• Please review our updated Terms and Rules here

Tandy 1400 HD -- PSU Bad I think

S

southbird

Experienced Member
Joined
Sep 11, 2009
Messages
316
So I got a 1400HD in good cosmetic shape for next to nothing, with the obvious disclaimer of "Don't have the power brick, can't test." Incidentally, I DID have a power brick from a 1400HD that I had years back but have long lost.

But alas, powering up just illuminates the backlight (sometimes) and all three keyboard lock LEDs light up steady; that's it.

Now, years back, with the old 1400 HD I had, I tore it apart because of a bad HDD, and found that the PSU will put out voltages without being connected to anything. Naturally these are the expected normal voltages of +12V, +5V, and probably -12V (used by the modem.) I even used this to power an AT board as a silly experiment back then.

Anyway, with THIS one, I've tested all the pin connections coming off the PSU, and about half are "dead", and the ones that work test out +5V. So it appears to me that the 12V side is the "dead" part. This was further hinted because the PSU has a 12V cooling fan on it which was not turning, clearly unpowered. I'm not an electrical engineer and am not too familiar with circuits more complicated than "Battery-Switch-Lightbulb", but I can use a DMM and traced 12V at least as far as the battery connector and a couple regulators.

Of note, CLEARLY somebody has messed with it before. There were two screws missing from the case and fuse "F1" has been directly soldered across on the underside!

Could this be the classic capacitor leak situation? Does that make sense as a possibility? Some of them do look a little sour on their undersides. I ordered myself a set of assorted electrolytic capacitors with ranges of parts I have seen on the PSU, as well as a capacitor tester (unless you guys just recommend I replace them all anyway while I'm poking around.)

And this is probably a long shot, but does anyone happen to have a pinout of what the PSU pins voltages SHOULD be? Or would be willing to open a 1400 to test them out? Of all portables I've ever opened, this one definitely is not difficult at all (just the screws on the bottom, and you can lift the case), so if you have a 1400 and are willing, it would be much appreciated...
 
Actually I found the service manual at http://www.oldskool.org/guides/tvdog/documents.html (14ltserv.zip) so I'll have a look at that this weekend perhaps... even specifically has a flowchart that breaks +5 from +12/-12/+15 and tells me a part to check

EDIT: Well, never mind, not that anyone's probably paying attention anyway, but I guess a 1400 LT PSU is different enough to tell me nothing about the 1400 HD's. Even the motherboard apparently is different. Oh well.
 
Last edited:
That must be right as I have a 1400FD which has a similar PSU to the LT I think (there is no fan in it). The problem on mine is that the (soldered) fuse blows if the battery and external power are connected, but it will run with either on its own.

Out of interest, what hard disk controller is in the HD? The FD has a 'modem slot' which I think is an ISA slot with the pins muddled up, so I was pondering on trying to get XT-IDE in it.

Cheers
 
pearce_jj said:
That must be right as I have a 1400FD which has a similar PSU to the LT I think (there is no fan in it). The problem on mine is that the (soldered) fuse blows if the battery and external power are connected, but it will run with either on its own.
Click to expand...

I can't speak for the FD model, but I do have the fan and alternate PSU. I started replacing leaking eltrolytic capacitors last night in a hope that might solve it, but I don't really know. I'm honestly wondering if it's the big transformer on the board. The LT's PSU diagram did note a transformer which produced the 12V, -12V, -15V voltages by stepping up the 5V rail (with very limited current.) Of course a 1400HD has a hard drive to worry about, so it might be done differently. In any case, prior to my cap replacement game, it did seem there was 5V on one side of the transformer and nothing on the other side... but given my limited electronics understanding, I'm not really sure if that's what was going on or not. I guess I'd also think it odd that the entire transformer would just be "dead"...

Of course, I don't REALLY care about the PSU itself. If I had the voltage pinouts (from an appropriate service manual for example), I could just replace it with anything else...


Out of interest, what hard disk controller is in the HD? The FD has a 'modem slot' which I think is an ISA slot with the pins muddled up, so I was pondering on trying to get XT-IDE in it.
Click to expand...

Well, the MODEM slot is pretty much just a serial connection. But if the FD has an expansion slot, that's really more of a muddled ISA. Check this: http://www.old-computers.com/download/Tandy/1400/1400FD-compiled-info.txt

I haven't seen the HDD controller in the 1400HD yet this time around, but from memory, I remember it as some proprietary connector board, and the interface is some kind of MFM (?) on a very narrow cable. But through the expansion port or the hard drive controller port (whichever you feel most adventurous about), you could probably wedge any ISA compatible part onto it. Now, of course, space is the real problem, but adapting an XT-IDE into the narrow space and using a pin header to connect internally would be a really neat project. You could probably start with a stock, known-working XT-IDE and build a pin header to ISA slot converter to make sure the card would even work.

If I get this sucker working I'll probably do the converter anyway to see what I might wedge out of this...
 
pearce_jj, you wouldn't happen to actually access or look at your PSU? I'm totally at a loss as to what could be wrong with mine (lack of electronics knowledge hurts.) Ultimately the wires all form into a single 15-pin motherboard connector. If you have something else then it's not mine...

From the LT Service manual, it's suggested that the transformer is what produces the +12 / -12 / -15V rails, so there's a chance that it is exactly what's wrong. But it's a black plastic monolith with no identifying marks, so I wouldn't even know where to begin thinking about it. I also can't absolutely verify this is the case with the HD's PSU, but I figure it's "most likely" the case, especially since +5V exists entirely on one side of the transformer and zip on the other. Unfortunately again, there's no marks that tell me which pins would have produced what voltage.

All connectors off the PSU merge into a single motherboard connector (except for one power line that I believe goes to the floppy drive) that has 15 pins, what I know are the following pins --

1 (White) UNKNOWN
2 (Dark Blue) UNKNOWN
3 (Brown) UNKNOWN
4 (Gray) +5
5 (Blue) +5
6 (Red) +12 * Transformer Upper Right (bottom view)
7 (Purple) -12 * Transformer Lower Right (bottom view)
8 (Orange) UNKNOWN Transformer Mid Right (bottom view)
9 (Black) GND
10 (Black) GND
11 (Black) GND
12 (Black) Ext Black **
13 N/C
14 (Green) Ext Green **
15 (Pink) +5

Ext Yellow +5 (probable floppy drive direct power)

* I traced these to the modem connector
** The Ext Black/Green do not connect to the PSU, just branch off the connector; I expect this is delayed power to the HDD.


So there's four pins which I have no idea what their voltage is supposed to be. Three of them come off the "dead" transformer (voltage goes in, none comes out.) 6-8 I've traced back to the transformer. 1-3 I just have no idea, they are top-side wires and hard to trace because they flow into and under tightly packed components. They might even not be voltage lines at all, perhaps some kind of feedback, who knows.
 
Last edited:
Thanks for the info on the slots. I will check that out.

I can check the pinouts no problem - will post up once I've had chance (this week hopefully!). Annoyingly I already did this..., but the envelope in question has got lost in the sea of mess that is my workshop :(

After years of storage my FD was running but with no screen backlight. This was solved by replacing all the PSU caps. The soldered NiCd CMOS battery was also dead but that didn't stop it from running.

My PSU has two outputs I think, the multi-pin connector and a smaller plug connected to a seperate board (modem perhaps). 5V is maintained even when system is off (to charge the CMOS battery maybe). If you have a main 4.8v battery, disconnect it as a dead one caused the relay in mine to buzz which I think lead to my charging fuse problem.

Anyway, will post back.
 
Right I've made a start on documenting the 1400FD, actually only the pinouts so far:

http://vintage-blog.peacon.co.uk/wiki/Tandy_1400FD

Orange (-22V) a problem for me with the failed caps. It runs the screen backlight, there is another transformer next to the screen connector which presumably creates some high voltage to run the backlight.

HTH
 
pearce_jj said:
Right I've made a start on documenting the 1400FD, actually only the pinouts so far:

http://vintage-blog.peacon.co.uk/wiki/Tandy_1400FD

Orange (-22V) a problem for me with the failed caps. It runs the screen backlight, there is another transformer next to the screen connector which presumably creates some high voltage to run the backlight.

HTH
Click to expand...

You are awesome! Thanks for the additional pins. I'm 99% convinced that what's failed is the transformer, though I'm not sure how/why. (Other than the generic "it burned up at some point".) I mean, power goes in one end, and zip on the other. I replaced a few bad looking caps, but none of them are the worst I've ever seen (and it made no difference; still regulates +5V and nothing else.) I was wondering (and fearing) what the OTHER transformer on the mobo ... backlight drive makes sense. I can say that even in its reduced state, the machine was driving the backlight. And honestly, all that might be stopping the POST is the fact that I'm missing the +5V voltages on the white and brown lines, whatever they power. I could even bet one of those is supposed to be the "POWER GOOD" signal which would totally explain why it looks like it's in a perpetual reset. The lack of the large negative voltage (which I was assuming was -15, based on the LT specs, but yours apparently is -22!) likely will mean I have no contrast on the LCD. But with any luck, jumping the missing +5Vs and building a custom regulator to replace the dead transformer might even render a working machine... Heck, I might even just replace the whole thing with a PicoPSU and build just the -22V backend.
 
I've not looked at the PSU board particularly closely, but there must also be a modulator to provide AC to the transformer to produce the -15/22, -12, -5V and +12V outputs... perhaps that is dead (and maybe because of a dead capacitor?).

There is presumable a power good pin but also there might be some mechanism to support the low battery light, unless it's just measuing the +12V for example.
 
Okay well I jury rigged a modern PSU into it matching your pinouts. The machine behaves the same (apparent perpetual reset, LEDs stuck on) and the only "improvement" is that the HDD spun up and the cooling fan kicked on. I guess I don't know what else to do with it at this point.
 
Not that this will be any help at all, other than to commiserate, but I have an 1400HD that behaves the same as southbird's device. Backlight comes on (sometimes) and all three light on the keyboard. I have dissassembled mine (just the power supply is out now) and I have come to the same conclusions, no obvious damage (caps, etc), it would be nice to know the other pin-outs, and if I could get those, perhaps a pico psu would work. There hasn't been any new info here in sometime, so I thought I would ask if there was any new information and go from there...
 
Sorry I'd missed the prior update on this one too. I've not looked at it since though, other than to tidy up my documentation on it, the PSU stuff on it's own page now: http://vintage-blog.peacon.co.uk/wiki/Tandy_1400FD_PSUs

Note that in mine, none of the caps looked bad. I would start by cutting out the CMOS battery, disconnecting the main NiCd battery and testing the main AC transformer, which should be giving a good smooth 12V regulated output.
 
I have disconnected the NiCd (I really thought that I had found an issue there, because the plastic that covers the two battery wires looked like it was blackened, as if from a short, but the wires were intact, so apparently not). Anyway, the Power supply is great and rock steady. That leads me to the PSU. CHecking this tonight, I have a blown fuse (F1) near the power input - when I jumper across it, I get a click at the relay and power on several of the pins (6V only no 12 or 22). It is late so I will test more later. Need to replace the fuse F1 though (PS-F1:10953750:RESISTOR,FUSE 3.15A 125V:NE0055 from the 1400FD Parts list)...I don't know if there is an issue with the other one (F2) by the power switch...check that tomorrow.
 
Last edited:
I did a more thorough examination of the board after fixing the fuse issue (only the one was bad). I tested with both the power switch on and off and looking at which connector pins were bad, I was able to get a feel for which traces on the board led to a possible issue. I had an issue with C66 leaking badly (I did not catch it before, but the caps were bulging out the bottom which was harder to catch) and C63 (probably just corroded from C66) and C31(leaking and expanded) and C4(slightly expanded). They are all Rubycon branded (except C63) and are the smaller ones 47 & 22 μF. I hope to see some improvement when they are replaced...
 
I have replaced those and a few others that looked suspect - replacing caps is a little like eating potato chips, once you get going, it is hard to stop. Only two tested out of spec, and I have one more wire in the main connector that works now. None of the 12v ones work, so I removed the transformer (actually not too difficult) and tested at the input pins with the transformer out. Mine has two markings on the case E 14 (on the top) 01 (underlined) with a little pi looking character next to it on the side away from the #1pin (and the power switch) and on the board (under the transformer) is 'ET-24AY'. I will have to take some time and transcribe my results, but pins 1,3,4,5 were all powered at 6v (on the board). Numbering with 6 directly across from 1 (I don't know if that is the correct way or not), but 7 and 10 tested continuous with ground. I'll pass along more later.
 
Well it's progress anyway, getting another line up. If you figure out the charging circuit please do share, as mine blows F1 instantly if the (now NiMH) battery and external 12V are connected together. Works OK from either in isolation though, other than always showing low-battery even if it's fully charged.
 
It has been a while, but there was not much to report, I was stuck with a power supply that I could not finish due to the inability to find a replacement for the transformer (still looking by the way). I purchased another 1400HD (I will refer to it as 1400HD-2) - it was severely damaged but the power supply lit the back light and it did not exhibit the 'all indicator lights on' issue, the fan worked, but it still did not appear to 'post'. It had several leaking capacitors, which I replaced and then I installed the 'better' power supply into 1400HD-1. The laptop exhibited the same 'all lights on issue', the fan did not work etc. So now I question whether there is an issue with the original 1400HD-1 power supply, since the 'better' power supply exhibits the same symptoms. I still owe some additional documentation, so I will work on that, before digging in further.
 
No, it doesn't, but it does do something - not sure what, but the lights come on and go out, then the screen comes on and after a while the hard drive spins up, never anything to the screen though. Anyway, having two of them has allowed me to do a little experimentation and I've found something interesting. I took the board with the yellow/black and green/black wires (under the HD) and removed it from the No.1 box (it boots to the 3 lights on) and that behavior moved to the No.2 PC along with the board. So now more experimentation is in order.
 
You must log in or register to reply here.
Back
Top