[C64] RS232 userport adapter works but gets hot

[Divarin]

I've had this device sitting around for years and have rarely used it. The reason is that it gets really hot, well U1 (one of the MAX232CPE chips) gets hot. Like hot enough that you can smell it burning and if you touch it you burn yourself.
U2 (which is the same chip) is fine.
I tried swapping the two chips to see if the heat moved (bad chip?) but it didn't
There's no shorts that I can detect, voltage into the chips is fine (just shy of 5v).

I tried to find any info about this (url is printed on the PCB: www.museo-computer.it) but even with the wayback machine I can't find any info. Schematics was what I was going for but so far can't even find evidence this thing ever existed (maybe from a parallel universe?)

I did see this text-only schematic of an RS232 interface design (http://www.zimmers.net/anonftp/pub/cbm/documents/projects/rs232/index.html) (at bottom) but this looks different from what I have.

Has anyone here used one of these or something similar?

 

[ClassicHasClass]

If the "heat" doesn't change by switching the chips, it can't be the chips and it must be something about the board or wiring. Rather than risk a fire, there are many other user port adapters you could use and none of them are very expensive. I use a GGLabs myself for my 128DCR which supports UP9600, has a USB port and connects directly to my Linux workstation.

 

[Chuck(G)]

I've used the MAX232 chips for years and they don't get hot. Basically charge-pump devices with the interface logic. Check the caps around the chips if swapping the chips still causes one to get hot. A short on an output is also a possibility.

 

[NeXT]

I ran into an issue with a 20ma current loop adapter and the MAX232 years ago which I designed that came down to because the MAX232 has multiple driver lines and we are only using one set for transmit and receive, in certain conditions unless you deliberately pull the unused lines low they can go into CMOS latchup.

 

[Chuck(G)]

Sure, but that's basic "Design for Dummies". Can't believe that the board designer didn't think of that; but maybe not.

 

[Divarin]

Yeah I'm basically using this broken piece of tech as an excuse to practice and improve my own skills, I have other means of connecting my Commodore so for me the end goal isn't so much having a working rs232 adapter as I could just go buy another one it's to get this one in particular working just to see if I can.

What I've done so far is swap the chips, check for shorts between any two pins on the chip, check that the voltage was correct, and did a real quick & dirty check for shorted caps (just checked continuity). I guess next would be to desolder the caps and check each one properly.

It is possible this board has a design flaw.

 

[syzygy]

I would not jump to any conclusions about a faulty design without evidence. It is simple enough for you to trace out and test
everything (including the connectors) if you really wanted to get into doing that.

The designer of your board is on twatterX - at least he was 5 years ago. You could probably write him and ask if he could get you the schematic - I mean if you really wanted to find out what is going on. He also made a V2 that appears to be available here.

For the sake of nostalgia, here is the one (a 1982 design) that I used in the distant pass (these were common as mud in the C-64 heyday). It was sold on the bay long ago, but I have to do something with the pictures

 

[Divarin]

interesting well I don't think I want to make a twitter account just to reach out to him I'd rather just stay out of that whole universe.

So nothing wrong with the caps as far as I can tell. One interesting bit I tried powering the device with a bench power supply (so it's not in a c64, just sitting on the bench) this was easy to do as j5 (unpopulated) provides access to the 5v and ground lines. Doing this, just power alone, no heat.