Image Map Image Map
Page 12 of 14 FirstFirst ... 2891011121314 LastLast
Results 111 to 120 of 135

Thread: Frustrating Problem

  1. #111

    Default

    In my last post here I mentioned trying Godards Silver dip to clean the pins of silver plated IC's. I gave it a go today. I didn't dip the chips, that may be a step too far, just applied with small paint brush



    Before



    After



    Don't use it on non silver plated pins, you'll know when they are silver plated as they will have gone black.

    Pete

  2. #112

    Default

    Make sure to wash it off when done. The pins are only sliver plated. The metal below the silver will act like a battery and dissolve one of the two.
    Dwight

  3. #113
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    33,710
    Blog Entries
    18

    Default

    I don't think that I'd soak any electronic thing in that stuff. Almost all similar products (and that includes Silver Dip) use a dilute solution of thiourea and either sulfuric or phosphoric acid with a bit of surfactant (detergent). A jeweler's rouge cloth might be less corrosive and do the same job.

  4. #114

    Default

    Quote Originally Posted by Dwight Elvey View Post
    Make sure to wash it off when done. The pins are only sliver plated. The metal below the silver will act like a battery and dissolve one of the two.
    Dwight
    Done and the item wasn't dipped just lightly brushed on and then wiped with a damp cloth

  5. #115

    Default

    I would be inclined to agree with Chuck(G). The Goddards is likely to be very ionic and conductive too. The only way to be 100% sure it would be gone would be to soak the IC in deionized water for a day. Still it did a good job improving the silver.

    I have attached a modified portion of your photo. Of course if your IC is going back into the same socket, the important part of the IC pin is the part that touches the socket claw. You can see where that is from the residual black mark, and clearly this is not a TI brand socket because it didn't grab the sides.

    I have found that this dark material in all cases seems to represent a reaction between the dissimilar metal of the IC pin, its plating and the socket claw. Its actually a lot worse on other brands of IC that have either tin plating or pre-solder surfaces. In any case, its a sort of oxide which is an electrical insulator. It needs to be cleaned off in all cases and it does require physical abrasion . I use a small folded over piece of 2000 grade (very fine paper) wet with WD-40 as a lubricant, then clean that off with contact cleaner. When its done, the black mark should have disappeared.

    If course if its a TI socket, its important to use this paper to clean the oxides off the sides of the pins, where the physical contact with the socket claw is.

    Its also better I think if the pins are lubricated prior to re-fitting the IC to the socket (it also reduces socket wear if they have to be removed a few times) and since some of the silver (or other plating) has been lost it is better to have a corrosion inhibitor. TI pins are steel under the silver and they can rust later if there is too little plating left.

    The question is which lubricant/corrosion inhibitor to use? I had used WD-40 in the past which is actually great at rust prevention on bare steel, however it does induce a negative effect on brass & copper.

    I have been running an experiment, and just updated the article.

    It appears, from the results so far, the better product to use would be either Inox mx3, or Servisol WD, but I'm still waiting on the servisol complete result. Others tested induce hygroscopic corrosion on copper, though they do not chemically attack it and the effect depends on the thickness of the original film. The oil left behind by Inox mx3 appears long lasting as does WD Servisol:

    http://worldphaco.com/uploads/THE_GREAT_WD.pdf

    PS: There is a product available in the USA in the WD-40 Specialist range that is a long term corrosion inhibitor. I would dearly love to test it with the copper plate experiment, but it is not available in AU yet and there are no ebay sellers that will send it.
    Attached Images Attached Images
    Last edited by Hugo Holden; January 11th, 2020 at 01:39 PM.

  6. #116
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    33,710
    Blog Entries
    18

    Default

    Hugo, have you considered using a thin layer of dielectric grease? That might at least keep corrosion out. I use it on electrical contacts on automotive stuff and it works well.

  7. #117

    Default

    Quote Originally Posted by Chuck(G) View Post
    Hugo, have you considered using a thin layer of dielectric grease? That might at least keep corrosion out. I use it on electrical contacts on automotive stuff and it works well.
    Yes, I think it would work well like Vaseline which is very good at waterproofing too, however I was looking for a spray on product for the task, mainly to do a similar job to WD-40 but without the effects on copper & brass components and with longer retention of surface oil. After 3 to 4 months WD-40 has left the building. I had also considered lithium grease.

    The Inox mx3 is looking good for the task of IC's and sockets lubrication & protection. The manufacturers claim its safe on electronic parts, doesn't attack plastics etc. Seems correct, I've tested it on a number of plastics with no reaction seen. Yet it smells quite different to all the others, almost like acetone, but it cannot actually be that, as there is no reaction with plastics & inks it must be the propellant that gives the smell.

    It would really be interesting to know what the ingredients were in all of these products. For example, why Inox's Lanox, WD-40 and Selley's RP-7 all give the brown spot effect on copper and whether or not that is due to a common ingredient. And what makes Inox's Lanox different from the Inox mx3 aside from the lanolin.

  8. #118

    Default

    Mike

    Somebody may have already suggested this but I've only read part way through the whole 12 pages of posts and I haven't read the code in any detail either. The main things that seem be to be discussed are poor pin connections or the FDC. I did try to find data sheet for an 8272 could not not find one. However, it occurs to me that you may be looking in the wrong place. If you are failing to read multiple sectors and getting less or more than you expected. It could suggest a memory, ROM, processor issue.

    If you have skewed disks, I’d guess the FDC can’t deal with that. It can’t read multiple skewed sectors as it has no idea what the skew is. So the processor must issue a sector by sector requests to the FDC which fetches the sector and the result DMA’s into memory, interrupt and then the processor issues the request for the next sector. The number of sectors will be in RAM and the code is in the ROM. Maybe you’ve got a problem there? A bad RAM chip, EPROM cell failure after all these years? Maybe trying the cold spray in that area could find something?

    Pete

  9. #119
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    33,710
    Blog Entries
    18

    Default

    The Intel 8272 is exactly the same chip as the NEC µPD765; Both have "A" versions and other than saying NEC or Intel, the datasheets are the same. The same applies for the Zilog Z765A and others of the many, many alternate licensed clones.

  10. #120
    Join Date
    Dec 2013
    Location
    Near Milwaukee Wisconsin
    Posts
    1,254

    Default

    I did not forget about this. I got sick and could not participate. It hit me hard. I could not go into the basement and work. In fact any kind of anything, other than sleeping was difficult. Well, it seems to be over for the most part.

    Over the last few days I finished the rewire of my 8272 FDC board. It really looks nice. I plugged it in and CPM came right up. I shut the machine off and tried again and it worked. I loaded WordStar and tried a few documents. All seemed OK. Then I decided to try some drive intensive DIR's and STAT's over and over. Then it happened the same problem. The fault occured on a much less frequent basis, but the next day, the fault was occuring most of the time. Back to square one.

    Today, I connected up the Logic Analyzer, and looked at the DRQ, HRQ, HLDA, DACK0, INT and Terminal Count. I had to increase the sample rate on the LA from 4 nSec to 1 uSec, in order to see enough data. But after a few runs, the fault occurred after 4 sectors. Looks to me that the DAM controller is working fine. This is verified by looking at memory and I see complete sectors in the proper memory spaces. But what I see is that the FDC 8272, after giving up the last correct sector, gets a terminal count signal from the DMA controller and then issues an INT. BUT.... the INT goes high and never goes low. The FDC should drop the INT after the first Result Phase byte is read. So the program isn't getting to the last Result Phase. This is bore out by the fact that the program gets stuck in the timeout loop of the WAITINT program.

    So..... I'm not sure, as of now, that the problem is with the PIC or maybe the CPU. Anyway, It's time for a nap. Maybe tomorrow, if I feel up to it. Mike

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •