Strange issue with floppy board

38 posts / 0 new
Last post
Offline
Last seen: 5 days 8 hours ago
Joined: Apr 27 2022 - 10:32
Posts: 15
Strange issue with floppy board

 

i have a new problem with Apple II europlus, a few days ago I have been having video interference problems with the floppy card inserted. I only have the floppy card inserted, I also tried with a super serial card but it does the same (maybe less).

 

has anyone had a similar problem? maybe some filter component on the motherboard bus?

 

 

Offline
Last seen: 5 hours 24 min ago
Joined: Jun 6 2020 - 10:50
Posts: 254
Actually, someone just

Actually, someone just started a thread within the last day or two with similar symptoms. He narrowed it down to a noisy PSU.

 

https://www.applefritter.com/content/disk-card-apple-ii-europlus

Offline
Last seen: 5 days 8 hours ago
Joined: Apr 27 2022 - 10:32
Posts: 15
now I'm going to read the

now I'm going to read the thread but the power supply I recovered entirely by changing all the capacitors both on the primary and on the secondary output

Offline
Last seen: 5 days 8 hours ago
Joined: Apr 27 2022 - 10:32
Posts: 15
nothing, I try to swap the

nothing, I try to swap the power supply of the IIe and the situation is perfectly identical

Offline
Last seen: 1 day 33 min ago
Joined: Apr 17 2022 - 12:26
Posts: 49
We indeed have the exact same

We indeed have the exact same problem

 

I can also confirm it is not the PSU noise.

 

I also believe this problem kills all the cards, except for some reaosn the language card (that maybe is special because of slot 0)

 

I think this is a problem with the clock signals, especially in the vincinity of B1/B2 ( 74S175) some curves have a funky shape to me and just don't look right.

Also swapping my 74S175 with another (working) one makes the apple not boot at all, which is super bizarre, and would let me believe some signal entering the 74S175 is not like it should be

Sad thing is this 74S175 is in the middle of a very complex circuitry

 

As some member said, this is in the end some problem with the DEVICE SELECT line, which are controlle by 74LS138 in H2/H12 (which work in my case) so I guess it's either the logic or the clock coming in to those that is out of specs.

 

I don't know how these clock signals looked like back in 1979 but today they are all very ugly and dirty, it's a miracle this thing is able to boot with clock signals so full of noise, bounces, echoes, etc. Which make debugging difficult because for someone with today's eyes, everything is a bit "edgy" on the oscilloscope...

 

I'm still searching. 

 

MacFly's picture
Offline
Last seen: 1 day 16 hours ago
Joined: Nov 7 2019 - 13:49
Posts: 177
Sounds like a fun session,

Sounds like a fun session, slowly narrowing down the root cause... ;-)

Looking at the images above: they clearly show a pattern, looks a bit like columns 4 + 6 of each character are affected - but never the others. Also, there is a pattern concerning the lines - looks like an issue involving a counter or address line. It's slightly shifting with each line - but that's not unusual, since the lines of the screen are not related to a power of 2 - so when there is an issue with an address line, or binary counter, then the issue will slightly shift with each screen line. It's not impossible that something like this could be caused by a weak PSU. But such distinct, fixed patterns are more likely caused by logic issues.

Anyway, as suggested before: pull the ROM from the disk II card. When this is not present, then the DEVICE SELECT line of the slot must not show any activity at all (must be fixed at HIGH logic level). In my earlier post I said, there could be brief activity during booting - that's not right. The ROM has it's own slot select signal (IOSEL), that's the one which would show brief activity during booting (when the CPU probes the slot). Without the ROM the DEVICE SELECT must be fixed at HIGH level all the time, since the card's registers are never accessed. If you still see any activity at all - then you have an issue with the address decoding logic somewhere.

My guess is, that you will still see regular activity on DEVICE SELECT. That's the logical explanation why plugging a good 74LS374 causes the system to show issues - while the issue is completely gone without the latch. The latch would incorrectly drive the bus - and hence "fight" with another latch or bus driver in the system. Btw, that could also cause some noise (maybe even on the power rails): when two logic ICs are fighting over the bus, they cause very brief moments of a short circuit (potentially on all 8 data lines). It'd be interesting to compare how the power-rails look, once you fixed the root cause of the logic issue...

Offline
Last seen: 1 day 33 min ago
Joined: Apr 17 2022 - 12:26
Posts: 49
V ery interesting point about

V ery interesting point about the noise indeed...

Since I got my II europlus i've been fighting with clock issues.

I had to replace two chips (detected as bad by a TL866) for it to start working, and even so there is something fishy because it only accepts to boot with ONE 74S175 out of my bunch of 5 that all register as OK in the TL866

So clearly this particular one is more "tolerant" to some weirdness of this entire system.

 

My main problem is making sense out of the clock logic, again this is pure "80s retro" in the sense the engineer (is that Woz ?) is using 100% of each chip and each little logic gate, which creates convoluted schematics that are hard to follow (or my brain is getting lazy). Nobody works like this anymore since the mid-90s , nowadays you would program a chip to do all that and don't care about the unused logic gates.

This is really a mind stretch to read those 80s schematics for a "modern" (and I'm not that young) electornics engineer. But it's part of the challenge.

 

 

The fact that two completely unrelated Europlus have the exact same issue makes me think thisis clearly a problem on the motherboard with one of the chips (I mean, there is no way we would be accident have the same broken trace, or the exact same cold joint on the same chip.)

However ALL TTLs on my Eurplus register as OK on my TL866 so either this thing should be taken with a grain of salt (possible) or it's one of the non TTLS that is broken (and I can"t test them)

 

Further reading on this ; It seems the TL866 only tests logic states (eg put this in the inputs, and check the outputs) - if I read this correctly it means does not fully test counters (let's say the test is very simple for counters) so in theory it is possible a counter is marked OK by the tester, but yet it's broken. I could maybe try to replace some of the couters involved in the clock logic (but I don't have spares so that'll be weeks of waiting again)... 

 

Current state of my europlus is ; If I dont use the expansion bus at all, I'm able to boot, load programs from audio input, write basic programs. I left it running for hours and it seems somewhat stable. (I do feel it's super sensitive to electrostatic and/or 50Hz but maybe that's because it is lying on my bench.) For me the power rail is super noisy, and it's NOT the PSU (i've recapped it for nothing, as I suspected - i'll post about that later.)

 

So all in all, what's NOT working is specific to the expansion bus and produces very high freqyency (>Megahertz) noise on the 5V rail, even when no cards are present but much owrse if some are present, making the system a bit unstable.

 

Offline
Last seen: 1 day 33 min ago
Joined: Apr 17 2022 - 12:26
Posts: 49
FYI here is the same problem,

FYI here is the same problem, yet again, but at VFCED (no solution found over there)

https://forum.vcfed.org/index.php?threads/apple-ii-plus-that-runs-passes-diags-but-has-thin-vertical-lines.76901/

 

 

MacFly's picture
Offline
Last seen: 1 day 16 hours ago
Joined: Nov 7 2019 - 13:49
Posts: 177
TL866

The TL866 isn't bad. If it detects a fault or doesn't recognize a chip, then the chip is indeed bad (or unsupported by the TL866 :-) ). Any IC which is completely dead will fail the TL866's test.

However, this device only makes some simple, "static tests" to verify the IC is ok (it will slowly set input values and then slowly check all outputs). It's unable to detect when an output is weak (unable to drive a signal with the specified maximum current). It's also unable to detect faults related to very quick logic changes, i.e. if a device really produces the correct output after X nanoseconds. When 40 year old semiconductors degrade, they are not always dead as a brick. Sometimes they are just weak or slow. The TL866 isn't able to pick this up. So, testing all components with the TL866 sometimes won't show all the issues.

 

Offline
Last seen: 1 day 33 min ago
Joined: Apr 17 2022 - 12:26
Posts: 49
Well an update, I've just

Well an update,

 

I've just tested all that MacFly suggested and it seems we're hitting a dead end

All signal behave exactly as you expected :-( 

 

- nothing in the slots : DEVSLCT high all the time on all slots

- card with no ROM : lines on screen (!), but DEV SLCT stays high

- card with ROM ; DEV sLct shortly pulled down at regular interval ( 2Mhz ? don't quote me on this)

 

I've also tried inverting the LS138 (io select and dev select) result is the same.

 

The 74S195 (C2) is suspicioucly hot on my system.

Again all things point to the freaking clock but I can't make sense out of it and I don't have enough spares to blindly replace ships (that all test ok in TL866)

Offline
Last seen: 4 hours 24 min ago
Joined: Apr 26 2016 - 08:36
Posts: 360
What do you suspect is wrong

What do you suspect is wrong with the clock?

Clock issues should be simple to diagnose with an oscillosope, checking the output of the primary clock at the 74S86 chip's output pins, as well as the video 7 MHz signal (or the equivalent PAL motherboard frequency) using an oscilloscope to check for proper frequency, amplitude and shape of the (theoretical) squarewaves.

 

Offline
Last seen: 4 hours 6 min ago
Joined: Jul 5 2018 - 09:44
Posts: 955
Have you tried a different

Have you tried a different controller card?  I'm wondering if one of the capacitors on your Disk ][ Controller Card is leaking interference when it is running?  I'd suspect C2 as it is an electrolytic type which tend to go bad over time, but it could be one of the small ceramic/poly ones I suppose (C3 or C6 maybe since they're nearest the power pins on the Apple II bus connector).

 

I'd also try a different (shielded) RCA cord for the monitor and make sure the floppy drive ribbon cable is properly grounded and routed as far away from the monitor or video connector as possible.

 

Does this machine have a PAL or SECAM color card in slot 7?  If so maybe that is picking up interference from somewhere?

 

 

Offline
Last seen: 5 days 8 hours ago
Joined: Apr 27 2022 - 10:32
Posts: 15
softwarejanitor wrote:Have
softwarejanitor wrote:

Have you tried a different controller card?  I'm wondering if one of the capacitors on your Disk ][ Controller Card is leaking interference when it is running?  I'd suspect C2 as it is an electrolytic type which tend to go bad over time, but it could be one of the small ceramic/poly ones I suppose (C3 or C6 maybe since they're nearest the power pins on the Apple II bus

I think at this point we have the exact same problem

the stripes are also with other cards such as the super serial card (although they are slightly less visible)

it seems to be the load on the expansion buses at this point because without cards it looks perfect

I have tried many cables but nothing changes

 

 

 

 

Offline
Last seen: 1 day 33 min ago
Joined: Apr 17 2022 - 12:26
Posts: 49
If I'm right AppleWoz : you

If I'm right AppleWoz : you did swap to PSU with the PSU of an Apple IIe and got the sam result right ?

This would rule out the psu as the cause

 

On my side I fully recapped my PSU (to no end) as it had a lot of noise.

Truth is : the noise is created (well, its level is doubled) when I insert the diskII card.

 

On my side, I can confirm (as you do) this problem happens with other cards, creating the exaxct same pattern

It also happens with just the Disk2 card (and no floppy connected)

And more than that, it does NOT happen if you remove the 74LS234 out of the diskII card (see my thread with MacFly)

 

So that would rule out the "load" poblem to me as well  ; it's unlikely just the load of that  single chip would trigger a difference in current cunsumptio,

 

One important question for you Applewoz ; despite of the lines ; does your floppy work ?

Because mine does not. It spins, itn looks like it's working - but it's not working if I try to boot a floppy or try to write/format a disk with INSTA-DISK (from audio in) or such. As I have two floppies, none of them work.

So more than lines, it thin the cards  themselves do not work (please confirm)

 

 

Offline
Last seen: 4 hours 6 min ago
Joined: Jul 5 2018 - 09:44
Posts: 955
AppleWoz wrote
AppleWoz wrote:
softwarejanitor wrote:

Have you tried a different controller card?  I'm wondering if one of the capacitors on your Disk ][ Controller Card is leaking interference when it is running?  I'd suspect C2 as it is an electrolytic type which tend to go bad over time, but it could be one of the small ceramic/poly ones I suppose (C3 or C6 maybe since

 

Load based issues I've seen in the past on a ][+ were related to the 8T97 tri-state bus drivers...  But that was usually with multiple cards in the system and more with lock-ups and other weirdness than glitchy display.

 

Offline
Last seen: 1 day 33 min ago
Joined: Apr 17 2022 - 12:26
Posts: 49
I'm starting to be out of

I'm starting to be out of options here really. And this seems a problem multiple people have so this must be something "stock" the the apple II+ but I wonder what.

 

Like I always did on that repair, I'm focusing on many things at the same time, so I have other thing that look abnormal and I'll fix those to be sure I end up with a single problem

 

Currently, my 74LS195 in C2 is very hot to the touch after like 5 min of running, i'm sure it would be burning hot after an hour. Also the socket of it seems to have "melted" ( softened a bit ) because of that. That is probably not normal. That doesn't prevent the system from booting or running.

Only one 74S195 out of my stock makes the system boot. So there is something fishy.

 

That 74S195 needed replacement when I first got this Apple II, so maybe what causes it to overheat is still present.

All the signals on it look OK (I mean they are the right frequency)

 

  Id be glad to solve that problem

Offline
Last seen: 5 days 8 hours ago
Joined: Apr 27 2022 - 10:32
Posts: 15
CedsRepairs wrote:If I'm
CedsRepairs wrote:

If I'm right AppleWoz : you did swap to PSU with the PSU of an Apple IIe and got the sam result right ?

This would rule out the psu as the cause

 

On my side I fully recapped my PSU (to no end) as it had a lot of noise.

Truth is : the noise is created (well, its level is doubled) when I insert the diskII card.

 

On my side, I can confirm (as you do)

With the change of PSU nothing changes except that it was an end in itself test, since both PSU are 100% recapped, even with only the floppy card (without the two floppies connected) but when I start a floppy (game or software) the lines are drastically reduced almost to disappear

Offline
Last seen: 1 day 33 min ago
Joined: Apr 17 2022 - 12:26
Posts: 49
So your floppy works despite

So your floppy works despite the lines ?

Offline
Last seen: 5 days 8 hours ago
Joined: Apr 27 2022 - 10:32
Posts: 15
CedsRepairs wrote:So your
CedsRepairs wrote:

So your floppy works despite the lines ?

the lines do not affect in any way the operation of the floppies, the 2 floppy drives work perfectly, I opened the thread to understand if it was a defect or not and how to fix it but the apple 2 europlus works perfectly

Offline
Last seen: 4 hours 54 min ago
Joined: Nov 29 2020 - 19:48
Posts: 64
74LS195 in C2

So does the 74LS195 get hot only under load or do you get similar behavior without any cards?  What does the bottom of the motherboard look like around the 74LS195? Are there any bodge wires on the bottom side?

Offline
Last seen: 1 day 33 min ago
Joined: Apr 17 2022 - 12:26
Posts: 49
Super hot doing nothing and

Super hot doing nothing and with no cards

 

the only bodges are the ones required by the videx  EnhancerIi keyboard ; not really into that area but I'll test without them as I'm out of ideas anyways 

 

as I had to replace that chip ; and that the ic socket seems to have been altered by heat ; whatever it is it's there since quite a while 

I'm also flabbergasted that only one 74s195 accepts to work ( or maybe that only with only a specific one the computer boots - maybe the problem is somwewhzre else ) 

Offline
Last seen: 4 hours 54 min ago
Joined: Nov 29 2020 - 19:48
Posts: 64
Replace C2 socket for 74LS195

Since the socket for the 74LS195 at C2 shows some deformation I would replace it and the 74LS195 (again). I'm wondering  if the deformation is causing poor contact and generating heat. That seems a little out there, but so is this whole issue at this point.  Start working backward from the socket if the replacements also generate heat.  I usually start closest to the known problems.

Offline
Last seen: 1 day 33 min ago
Joined: Apr 17 2022 - 12:26
Posts: 49
I was thinking of replacing

I was thinking of replacing the socket too.

 

I already desoldered the keyboard socket out of the motherboard for solving a problem I did not have, so why not desolder another one :) 

 

Thankfully I have a nice hakko 951 that makes the unthinkable possible (I didn't say "easy")

 

I'll do that this evening.

Offline
Last seen: 1 day 33 min ago
Joined: Apr 17 2022 - 12:26
Posts: 49
I also ordered 74S from

I also ordered 74S from another seller. There's a lot of bizarre stuff out there. I wouldn't say counterfeits (nobody would waste time counterfeiting something that is as low volume as these old parts) but I'm sure the origin of many 74S/LS on ebay is fishy. I wouldn't be surprised if some of them are litterally trash picked, came from stuff that burned or was in use for decades.

 

Of course the seller alwyas states their parts are NOS. I don't think anybody has NOS 74 Shottky (S/LS) in 2022... 

 

 

Offline
Last seen: 4 hours 24 min ago
Joined: Apr 26 2016 - 08:36
Posts: 360
CedsRepairs wrote:I also
CedsRepairs wrote:

I also ordered 74S from another seller. There's a lot of bizarre stuff out there. I wouldn't say counterfeits (nobody would waste time counterfeiting something that is as low volume as these old parts) but I'm sure the origin of many 74S/LS on ebay is fishy. I wouldn't be surprised if some of them are litterally trash picked, came from stuff that burned or was in use for decades.

 

I try to stick to reputable sellers for vintage TTL logic.  Jameco Electronics, Unicorn Electronics, and if possible, see if new production parts are available from Farnell, Mouser and Digikey.

 

It's not that the parts are too low to counterfeit - the chinese will counterfeit anything they think will sell.  And harvest any number of parts from descarded electronics whether known working or not.

 

Offline
Last seen: 4 hours 54 min ago
Joined: Nov 29 2020 - 19:48
Posts: 64
Fake Chips Concern
baldrick is once again dead on, relabeled fakes and bad chips abound.  I too prefer Jameco, Mouser, Digikey and Unicorn.  I have been working on a number of ][ Plus machines and clones and ordered a complete set of 74LSxxx for the board to speed my troubleshooting just a couple of months ago. Good parts are available.  Shipping is the killer, that is why I ordered everything I thought I would need to test with good parts and replace common failures in one order. 
Offline
Last seen: 4 hours 6 min ago
Joined: Jul 5 2018 - 09:44
Posts: 955
8BitHeaven wrote:baldrick is
8BitHeaven wrote:
baldrick is once again dead on, relabeled fakes and bad chips abound.  I too prefer Jameco, Mouser, Digikey and Unicorn.  I have been working on a number of ][ Plus machines and clones and ordered a complete set of 74LSxxx for the board to speed my troubleshooting just a couple of months ago. Good parts are available.  Shipping is the killer, that is why I ordere

 

I will have to agree with both of you...  too many bogus chips out there.  Another source I might use is JDR Microdevices.  They used to be a major mail order vendor but now they seem to primarily sell through eBay.

 

 

 

Offline
Last seen: 1 day 33 min ago
Joined: Apr 17 2022 - 12:26
Posts: 49
This is not yet an

This is not yet an explanation, but for those interested in this "thin vertical lines" drama, here are my latests measurements

 

This is the ripple & noise on the 5V DC rail, with an appleII Europlus working and powered on.

(This is measured with the tiny ground probe at the tip of the oscilloscope probes and at the psu socket) so the best I can do.

 

 

And this is the same, with the DiskII card inserted

 

 

As you can see the ripple and noise is 4 orders of magnitude higher, and seems to have two frequency components (which would maybe explain line the pattern we see but I'm not a PAL expert). Yes remember this is eurplus so this is PAL output. Probably AppleWoz is PAL as well ?

To be fair 160mV peak to peak is not the end of the world. I believe this is more a sign of something not working properly than the actual cause of problems.

 

So can this ripple/noise you see explain the "thin lines" PAL pattern we see, if a Video expert is out there, let us know. If the frequencies are completely off, let us know as well so we stop digging in that direction.

 

It is difficult to confirm this ripple/noise is sent "as is" on the PAL output itself due to the nature of the PAL signal, but I suppose it's fair to assume it's at the very least disturbing the signal generation in some way.

 

My gutfeeling is that - as someone pointed out earlier - this is the "signature" of something doing a short circuit in somekind of bus. 133Khz is very close to 1Mhz/8 buut maybe I'm just seeing what I want to see.

 

And I'm still concerned about this burning hot 74S195 out there. I don't have enough 74LS spares to replace the whole clock section...

Offline
Last seen: 4 hours 6 min ago
Joined: Jul 5 2018 - 09:44
Posts: 955
CedsRepairs wrote:This is not
CedsRepairs wrote:

This is not yet an explanation, but for those interested in this "thin vertical lines" drama, here are my latests measurements

 

This is the ripple & noise on the 5V DC rail, with an appleII Europlus working and powered on.

(This is measured with the tiny ground probe at the tip of the oscilloscope probes and at the psu socket) so the best I can do.

 <

It really seems like something in the Disk ][ Controller Card or the floppy drive itself is introducing noise or failing to filter it.  I'm really suspecting a bad capacitor on either the Disk ][ Controller Card or the Disk ][ Analog Board.

 

Ideally test with a completely different drive and controller card and if the problem goes away, then you know it has to be there.

 

 

 

 

Offline
Last seen: 12 hours 58 min ago
Joined: Jun 18 2010 - 13:54
Posts: 508
Since the CPU seems to be

Since the CPU seems to be operating correctly and this is just a display issue, my bet is on bus crosstalk between the CPU and video circuits. Might be a slow bus transceiver or other chip. The 74S195 will run hotter than the rest.

Offline
Last seen: 1 day 33 min ago
Joined: Apr 17 2022 - 12:26
Posts: 49
Hey quick question, as we

Hey quick question, as we both have this line issue

 

1) Do you have a PAL encoder card

2) Do you have a possiblity to test it's output

3) Your text on your screen seems way too sharp to be out of PAL encoder to me ; I guess you're using the "onboard" monochrome connector  ?

 

I'm asking all this because strangely enough, I don't have the artifacts out of the PAL Encoder (but I hate the picture that gets out of this, way too soft)

 

Unless I'm mistaken, the onboard connector of an appleII Europlus is "PAL monochrome". From my reading, the motherboard has been modified at the factory for that ( Xtal, jumpers). Which means that if it is a timing issue; it's specific to PAL implementations.

 

Interestingly too, when using the onboard 2pin connector to feed the video signal in my PAL encoder card, I need to crank up the output level of composite nearly to the max. So I've used the other connector (the 4 pin one) that doesn't go thru the potentiometer. But that's not how the pal encoder is supposed to be connected. I'll later check if the composite "voltage divider" potentiometer is still within specs.

 

In short ;

1) I wouldn't be surprised if this problem is both stock, frequent, and very specific to only the Europlus computers. Which are a sub-niche of remaining APpleIIpluses. If it is a timing/crosstalk issue, it probably happens ONLY on those "modified" motherboards, with the PAL XTAL, which would explain why this problem is both widespread (amongst europluses) and we get little feedback ( 80% of people reading this forum probably use NTSC ). Someone on another forum mentionned that a vintage picture from Apple marketing themselves show an Apple screen with that same vertical lines pattern (!).

2) I wouldn't be surprised if this problem disappears if we "mod" the europlus to switch it back to NTSC (which I've been thinking of doing as modern LCD have no problem whatsoever with NTSC, and that's how I use my Apple //c)

 

This issue must have a cause, and I'd be glad if I can find it ; but I'm more and more enclined to believe it is a stock problem that occured even back in the day.

 

 

Offline
Last seen: 1 day 33 min ago
Joined: Apr 17 2022 - 12:26
Posts: 49
I've solved my 74S195 issues

I've solved my 74S195 issues by

- desoldering the socket (which imho had been melted by the heat of the chip making bad contacts)

- soldering a new one

- switching to 74LS195 now reliably works (?) and what's even better ; it seems the 74S86 (next to it) is now much less hot to the touch as well.

 

So I know this all sounds a bit voodoo and bizarre - but it seems "from my observations" that my super-hot 74S195 was caused by bad contacts on the socket (but that socket was good enough for having the computer boot and work), which turned into heat issues all over the clock circuitry for some reason.

 

MacFly's picture
Offline
Last seen: 1 day 16 hours ago
Joined: Nov 7 2019 - 13:49
Posts: 177
A bad (high resistance)

A bad (high resistance) contact cannot cause heat itself. A high resistance limits the current and reduces the overall power-loss.

Also, the 74195 does not have any three-state outputs, so it does not have any pins which are either inputs or outputs - and hence could cause an issue, if some input signal was not properly controlled. It only has fixed inputs and fixed outputs.

The (only) logical explanation for your issue is that the one of the five 74195 outputs had low-resistance to ground, to 5V or to another of its outputs (= a short circuit at one of its five outputs - probably a bad solder joint, connecting two pins). If the machine was otherwise still working fine, then the IC was still able to "overpower" the short circuit and drive the output pin to an acceptable TTL voltage level - but it required much more power than normal, thus the IC got hot. This would also explain why some ICs would work, while others wouldn't. The short circuit at its TTL output likely required a current beyond the specs - so while some ICs happened to be just strong enough, others weren't.

Offline
Last seen: 1 day 33 min ago
Joined: Apr 17 2022 - 12:26
Posts: 49
Okay back to this thin

Okay back to this thin vetical lines problems

 

I can now confirm that the more you load the thing, the more the lines are prominent.

As i' now succesufly repaired

- my motherbaord

-my psu

- my 16K card

-my pal encoder

 

And that I just added an 80col card : yes, those lines pop up more and more the more cards you had.

And interestingly, they do not appear on the output of the PAL encoder.

 

This must be in some way related to the PSU. I could feel it. I don't understand why the OP here has the same problem even after swapping PSU but I'd like to see that with my own eyes on my own setup.

 

My PSU is fully recapped (and as explained in my own topic I had no hope whatsoever this would solve anything) I also did en ESR test on all capacitors after they have been removed and all of them tested OK. I'm not saying it's entirely useless to replace 40yo stuff with new ones, just that this particular "thin lines" problem here isn't solved by that. Everything else is also recapped (the motherbaord and all the cards) and all TTL logic is tested.

 

After my holidays , in a couple of week, I'll power the Apple II with a modfiied mini ATX and i'm curious to see if the lines will be gone.

 

If so, I might take the time to get to the bottom of this problem with an external load and the recapped PSU, but I suppose it is down to a semi conductor that is acting up (and there aren't that many in an ASTEC PSU)

Offline
Last seen: 1 day 3 hours ago
Joined: Jun 12 2022 - 23:35
Posts: 22
CedsRepairs wrote:Okay back
CedsRepairs wrote:

Okay back to this thin vetical lines problems

 

I can now confirm that the more you load the thing, the more the lines are prominent.

As i' now succesufly repaired

- my motherbaord

-my psu

- my 16K card

-my pal encoder

 

And that I just added an 80col card : yes, those lines pop up more and more the more cards you had.

And in

Do you have an oscilloscope?  You could look at the power rails and see if there is some oscillation due to the switching power supply on the rails.

Also, if there is a bad ground connection in the PSU, it could result in some of the switching power supply signal being present on the power supply lines.   If this is the case, you may have to reheat solder joints in the PSU to reestablish a good ground connection.

Definitely swapping in a different supply seems to be the easiest way to eliminate the PSU as a possibility.

 

 

Offline
Last seen: 4 hours 24 min ago
Joined: Apr 26 2016 - 08:36
Posts: 360
MacFly wrote:A bad (high
MacFly wrote:

A bad (high resistance) contact cannot cause heat itself. A high resistance limits the current and reduces the overall power-loss.

 

Oh yes it can.  In fact, high resistance poor contact (like corrosion between juncitons of old aluminum wire improperly connected to modern copper wiring, for instance) is the very cause of thousands of house fires every year.  You're right that a given high resistance junction cannot cause heat on its own, but pass a current through it, and it can potentially generate lots of heat.  And if iit cannot radiate that heat faster than it's gnerated then its temperature will increase.

 

If you have a high resistance contact, you do get a voltage drop, but you still get current passing.  So yes, the voltage drop can be described by Ohm's law, V=IR where V is the voltage drop for the given resistance and said amount of current passing through the high resistance junction, but you also have to take into account the power law, P=IV where P is the amount of heat generated (in watts) for the given current passing through the given resistance.  This is also the basis for resistor power ratings.

 

Sorry for going off topic.

MacFly's picture
Offline
Last seen: 1 day 16 hours ago
Joined: Nov 7 2019 - 13:49
Posts: 177
You are missing the point,

You are missing the point, baldrick. Of course high resistance contacts gets a bit warm. But a simple logic IC or RAM converts all its power into heat anyway. So, when a contact has a high resistance, yes, it gets a bit warm - but it limits the current and slightly reduces the dissipated power by the IC itself. So overall, IC socket will not disspate more power than before - and a high resistance contact cannot explain why an IC socket suddenly gets burning hot.

That would only be the case, however, if the voltage drop caused some misbehaviour of the IC. Like when a chipselect line is not properly connected and suddenly the IC drives its output when it shouldn't. But that's impossible in this case since this logic IC has only fixed inputs and fixed outputs. No misbehaviour is possible, which could suddenly cause this IC to draw excessive power.

You can try it on a breadboard. Pick a simple logic IC like a 74LS00, use a 1K Ohm resistor on an any input or output - and see if the overall current suddenly increases or the IC got burning hot. That's not possible. The only explanation for the misbehaviour above is a short-circuit on an any of the ICs fixed outputs.

Offline
Last seen: 4 hours 24 min ago
Joined: Apr 26 2016 - 08:36
Posts: 360
I get your analogy.My guess

I get your analogy.

My guess is that the hot socket is due to a poor contact on the +5V line, not one of the data lines.

The currents generated are way to small (because generally they feed to high impedance destinations) to make any sigificant amount of heat on the signal pins.

 

 

Log in or register to post comments