On Wednesday 02 January 2008 22:10, Gene Buckle wrote: With an appropriate current-limiting resistor I would assume? The little tester I have for rs232 stuff uses bi-color LEDs, makes it easier to see what's going on. Could be, I've certainly replaced my share of those. And put sockets in while I was at it. :-) Just make sure it's getting the proper power supplies, as if one or both of them are missing you won't see proper output then, either. Whenever you're trying to troubleshoot anything, always verify the power supplies first! I forget that myself from time to time, and have regretted doing so. -- Member of the toughest, meanest, deadliest, most unrelenting -- and ablest -- form of life in this section of space, ?a critter that can be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters" - Information is more dangerous than cannon to a society ruled by lies. --James M Dakin

On Wednesday 02 January 2008 22:50, Gene Buckle wrote: Why I salvage all sorts of parts... Well, there ya go then. Both plus and minus 12V? You do need both. If there's no change in the LED indicator, and the input that drives it is changing, then yeah, it's probably toast. -- Member of the toughest, meanest, deadliest, most unrelenting -- and ablest -- form of life in this section of space, ?a critter that can be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters" - Information is more dangerous than cannon to a society ruled by lies. --James M Dakin

Checked it. -0.60v There's a Elpac/TDC CB-3811 DC-DC converter on the board that may have died. I'll pull the board tomorrow and test right at the legs of the device. Thanks for questioning whether or not I'd checked the -12v supply! Finding a replacement is gonna be a stone b*tch I think. :( I may have to go to another DC-DC converter that is near the same size that I can wire in. g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind.

One comment on that. MAX232 requires 4 cap of 1.0 ?F (tantalum). The "A" version (MAX232A) will work fine with 4 caps of 0.1 ?F. - Henk.

On Thursday 03 January 2008 00:48, Gene Buckle wrote: Something might be loading it down, too. I'd check for that before assuming it's bad. -- Member of the toughest, meanest, deadliest, most unrelenting -- and ablest -- form of life in this section of space, ?a critter that can be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters" - Information is more dangerous than cannon to a society ruled by lies. --James M Dakin

On Thursday 03 January 2008 11:04, Mr Ian Primus wrote: I don't recall too many of those with that failure mode... Like what sort of stuff? The most I ever saw those fail was one company that had a wire running from one building to another and that used to pick up a lot of transients from lightning. They'd bring those terminals in fairly regularly. I tend to not rule stuff out completely, and avoid surprises. :-) -- Member of the toughest, meanest, deadliest, most unrelenting -- and ablest -- form of life in this section of space, ?a critter that can be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters" - Information is more dangerous than cannon to a society ruled by lies. --James M Dakin

On Thursday 03 January 2008 12:20, Mr Ian Primus wrote: I always thought those chips were rather more robust than that. I hate it when that happens... That sounds pretty seriously zapped, all right. I'd disable it (if you can) and plug something else in there mapped to that location. Stuff tends to want to surprise me anyway, too much of the time. -- Member of the toughest, meanest, deadliest, most unrelenting -- and ablest -- form of life in this section of space, ?a critter that can be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters" - Information is more dangerous than cannon to a society ruled by lies. --James M Dakin

Well as I mentioned earlier, I'm going to pull the converter and test it out by itself. The power supply may need repair as well as it's only putting out 4.65 volts under load. The Littleboard doesn't seem to mind, but it bothers me. :) g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind.

On Thursday 03 January 2008 20:43, Gene Buckle wrote: That's exactly how you do that. Just disconnect that output and see what's there. If it gets close to what you could call normal, then there's some other problem, like a shorted bypass cap or similar. I never used to see all that many problems with shorted caps, but discussions on other lists (tek scopes particularly) seem to indicate that it's getting to be a common problem these days. -- Member of the toughest, meanest, deadliest, most unrelenting -- and ablest -- form of life in this section of space, ?a critter that can be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters" - Information is more dangerous than cannon to a society ruled by lies. --James M Dakin

That's basically how you do it. Remove the converter, power it up and see if oyu get the right output. If you do, then it's worth trying a sensible load on it (A 1k resistor?) to make sure the output doesn't drop spectacularly. If the convertor works like that, but doesn't work when fitted to the PCB, then something is overloading it. -tony

It generates -12v from the +12v coming from the power supply. I don't know if it's potted or not. If it is indeed dead, I'll crack it open and find out. :) -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind.

The highest value I have in my "spare" stuff is 33Ohm. g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind.

OK so far [OK, on first reading I read 'DART' as 'LogicDart' -- HP's 'advnaced logic probe'. I now of course realiuse you mean the Z80 serial chip of that neam...] For the output to be correct, 4 things have to be right : 1) The 1488 gets the right input signal(s) 2) The 1488 gets power 3) THe 1488 is working correctly 4) The output must bot be shorted to something else You have checked (1) -- or have you. The 1488 contains 4 drivrs. One is an inverter ,the other 3 are NAND gates. _Often_ the inputs on the latter sections ae strapped together turning them into inverters too. But is that the caser here? Have you made sure there's not another signal being fed into the other input? Now as for (2), check there's +12V and -12V on the right pins of the 1488. And that the ground pin is, indeed, groudned. Remember a supply might be misisng because osmething is overloading it. Maybe a 1488 or a dcoupling capacitor has shorted. Nwo, if this is, indeed, the TxD line, the output shoculd be about -10V when the thing is idle. Check that with a meter. If it isn't, then either the 1488 is playing up, or the output is shorted to something. The voltage you get there might tell you what it could be shorted to. If in dout, remove the chip (or lift the output pin) and see if the votlage on the TxD trace on the PCB goes away -- is it sourced by the 1488 or not? If you get this far, I'd then take one of those bicolour LEDs with 2 wires. The have a red and green LED in inverse parallel inside. COnnect that in series with a suitable reisstor -- 3.3k would proaly do -- between the TxD output of the 1488 and ground. You should see one LED glowing when you turn the machine on./ See if you get flickers from the other one (you can often see the 2 dice inside the package, BTW) when the machine tries to transmit something. Having used that sort of tester on 9600 buad lines in the past, I will say that the flicker is clearly visible. At htis point I'd proably replave the 1488. Put the new on in a socket if there's space for it. I've had them fail in the past... -tony

3 of the 4 sections of the 1488 are NAND gates with 2 inputs. Those 2 inputs must both be drivein somehow (they should npt be left floating). Basically there are 3 things you can do : 1) Link then together, and to the logic signal (for example TxD from the Z80-DART), thus turning that section of the 1488 into a NOT gate with RS232 level outputs. This is the most common thing to do. 2) Connect the logic signal to one of the inputs, connect the other input to +5V, maybe through a resistor. 3) Drive the 2 inputs from separate logic signals. Maybe the TxD line and a separate 'force break' line. Or the main and back channnel signals in a modem. My point was that you'd checked that the TxD line from the Z80-DART had got to the 1488, but you didn't say what happened to the other input of that bit of the 1488, if indeed there is a second input (the TxD could be using the single-input part of the 1488, of course). If case (3) applies, you could have had a problem in the logic that drives the other input, for example. Right. That, of course, is somethign you need to correct. Maybe the converter has failded. But 0.6V sounds like a diode-drop to me, so it might be that a chip has failed as is now acting like a forward-biased diode from the -12V line to ground, pulling it down. -tony

I am hampered by the facts that (a) this machine is not at all common in the UK, so I've never seen one and (b) I don't have schematics. But I don;t believe that for an isntant. I do not believe that the pin that's convnetioanlly DTR is driven by a combination of the RTS and TxD signals. RTS on its own would be reasonable. Let me re-state what I am asking. The 1488 has 4 sections. These sections are independant of each other, sharing only pwoer connections. 3 of the secions are NAND gates with 2 inputs each, one is an inverter (NOT gate) with only 1 input How are thse sectiions used? In particualr, the section that drives the TxD ouput (pin 2 on the DB25 connector), does it have 1 or 2 inputs? If the latter, where do _both_ of the inputs go? Are they linked together, do they come from separate signals? Again, I don';t know wnough about the machine. Is the -12V line used by anything else apart from the RS232 drivers? How many 1488s are there on the board? You want to remove either the DC DC converter or all the loads on it. If the altter are just 1 or 2 1488s, it might be easier to remove those. If the -12V line is still missing with those off the board (and again assuming tyhat -12V line doesn't power anytthing else), you then want to remvoe any capacitors between that -12V line and ground (they can short too). If it's still missing, then the converter is bad, but I'd still test it after getting it off the board. If you remvoe the converter first, and it tests fine out-of-circuit, you then need to find which of the on-board compoentns that connect to the -12V line (1488s, decoupling capacitors between the -12V line and ground, anything else?) is shorted. aybe by re-fitting the covnerter and removing the components until the -12V line comes back. -tony

Update - I pulled the 1488 and sure enough, the -12V came back up. I stupidly picked up a 16 pin socket instead of a 14, so it'll be some time next week before I get the new part installed. Can someone recommend a place were I could find a 40 pin chip-clip? Jameco, Mouser and Digikey haven't got any - unless of course I'm calling it by the wrong name. :) tnx all for your help! g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind.

On Sun, 6 Jan 2008, Glen Slick wrote: That's exactly it! Thanks Glen! g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind.

Sounds like that 1488 is shorted, then. Of course it _could_ be that the DC-DC converter can't supply the rated load any more, but I doubt it. Err, clip 2 pins off the end of the socket? I've done that when I needed to get something working _now_ and I had the wrong socket. -tony