I'm also a long time Weller user. I bought my first WTCP station in 1974, used it for over 25 years, then gave it to my brother, who's still using it now... but I'm pretty sure he's replaced the heating element and the magnastat temp sensor once. He's also gone thru a handfull of soldering tips.

I recently upgraded my soldering equipment to Weller's WXP line. A WX2 (2-channel supply) and their 65-watt iron and their pico handle for SMT work. I doubt I'll ever replace them, as they will likely be running past my personal end date

Some equipment does fail... it happens. It doesn't have anything to do with using it.

_________________
Regards, KM
https://github.com/floobydust

Alright so the new iron arrived and it still has the same issue. So apparently it's the station that's misbehaving.

It's weird though, when no iron is inserted it seems to work normally but as soon as either the old or new one is plugged in the display starts to flicker a bit, which gets stronger and stronger until it just gets stuck on the screen I posted.
It seems like it's constantly resetting the internal microcontroller. But why? Sole broken cap or over current protection?

I'll probably just order a replacement soldering station. I can then try to repair the old station and if that works I will keep it for myself so I have an iron at home and the money won't be wasted.
(Unless it's a really bad idea to try and DIY repair soldering stations?)

But it still makes me wonder what killed it... it was the first time I ever set the station to 400C, the iron says it supports up to 450C...
Maybe some unluckily large current spike when I set the temp and held it to the solder blob somehow destroyed some small internal current limiting circuit?

EDIT: one of my colleagues had a better idea, I'll just tell my boss that it broke while doing actual work instead of private stuff. Honestly I kinda feel dumb for not thinking of that myself, guess I was panicking too much.

Those display problems remind me of a similar issue I recently had with a 1980s Casio calculator. Turned out to be dry solder joints. I just retouched some of the main joints and it’s back to working like new.

_________________
I like it when things smoke.
Blog – Zolatron 64 project

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

Alright, finally some good news for once!


While the replacement soldering station didn't arrive yet, i did remember that we have some older soldering stations in storage.
And i was fully ready to use a GDR-era looking soldering station to do this, but i was surprised to find the previous good soldering station that we had:

The now broken station had issues melting all the solder blobs at 400C, but this one was able to handle it easily at ~275C.
Maybe this station is just more powerful, or was it a sign of the other one being on the brink of death? who knows.

Either way i had no issues with the vertical flow technique as Garth described it. (i used more solder and an ungodly amount of flux)
I didn't notice any solder bridges between pins and all of the pins seem to be connected to their respective pads. i'll attach some images. (I know U6 is crooked, but i only noticed after having soldered the right side)
Also i used a completely new board for this, i consider the other one a lost cause. maybe i can harvest the RAM IC and FT240X from it, but for now i want to continue this board with new parts.
Tomorrow i can finish the rest since it's just the FT240X and the other TH Parts.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

I didn't get done today because i was missing some parts (the 32-pin PLCC socket 8 pin DIP socket, and 3.3k Resistors).
i also ran out of bypass caps (100nF) and noticed that i don't have any 47pF caps for the USB Port. i can desolder both of these types from my current 65C02 SBC. or i could use 22pF caps that i have lying around for the USB Port (as i don't think it would be that picky, i'll first try to use the USB Port without any caps as a test), and i have some 150nF caps that i could use for bypass instead of 100nF (should be close enough)

I have a 6502 SBC implemented as 100mmX100mm 2-layer PCB. I intentionally built it without any bypass cap. It worked just fine at 14.7Mhz. I did that to check my design margin, your mileage may vary.
Bill

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

Ok, it's (maybe) finished. i salvaged the 47pF Caps from my old 65C02 SBC (ironically USB on that SBC still works perfectly fine without them), there are still 2 bypass caps missing but it should be fine for now.
first of all, an image of the whole board:

There is already a botch wire, between the 2 SRAM ICs, because U6 was so tilted that the top left pin wasn't making contact at all no matter how much solder i threw at it, so i just added a small bridge to the IC next to it, since it goes to the same connection. i also checked it with a multimeter to make sure the bridge wasn't making contact with the pins next to it.

So, power on. the LM340T does get really really hot with all ICs in place. i only have small heat sink i can put on it for now, but i'll order some actual TO-220 ones for it later. (any tips for cooling? active/passive? is a voltage closer to 5V better for avoiding heat? i did use a 12V DC Adpater but i have an adjustable one that i can set to 7.5V or 6V for example)
Checking the clock circuit it appears to be working right. the slowest DIP-8 clock i have is 1.8MHz, which generates a ~900kHz GCLK and PHI1/PHI2.
but that's basically where the good things end, the CPU seems to be stuck in a reset loop for random amounts of time. i checked VPB with my scope and it's constantly going until it randomly decides to stop.
welp, turns out i left the BE pin floating... well frick, i'm gonna have to connect it to 5V via a resistor tomorrow.

another thing that doesn't work, USB. my PC sees that there is something plugged in but cannot figure out what, so basically it cannot communicate with the FT240X.
it is being powered correctly, i also checked the 3.3V and 1.8V outputs and they are correct as well (meaning the IC isn't fully dead). i compared the traces with my old 65C02 SBC and the USB connections are also right, the 27 Ohm Resistors have the correct (measured) value, same with the capacitors.
well well well, like with the CPU i think i found the issue while writing this post.

the datasheet for the FT240X says that RESET can be left unconnected when not needed, but the same datasheet also shows it being pulled high to 3.3V in the examples. luckly the pin is right next to a 3.3V powered pin so using a small wire i hand-bridged the pins and powered on the board... and tada it appears in device manager like nothing bad happend 5 minutes ago.
i stopped bridging the wires and power cycled it (fully unplugged for like 10s), and after power up it still works????? apparently this was just a 1 time issue where it wasn't able to properly start up itself or something because RESET was "floating", and now that it finished it's inital power up it just works everytime now?
I think i'll still solder a small bridge between the pins to always have RESET pulled up, as i don't trust this thing fixing itself like that...
but atleast it's a massive relief that the IC itself is not broken, so something good i guess?

i'll report back with more tomorrow as without any CPU signals i cannot check the rest of the system. I'm hoping that by the weekend i'll have a functional SBC so i can start writing the ROM code

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

i might want to use it in the future, so just to be on the safe site i'm gonna use a resistor.



from my experince that is not a good idea, because my 65C02 SBC v1 had it done like that and it was very annoying.
Because resesting the FT240X disconnects it from the Host which means that after the reset signal goes high again you have to wait a few extra seconds for the chip to reconnect with the host device (ie the PC) before you can use it again with whatever terminal software you have.
so i'm perfectly fine with just having it's reset pulled high and manually flush it's internal buffer after the CPU resets.


They are tactile, but bonestly they are just cheap buttons i found on amazon by searching "breadboard push buttons" and were like a few bucks for a 100pcs pack.
but i did some searching on mouser and found a lot that look identical to the ones i have: https://www.mouser.com/ProductDetail/TS ... vxdQ%3D%3D



i'll see tomorrow if 7.5V is enough, and if not i can use 9V instead. it should still produce less heat than 12V like you mentioned.

ok, update time!

after doing the botches i was able to get some basic code running!
the test program just writes a "!" to the FT240X and constantly flips output value of Port A on the VIA. both work so that means the decoding logic for ROM and IO is working perfectly! the boot up code is in Bank 0, while the actual test code is in Bank 1, so the Bank latching also seems to work fine.

i have the NMI ISR print a single "?" to the FT240X and pressing the NMI button makes it appear exactly once each time. but when i run the system at 25MHz (ie 50MHz Oscillator) it crashes when trying to return (as it prints the "?" and then stops)
hmm, Both IO and ROM stretch the clock by 1 cycle so while accessing those the CPU is pretty much running at ~12.5MHz, while accessing RAM it keeps it's full 25MHz. so something must be going wrong.
maybe the CPU just gets unstable at 25MHz, or the decoding logic just takes a hair too long, causing it to read garbage data. it's really hard to tell.
i also tried it with 20MHz (ie 40MHz Oscillator) and it works perfectly fine, so i'll take that as the current speed limit for now. (which is still great tbh, a 20MHz 65C816 is basically a SuperCPU!)

but this will defineitly make it harder to build a VGA card for this without Dual Port RAM... i do have the WSE pin on the expansion port, so i can tell the CPU to chill for a cycle but i don't know if that will be enough to build a semi-asynchronous interface to the Video RAM.
but that's an issue for the (not so far) future. right now i want to write the serial bootloader for the ROM so i can test programs without having to touch the ROM again.

on another note, 7.5V seems to be enough for it to have a stable 5V output. the heat it generates is also not as bad. i had it running for like 30min and it's just about too hot to touch, while before it would be too hot after just a few minutes.

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!