• Please review our updated Terms and Rules here

I wish to create a new DMA/RAM expansion card for the Tandy 1000 line.

Is there anyway they can do a proper double-tall male header with offset shroud like on the original RAM expansion Plus board? That way, whatever designs come out of this thread could be a proper replacement for the Plus RAM expansion allowing two more Plus boards to mount on-top.

I haven't tested this yet because I'm currently not caring a lot about cross compatibility with original boards, but my ad-hoc measurements suggest that if I were to use a "double-ended" male header (they sell trimmable 2x40s at my local shop) and shim it while soldering I could get the "low" slot to the proper .7" spacing. (If you use a normal header you end up with .5".) Likewise using a normal header for the "high" slot and plugging in a set of trimmed "extra tall" stacking headers as an extender gets you about the right height for the upper slot. So it is *possible* to cobble it together. But it certainly isn't optimal. Shrouds would also be really nice. You could probably 3d print a sleeve to fit around the suggestions above but it wouldn't be fantastic.
 
Yeah, the 62-pin headers I've been getting from Digi-Key are too tight. The first couple that I ordered from a different manufacturer actually did fit properly, but they went out of stock. I'd definitely be interested if getting a handfull if you can get or find ones that actually fit properly.

I don't have the original Tandy board to compare it with, but it does definitely give me the willies how much insertion/removal force it seems to take to get boards in and out of these machines with the headers I've been able to find. I'm frankly worried the traces are going to start cracking after too many cycles.
 
I don't have the original Tandy board to compare it with, but it does definitely give me the willies how much insertion/removal force it seems to take to get boards in and out of these machines with the headers I've been able to find. I'm frankly worried the traces are going to start cracking after too many cycles.

I'd say about 2-3x as much. This is fine if you're card doesn't have a back panel mounting bracket, and the socket is all that's holding it in place. Then it's not going anywhere.

But you need to carefully wiggle it out a little bit each corner at a time.
 
I am working with a vendor in China to have some proper 2x31 headers made up rather than piecing together 2 other sized headers, or getting a 2x32 and chopping off one row. Once I get the particulars worked out, I will share my vendor, or if anyone wants a few.. I am planning on ordering plenty of extras.

Did you manage to get those headers made? Adrian's latest video said you did. If so, I think everybody on this thread would be interested in getting a few.
 
They have been made, 250 of them are in a DHL plane as I type this. I should likely have them on Friday or Monday.

I did get a small sample of 5 that I have tried out, and I am quite happy with them. Yes, they are tighter than the Tandy board, but I really chock that up to being a brand-new connector with not yet stretched out contacts. I find that after a dozen insertions, they loosen up quite a bit. Since the Tandy cards have most likely been installed in computers for hte last 30 years, they are already pre-stretched out.
 
They have been made, 250 of them are in a DHL plane as I type this. I should likely have them on Friday or Monday.

I did get a small sample of 5 that I have tried out, and I am quite happy with them. Yes, they are tighter than the Tandy board, but I really chock that up to being a brand-new connector with not yet stretched out contacts. I find that after a dozen insertions, they loosen up quite a bit. Since the Tandy cards have most likely been installed in computers for hte last 30 years, they are already pre-stretched out.

The ones I've been getting from Digi-Key do START to loosen after a while, but after pulling it out over and over on my XT-IDE card (probably at least 20 times for testing), I ended up lifting the solder pads joining the back plate PCB. I'll have to add some THP vias to the pads for the next rev before I put it up on GitHub, maybe 30-40 mils or so, something that you could flow solder into for reinforcement.

How much you figure to sell them for?
 
After shipping, they worked out to about $0.47/ea in the quantity that I bought. I'd just round up to $0.50/ea plus shipping.
 
After shipping, they worked out to about $0.47/ea in the quantity that I bought. I'd just round up to $0.50/ea plus shipping.

So that rounds to, what? ~$120 USD to get the batch of 250 made? If the quality's reliable, that's a really good deal. Who did you go with to get them made? There's probably other (now) defunct connectors we can think of to get made up.

Do you have a Paypal or something set up? I figure to order a batch of 10. That should keep me going for a little bit.
 
The connectors themselves were $0.30/ea but DHL shipping and bank fees are what increased the price by 50%. Even then, they are far cheaper than $3-4 per connector from Mouser or Digikey..

It's a good deal as long as the 250 are the same quality as the sample 5 they sent me. Once they get here, and assuming they are all in good condition, I will be more than happy to share the manufacturer. But yes, I dont mind parting with small quantities once they arrive.
 
The connectors themselves were $0.30/ea but DHL shipping and bank fees are what increased the price by 50%. Even then, they are far cheaper than $3-4 per connector from Mouser or Digikey.

Would you be able to share the contact info for this vendor you found to do the work? I'm wondering if they'd charge in the same ballpark for custom-height shrouded male connectors. Maybe we could put together a group buy for people working on boards that include the stacking functionality.

I find myself more interested in finding an easy off-the-shelf solution for getting the proper .7" spacing for the lower piggyback board since I finally found what was causing a maddeningly intermittent issue with the serial ports on my new board. I kept having this issue where the mouse driver would tell me "device not found" when booting in the 1000HX; it didn't happen in the 1000 EX, so I churned through all sorts of possibilities trying to find it. The punch line turned out to be that one of the jumpers on the RAM board in the HX was *just* tall enough to touch the bottom of the piggyback board and screw with it. (Just eyeballing it it *looks* like the battery holder is the tallest part, and I had a piece of electrical tape over the positive contact to make sure that wouldn't be a problem, but apparently eyeballs can deceive you. And of course this meant if I shuffled just the serial/IDE boards between systems, which I was mostly doing because I'm still running the old prototype RAM board in the EX which doesn't allow the optimal memory map for the HX, the problem went away because the component positions were different.) Taping an insulating sheet on the upper board solved the problem... but so would the proper extra .2" of clearance. (And also would be the "right" way, just in case someone actually wanted to mix new boards with an old Plus card.) Feh.
 
Would you be able to share the contact info for this vendor you found to do the work? I'm wondering if they'd charge in the same ballpark for custom-height shrouded male connectors. Maybe we could put together a group buy for people working on boards that include the stacking functionality.

I find myself more interested in finding an easy off-the-shelf solution for getting the proper .7" spacing for the lower piggyback board since I finally found what was causing a maddeningly intermittent issue with the serial ports on my new board. I kept having this issue where the mouse driver would tell me "device not found" when booting in the 1000HX; it didn't happen in the 1000 EX, so I churned through all sorts of possibilities trying to find it. The punch line turned out to be that one of the jumpers on the RAM board in the HX was *just* tall enough to touch the bottom of the piggyback board and screw with it. .

Oh man I "love" those gotcha type problems. A few years ago I restored a Xenon pinball for my collection and on initial power up the computer controlled bulbs were firing off like flash bulbs. Long story short, it turned out a prior owner had replaced a machine screw with a wood screw that had the exact same head, and it passed through the playfield and bridged 70v onto the bulb bus which is like 5 or 7v. It took days (weeks) of tearing apart the harness and looking for the cause to find that problem.
 
Oh man I "love" those gotcha type problems.

Man, I didn't even mention the best part of the circus (here). By fantastically unlucky coincidence a few weeks ago I happened across a stash of old CF cards of various sizes; mostly 128MB-512MB, but there were two 2GB ones, and for some idiotic reason I thought it a good idea to swap these CFs into *both* machines. (In one case because, you know, bigger than 256MB, and the other to free up the PATA->SD adapter I'd initially been using.) Apparently those CFs have compatibility issues with XT-CF though; despite seemingly working fine at first both machines gradually started acting buggier and buggier. The most common symptom was they'd crash with an "Invalid command.com" error when exiting programs, but they'd also crash halfway through booting, I'd get strange random crashes or obvious corruption inside software... basically both machines started coincidentally going to hell around the time I'd swapped boards a few times. To say the least I was getting to the pulling-my-hair-out-with-frustration stage of W-T-F. I was convinced that maybe I'd started cracking the motherboard traces from the stress of pulling the boards in and out. Finally, around the time I started realizing there was actual file corruption going on even with files that I'd never written back to disk it occurred to me that, wait, I'd swapped both drives around the same time...

Anyway, put the PATA->SD (and smaller CF) back in and everything was suddenly fine again. AAAARGH. So, yeah, that totally didn't help with the narrowing it down. ;)

A good reminder of how it's a bad idea to change multiple variables at once, I guess. :p
 
The connectors arrived yesterday, and while I have not inspected every single one.. they seem to be perfect. They have a snug fit right out of the package, which I am very pleased with.
20191204_221331974_iOS.jpg
20191204_221356469_iOS.jpg

The company I used is called Shenzhen RealRun Electronic Co. They typically have a MOQ of 1000, so I did end up paying a bit more per piece than normal for "only" ordering 250. They have a contact link on their Alibaba page here: https://realrun.en.alibaba.com/?spm=a2700.icbuShop.88.21.621118397jvQ2r

The part number is: "FH2.54*8.5-2*31P 0.64*0.4 Gold 1U PA6T"

Otherwise, if anyone wants some small quantities, feel free to PM me here or since some of you know me on facebook, you can contact me there.
 
Man, I didn't even mention the best part of the circus (here). By fantastically unlucky coincidence a few weeks ago I happened across a stash of old CF cards of various sizes; mostly 128MB-512MB, but there were two 2GB ones, and for some idiotic reason I thought it a good idea to swap these CFs into *both* machines. (In one case because, you know, bigger than 256MB, and the other to free up the PATA->SD adapter I'd initially been using.) Apparently those CFs have compatibility issues with XT-CF though; despite seemingly working fine at first both machines gradually started acting buggier and buggier. The most common symptom was they'd crash with an "Invalid command.com" error when exiting programs, but they'd also crash halfway through booting, I'd get strange random crashes or obvious corruption inside software... basically both machines started coincidentally going to hell around the time I'd swapped boards a few times. To say the least I was getting to the pulling-my-hair-out-with-frustration stage of W-T-F. I was convinced that maybe I'd started cracking the motherboard traces from the stress of pulling the boards in and out. Finally, around the time I started realizing there was actual file corruption going on even with files that I'd never written back to disk it occurred to me that, wait, I'd swapped both drives around the same time...

Anyway, put the PATA->SD (and smaller CF) back in and everything was suddenly fine again. AAAARGH. So, yeah, that totally didn't help with the narrowing it down. ;)

A good reminder of how it's a bad idea to change multiple variables at once, I guess. :p

Somebody needs to start a thread for which CF cards are known to work/not work, and whether with the ISA-CF or the XT-IDE (8-bit PIO/16-bit mux). As well as which adapters they've had good results with.

Within the collection that I have, I've also noticed that some cards are picky about master/slave configurations, which can also depend on which CF adapter I'm using, though all of them are fine as standalone devices. Some do not like to be a master when a slave device is present, but have no problem being a slave. Most people aren't going to run into this problem, though I should try to acquire a matching set of adapters just to make sure that's not what's causing the issue.
 
I tend to stick with smaller cards for my XT class computers. 32-256mb in size, and I find that compatibility with these cards is very high.

The only oddball are the Cisco 64mb cards which do not seem to boot on any of my Tandys. The boot fine on my NuXT, and an IBM PS2 model 35. What is weird is that I can read them just fine on my Tandys, just not boot from them. I think it may have something to do with the 2048 byte allocation size that they end up getting formatted with.

32mb cards are my go-to nowadays since they are "large" for an XT machine, and do not have stupid long free space calculation times. Every brand of 32mb card I have tried on my XT-CF-Lites have worked so far.
 
I tend to stick with smaller cards for my XT class computers. 32-256mb in size, and I find that compatibility with these cards is very high.

The only oddball are the Cisco 64mb cards which do not seem to boot on any of my Tandys. The boot fine on my NuXT, and an IBM PS2 model 35. What is weird is that I can read them just fine on my Tandys, just not boot from them. I think it may have something to do with the 2048 byte allocation size that they end up getting formatted with.

32mb cards are my go-to nowadays since they are "large" for an XT machine, and do not have stupid long free space calculation times. Every brand of 32mb card I have tried on my XT-CF-Lites have worked so far.

My CF cards are between 16-512 MB, but I couldn't get any of them to boot until I used Diskpart to "cleanall" on those cards in my Win10 machine. All said, you're right, there's no point in having anything larger on an XT class machine, even if you're using DOS 5 or 6, and can see 2GB partitions. However, sometimes you just use what you have lying around.
 
Even after something like an FDISK /MBR? That tends to do the trick for me.
 
The only oddball are the Cisco 64mb cards which do not seem to boot on any of my Tandys.

All of the CF cards I have, including those evil 2GBs, are Cisco labeled, but of course that doesn't really mean anything. I think I'll spend an hour at some point documenting what they identify as in the CF BIOS. The 128MB and 256MBs I haven't seen issues with are a different vendor from the 2GB ones; I think I got a 64MB in the last batch, I'll see if it's the same OEM as either of those or something else entirely.
 
The majority of my Cisco cards identify as "STI Flash" with a version number that follows. Based on that, I believe them to be rebranded Simpletech cards. There are a couple that identify as something else, but I do not recall off the top of my head what those were.
 
Back
Top