvent for it. Here is a picture from a TDS520A https://groups.io/g/TekScopes/photo/260554/0?p=Name,,,20,1,0,0 Interestingly sometimes one or two pins are covered but neighboring pins are OK (maybe voltage at the pin making a difference?). I have seen this in CPU card of 2467B, both acquisition and CPU cards of TDS520 and TDS520A. Ozan
|
|
|
|
On Wed, Feb 10, 2021 at 01:13 AM, Ozan wrote:
That's the result of electrolyte leakage from defective SMD ElCaps. It damages PCB's, vias and traces below a PCB's surface. It's a classic with 'scope models that you mention. If you want to prevent future failure of the 'scope, it's important to thoroughly clean the PCB's. This forum is full of references to what's happening, what to do about it and what happens if it isn't dealt with in time. Raymond
|
|
|
|
Ozan
Hi Raymond,
toggle quoted messageShow quoted text
I know SMD caps are the reason and I know it needs to be cleaned as soon as possible. I am looking for chemistry gurus in the group to understand more about its composition, why it happens on some pins. Could it be left over flux interacting with the electrolyte? What will dissolve it? I usually do DeOxit (hoping to stop oxidation - probably unnecessary), isopropyl alcohol, 409, distilled water in sequence but I am still concerned about under the ICs. I might have missed it but I couldn't find any posts talking about the chemistry of it. It gas from the cap (I don't see a liquid trail) somehow reacting around certain pins. Ozan
That's the result of electrolyte leakage from defective SMD ElCaps. It damages
|
|
|
|
Harvey White
Most people seem not to like 409 because of its effects on plastic. I've heard several people suggest that simple green should be used for cleaning off PC boards (and rinsed thoroughly afterwards!). Another option might just be dawn dish soap, which seems to be lacking in some of the chemicals in the (above two?) products that tend to eat plastic. Certainly, there are stories about 409 being very nasty on plastic.
toggle quoted messageShow quoted text
If you wanted to be quite safe, I'd use whatever Tektronix used on their scopes, realizing that most scopes of that era did not use PC boards. Harvey
On 2/9/2021 7:43 PM, Ozan wrote:
Hi Raymond,
|
|
|
|
On Wed, Feb 10, 2021 at 01:43 AM, Ozan wrote:
Hi Ozan, After I posted, I reread your post and realized that you already knew and what you're really looking for. A lot has been written about cleaning methods. AFAIR, thorough washing with an acidic solution has been suggested and neutralizing afterward by thorough neutral rinsing. I have used soaking for about an hour in 50% diluted household vinegar, rinsing and brushing off solids, soaking in washing-up liquid for about an hour, and very thorough rinsing and soaking in clean water, followed by thorough drying in a ventilated place at about 40C for at least an hour. I have treated two 2467B's, three 2465B's, a 2445B, and (AFAIR) two TDS520A's this way, all successfully, between three and five years ago. One 2465B and one TDS520A needed restoration work to become functional again. I have preemptively removed and replaced a number of devices, mostly opamps, where I feared damage underneath. I don't remember if others have given different advice, because I had already finished most of my 'scopes before it and didn't consider changing my method. Raymond
|
|
|
|
greenboxmaven
I have restored some Sony video recorders that were loaded with the bad condensers. I made notes about ratings and placement, removed them all, and scrubbed the boards with hot soapy water. After a good rinse and drying, I re-tinned the spots where the condensers go, and repeated the wash. After that, installed the new condensers. I had complete success.
toggle quoted messageShow quoted text
Bruce Gentry, KA2IVY
On 2/9/21 19:13, Ozan wrote:
Hello,
|
|
|
|
On Wed, Feb 10, 2021 at 02:38 AM, greenboxmaven wrote:
The affected Tek 'scopes use multilayer boards, making a slightly more intricate procedure advisable. Raymond
|
|
|
|
greenboxmaven
The Sonys have three layer boards, I didn't have any problems. However, I can imagine some very bad problems if the electrolyte got into a connection that connects with one of the inner layers. How would you deal with that?
toggle quoted messageShow quoted text
Bruce Gentry, KA2IVY
On 2/9/21 20:44, Raymond Domp Frank wrote:
On Wed, Feb 10, 2021 at 02:38 AM, greenboxmaven wrote:I have restored some Sony video recorders that were loaded with the badThe affected Tek 'scopes use multilayer boards, making a slightly more intricate procedure advisable.
|
|
|
|
On Wed, Feb 10, 2021 at 03:03 AM, greenboxmaven wrote:
Electrolyte does go into vias and damages them and inner layers. That's why your best hope is the thorough and time-consuming procedure that I described earlier. It has worked for me in most cases, at least in the 'scopes I mentioned earlier. Raymond
|
|
|
|
Ozan
Hi Harvey,
Most people seem not to like 409 because of its effects on plastic.Good info about 409. Although I don't keep it long on the PCB it is quite strong, time to switch to Simple Green. Ozan
|
|
|
|
Ozan
Hi Raymond,
A lot has been written about cleaning methods. AFAIR, thorough washing with anGood idea with 50% diluted vinegar to neutralize the deposit. The deposit seems to be basic (opposite of acidic). I thought DeOxit would be acidic (pH is not disclosed in SDS) but it seems to be too mild to neutralize electrolyte deposit. Ozan
|
|
|
|
Chris Wilkson
The classic recipe for battery leakage on PCBs is a toothbrush scrub (actually short bristled hogs hair) in a white vinegar rinse. Follow that with a scrub in a baking soda rinse to neutralize the acid. Followed by a "clean water" rinse: de-ionized water if you have it or can afford it, but distilled water is plenty good enough. There's no reason to use tap water...distilled is cheap enough. Last step is a compressed air dying and an oven bake at maybe 110-120C for a "long enough time".
With SMT devices a warm soak is probably better at each stage. To give the cleaning chemicals a chance to get in under the parts. I was given a 7104 whose filter caps leaked and destroyed the power supply boards. I don't know the chemistry of the leaked electrolyte, but that was some *nasty* stuff! It ate the metal and the substrate of the PCBs. Of course it had been sitting undisturbed on a shelf for probably 20 years, eating away. Having nothing to lose I tried the battery cleanup process and it worked great. The vinegar fizzed up pretty nicely too. About 7-8 years later I peeked inside again and there was no obvious sign of further damage.
|
|
|
|
Ozan
Hi Chris,
Thanks for the recipe, in old times most cells were acidic so baking soda was first but for an alkaline electrolyte vinegar first makes sense. 409 is very basic (probably too much, Simple Green is suggested) so it might do the job of baking soda. One mystery is why only some IC pins get this dark deposit, there are ICs nearby with no deposit and in some ICs only few pins would get this deposit. It is not always the closest to the cap. Ozan
|
|
|
|
Harvey White
I'm not the first to mention the problems of Formula 409 on plastic. However, I should mention that Simple Green may have some of the same chemicals, which indicates that a thorough rinse is highly suggested. The TDS 500 and 600 series have potentially nasty capacitors in them, so part of the re-capping process is to go through and do a scrub with simple green, rinse, dry thoroughly, (possibly use 91% alcohol to help remove the water), then re-cap as needed.
toggle quoted messageShow quoted text
Harvey
On 2/9/2021 9:20 PM, Ozan wrote:
Hi Harvey,Most people seem not to like 409 because of its effects on plastic.Good info about 409. Although I don't keep it long on the PCB it is quite strong, time to switch to Simple Green.
|
|
|
|
Jim Ford
3-layer boards? Never heard of them. Maybe you mean 3 dielectric layers and 4 metal layers? PCBs are built up from the inside out, starting with a core (fully cured dielectric sheet plated on both sides with copper foil) and adding more cores and/or prepregs (partially cured dielectric sheet plated on one side with copper) symmetrically on both sides until the stackup is done. Hence there is always an even number of (metal) layers. Always!
toggle quoted messageShow quoted text
Jim Ford
------ Original Message ------
From: "greenboxmaven via groups.io" <ka2ivy=verizon.net@groups.io> To: TekScopes@groups.io Sent: 2/9/2021 6:02:16 PM Subject: Re: [TekScopes] Dark deposit on pins near leaking SMD electrolytic caps The Sonys have three layer boards, I didn't have any problems. However, I can imagine some very bad problems if the electrolyte got into a connection that connects with one of the inner layers. How would you deal with that?
|
|
|
|
Harvey White
Physically, one layer, copper on top.
toggle quoted messageShow quoted text
inner layer (* see note) outer layer, copper on bottom. That inner layer can be single sided or two sided, depending. I don't think that anything in the manufacturing process keeps it from being so. I've actually made 3 layer boards, although they're quite awkward without plated through holes. One side of that inner layer can be missing, so yes, a 3 layer board. I guess it was cheaper to make for some reason, but that was then, long ago (in electron lives) then... Harvey
On 2/9/2021 11:33 PM, Jim Ford wrote:
3-layer boards? Never heard of them. Maybe you mean 3 dielectric layers and 4 metal layers? PCBs are built up from the inside out, starting with a core (fully cured dielectric sheet plated on both sides with copper foil) and adding more cores and/or prepregs (partially cured dielectric sheet plated on one side with copper) symmetrically on both sides until the stackup is done. Hence there is always an even number of (metal) layers. Always!
|
|
|
|
Ed Breya
The dark stuff is probably mostly iron (Fe+3) compounds from the steel in the IC leadframe. The electrolyte from the capacitor plague results in all sorts of reactions - galvanic, just from the different metals sitting there, and electrolytic, driven by currents when stuff is powered up. It makes a real mess, and has likely ruined countless pieces of equipment. I have two Tek scopes out of commission from it. One (TDS540 I think) is because it's so tedious to track down the damage, and the other (TDS820) is because I can't find schematics needed to track down the damage, making it nearly impossible.
Ed
|
|
|
|
Ed Breya
Forgot to mention - regardless of the corrosion reactions and products and pH, always start with an acidic wash, and end with an alkaline one, then lots of flushing with hot water, then thorough drying.
Ed
|
|
|
|
Jim Ford
OK. I will change my story to I've never seen 3-layer boards.
toggle quoted messageShow quoted text
Usually when I see something that doesn't make sense, I attribute it to "It must have been less expensive to do it that way." Thanks, Harvey. Jim
------ Original Message ------
From: "Harvey White" <madyn@dragonworks.info> To: TekScopes@groups.io Sent: 2/9/2021 8:53:30 PM Subject: Re: [TekScopes] Dark deposit on pins near leaking SMD electrolytic caps Physically, one layer, copper on top.
|
|
|
|
Chuck Harris <cfharris@...>
Back when 2 layer boards were not quite good enough, and 4 layer
toggle quoted messageShow quoted text
boards attached a pretty hefty premium, some, including myself, made 3 layer boards. A 4 layer board can be made in a couple of different ways: 1) SS bonded to DS bonded to SS 2) DS bonded to blank bonded to DS In either case , there are 3 fiberglass layers, and 4 metal layers. A 3 layer board is made by a SS bonded to a DS. The 3 layer board has only 2 fiberglass layers, and 3 metal layers. They work well when the boards are small, but larger sized boards are bound to be warped from the start. Other than cost, 3 layer boards work well with TTL circuitry, as you typically use the middle layer as a ground plane, and salt little ceramic capacitors to every chip's VCC pin. -Chuck Harris Jim Ford wrote:
OK. I will change my story to I've never seen 3-layer boards.
|
|
|