[weblacky]

While I kind of get that, I ordered items there were in stock and my last order (which was like 3 weeks ago, was shipped out same day). My experience has been 3-4 days before shipment at worst (BEFORE Pandemic), I hating having to chain my orders but unless I want to pay $40+ a shipment for next day shipping, it's always on a delayed schedule anyway...ughhh.

I wish I lived next to their warehouse and could just walk in with a basket and grab what I needed! Why don't those exist anywhere in a major US city? I'd even drive down to downtown if there was a such a place (pick-up in person). The only good thing is that as more PSU parts are discovered by me, the larger each order can get. So hopefully I can do larger orders (which means fewer orders).

[robespierre]

While I kind of get that, I ordered items there were in stock and my last order (which was like 3 weeks ago, was shipped out same day). My experience has been 3-4 days before shipment at worst (BEFORE Pandemic), I hating having to chain my orders but unless I want to pay $40+ a shipment for next day shipping, it's always on a delayed schedule anyway...ughhh.

Note that unless they changed it, you can get free ground shipping with Digi-Key if you mail them a physical check. It's a little perk that few know about. There are also free shipping above $XXX offers from several distributors (Arrow...) that can save some money if you batch together projects into one order.

I wish I lived next to their warehouse and could just walk in with a basket and grab what I needed! Why don't those exist anywhere in a major US city? I'd even drive down to downtown if there was a such a place (pick-up in person). The only good thing is that as more PSU parts are discovered by me, the larger each order can get. So hopefully I can do larger orders (which means fewer orders).

They don't exist in major cities because A. the rent is too damn high (cue Jimmy McMillan) B. few if any local TV/Radio repair shops to buy parts and C. the hobby is much smaller than it once was.

Al Lasher Electronics in Berkeley, CA was one of the last of the old-time shops in a major city.
https://www.berkeleyside.com/2020/12/03/...oronavirus

[weblacky]

Good thing I selected US Postal Priority Mail (I live close to a major airport), the caps arrived a day earlier than predicted, today!. So we'll see when I can get them in. I still need to rig-up a PSU connector harness to my new DC load tester, so that will slow down my first bench-start attempt for a few days.

[weblacky]

Hi All,
Sorry for the huge delay, real life stuff BS, as always...anyway.  I went ahead and installed all the new caps I researched for the Indy PSU project and created a DC load testing harness (specifically to hook the 3.3v and 5v lines for now, but I have wires for everything).  I used a 150W limiting light bulb in my initial PSU run, I didn't run the PSU to max), for basic 2A-3A draw on both lines, the PSU started right up and the voltages appear OK, for now. 

See: https://siliconimage.irixnet.org/index.p...k/IMG_1244

I'm not 100%, sure but I think this is the PSU I bought from Slime-Eater, as it was so clean, that's why I started with it first.  So right now things look good, I removed all the adhesive so I do need to reapply something for proper final assembly.  I'm likely going to try the online recommendation of "GE Silicone II" clear.  

I have caps for two more PSU rebuilds, so I'll look into starting those.  Once I have a few PSUs I need to troubleshoot a tant-burnt Indy and I need to rebuild my Primary collection's Indy PSU so I can update it with all the wonderful upgrade parts I've purchased over the years!

Due to recent events/purchases I need to finish the basics on what I've started on the Indy PSUs, then I'm going to move directly to Tezro PSUs as I now need to get my Tezros up for firmware work and hardware upgrades.    It's more than possible this effort will put a pin on the Tezro/Origin 1.8v Aux line dip or not.  I plan on doing the PSU and then running it as-is, if the line ends up sagging then I'll do the SMD cap replacements in hopes that will cure it (should it reoccur).  On mine, it did it very quickly after startup.  So 20-30 minutes of running should prove it to me.  However even if it doesn't, the bad ripple exposure may mean I still need to do a few SMD cap replacements regardless, for safe measure, but 'll try to be systematic about it so we get some form of answer.

Stay tuned for more!

[robespierre]

The glue used was probably DOWSIL 738. It is thicker than Silicone II and more rubbery.

[weblacky]

The glue used was probably DOWSIL 738. It is thicker than Silicone II and more rubbery.

I'll look into it, but what's in there currently is a transparent-brownish, golden rubbery stuff that smells like a moldy attic when you melt it....

[weblacky]

The glue used was probably DOWSIL 738. It is thicker than Silicone II and more rubbery.

 
I did some research and took your advice, DOW does claim this formula is used to hold caps and coils (stabilizer).  Even though that's not what was used in '93, I know that's what they use now!  I bought a few small tubes.

I also rebuilt another Indy PSU, works fine.  Now I only have one rebuild kit left, but upon revisiting digikey, mouser, etc...the caps I'm using are all going out of stock and new stock is not expected until Dec 2021-Feb 2022!!! One of the special caps has gone out of production (and out on DigiKey) but I was able to order 100 of them on mouser and that should do 50 PSUs!  The others aren't out of production, just out of stock.

Upon thinking this over, I've decided that AFTER I rebuild all my Indy 7 PSUs, I'm going to try to offer a rebuild CAP KIT for the Indy (NIDEC), I've not gotten into my Sony to get a BOM for it yet.  These are the correct family upgrades from vintage Nichicon and current Nichicon production lines! (Not lower ESR and other troubling possibilities).

I have learned several things though, 
1. Some of the holes in the PCB are so tight that you really NEED to have a desoldering pump to even hope to get them fully cleared.

2. I'm using the 400F knife-tip soldering iron to melt through the old factory stabilizer.  Three of the caps are heavily gooped onto other components and one of those is directly against the logic board PCB with goop.  You need to be to be able to melt through this stuff correctly, else you risk damage from wrenching, pulling, breaking using force on the goop.

3. Some connectors are gooped and require pliers like this to even free (caps can be pulled with these too).  I recommend these pliers: EZRED KWP2 Kiwi Bent Head Needle-Nose Pliers: https://www.amazon.com/dp/B007QV4PZM/ref...Z05KM6S30J

So in order to do a rebuild you need a fair number of high-quality tools, without them, the risk of damage/failure is much higher.

[robespierre]

That's worth knowing. If you can't cut through the glue with side-cutters, it's definitely not silicone. Maybe some kind of acrylic. I suppose you have tried solvents: they are sometimes amazingly effective at softening such materials. But obviously anything like acetone or stronger can remove silkscreen labels and even swell the PCB if it drips, so you have to be careful.

[weblacky]

"Cutting" wasn't really an option due to space constraints and time. Like a heated wire threaded (then pulled) through the material would also work. The coils have bulging wiring (uneven surfaces) that interfere with getting a good, straight, path. Most of these glued components are so close together that you need to melt into them, then as you melt further down, you are actually pushing the neighboring components away from each as you are forcing the soldering iron deeper.

Several are clusters of three components glued as a union. Given the age of the components and lack of documentation I'm not going to warp or torsion the parts by using mechanical force.

I will try using the knife tip on the 738 RTV, most RTVs will liquify before burning, so I still expect the hardened RTV to be separable via melting.

In terms of solvents and such, these PCBS are....special? I've never seen this done...so instead of laying thick copper tracks, the manufacturer built up the track on the base copper with lots solder THEN applied some kind of solder mask OVER IT...yeah...bilt up hills of solder, with green solder mask then placed over the solder. I assume it's a cost-cutting measure. So when you go to melt the solder, you can actually melt/remove the solder mask applied on top. This "mask" is actually very fragile, even light mechanical rubbing will start to remove it). So while I quick shot of IPA isn't a problem...I think harsher stuff will affect it.

The board is essentially single-sided, even using the above trick.

[robespierre]

I think I've seen that "trick" used on PSU boards before. It's to get higher current capacity out of the traces. They are probably made with 2oz copper already, but there is a limit to how thick it can be and still be etchable.
High voltage creepage requirements usually mandate single-sided boards.
The methodology may need to be, remove the solder mask (solvent), desolder components, remove glue, remove components, replace components, solder components, add solder to heavy traces, clean, finally apply green acrylic coating.