MegaSTe hardware problems -help locate it please!

Troubles with your machine? Just want to speak about the latest improvements? This is the place!

Moderators: Mug UK, Zorro 2, Greenious, spiny, Moderator Team

User avatar
Fujiyama
Atari Super Hero
Atari Super Hero
Posts: 580
Joined: Thu Jul 12, 2007 8:21 am
Location: Norway

MegaSTe hardware problems -help locate it please!

Postby Fujiyama » Sun Jun 04, 2017 10:49 pm

I'm having multiple hardware problems with my MegaSTe and troubleshooting is futile. I'm looking for some advice. Specs:

- 4MB RAM
- TOS 2.06
- MonSTer board with an IDE-CF adapter and CF card acting as a hard drive replacement (boots from it as the ACSI IRQ pin is attached, and HDdriver 10 is used)
- Atari SM-144 monochrome hires monitor
- upgraded with a Sony MPF 520-1 (HD) floppy drive, GAL and AJAX chip
- standard internal SCSI hard drive (with MegaSTe hard drive interface) -the hard drive is currently removed

The computer hasn't been used for years, but decided to take it out of storage not long ago, my main use being MIDI using Notator SL.
These are the symptoms:


1) no keyboard/mouse response when booting (sometimes).
I should be able to skip the TOS 2.06 RAM test by pressing a key (and also hear that familiar "click" sound when pressing keys) but often nothing happens. I've checked for a contact problem with the keyboard connector, but it all seems fine. Recently I've opened up the keyboard as well, attached/reinserted the connectors but they all seem fine. Also, that part works again now, so it's hard to say until/if it reappears

2) 4 bombs when booting (sometimes).
I've heard this can be due to the floppy drive not being connected, but in my case I reckon it might be a bad floppy cable as the casing has been opened so many times and it seems a bit bent/folded messed up. I've just ordered a new one, so for now the floppy drive and cable has been removed.
The strange thing is that booting most of the time recently works fine (no bombs) from the MonSTer/IDE-CF interface.

3) MIDI IN isn't working.
For some reason MIDI data is received from a synth without playing it as free memory keeps decreasing constantly (Notator SL) and rapidly as long as a cable is attached to the MIDI IN port. I assumed it was "active sensing", but at least one of the synths I tried it with has it turned off. I couldn't figure out what kind of data it was, but it apparently fills up the buffer and slows down the computer until I remove the MIDI cable from the Atari's IN. I don't think it's a loop problem either because the issue is still there if the OUT cable is removed.
If I enable the sequencer for recording and play something, nothing will be recorded. Is appears not to see any MIDI input at all.
MIDI OUT however works fine (playing sequence files I've done from before).

I've checked for bad connections using the computer's schematics and a multimeter for continuity and all connections seem fine. The same goes for the (QWERTY) keyboard connector. I've also re-soldered a lot of connections on the motherboard in case of hairline fractures or otherwise bad connections. ICs in sockets have been removed and re-inserted.
Suspecting the MonSTer I've used the computer without it but the symtoms are still there, so apparently that's fine.

What else can I try out/look into?
Mega STe | MonSTer with dual IDE-CF memory card adapter | STe | SM-144 |NEC Multisync 1990SXi | IDE doubler | ST_ESSC | RSVE | ICD Link II | Link '97 | HD floppy drive/AJAX | HD floppy module | Minolta PCMCIA card-drive | Realtime Clock module | Discovery cartridge | Unitor-2 | Export | Combiner | Steady Eye | Human Touch | Unicorn USB

User avatar
Shockwav3
Captain Atari
Captain Atari
Posts: 224
Joined: Thu Feb 02, 2006 5:46 pm
Location: Berlin, Germany
Contact:

Re: MegaSTe hardware problems -help locate it please!

Postby Shockwav3 » Sun Jun 04, 2017 11:05 pm

MIDI In: Check U301, U302, U303 - if you're lucky it's just a gate on U301 gone bad.

Can I haz hardware?

User avatar
Fujiyama
Atari Super Hero
Atari Super Hero
Posts: 580
Joined: Thu Jul 12, 2007 8:21 am
Location: Norway

Re: MegaSTe hardware problems -help locate it please!

Postby Fujiyama » Mon Jun 05, 2017 8:29 pm

Let's hope so as that 7406 (U301) should be easy and cheap to get hold of. Any type will work (LS, HC, HCT etc.), right?

As for the PC900 opto-isolator (U302) -they seem to be obsolete these days, but aren't there pin-pin replacements with higher performance these days?

Finally, the MC68850 (U303) also appears to be obsolete as I can't even find a datasheet for it! What kind of IC is this and where is it located on the MegaSTe PCB? Perhaps underneath my MonSTer board as I've looked everywhere else?
UPDATE: never mind the location of U303. It appears to be underneath the MonSTer board, unsocketed.
Mega STe | MonSTer with dual IDE-CF memory card adapter | STe | SM-144 |NEC Multisync 1990SXi | IDE doubler | ST_ESSC | RSVE | ICD Link II | Link '97 | HD floppy drive/AJAX | HD floppy module | Minolta PCMCIA card-drive | Realtime Clock module | Discovery cartridge | Unitor-2 | Export | Combiner | Steady Eye | Human Touch | Unicorn USB

User avatar
Fujiyama
Atari Super Hero
Atari Super Hero
Posts: 580
Joined: Thu Jul 12, 2007 8:21 am
Location: Norway

Re: MegaSTe hardware problems -help locate it please!

Postby Fujiyama » Thu Jun 08, 2017 11:37 am

OK, I've replaced U301 (7406) with no change, but I just found out that U609 (LM78L82ACZ) which I've identified as being an 8.2V regulator had one or two if its legs cut off, causing contact problems (likely due to being in close proximity to the power supply which has been removed/reinserted many times and likely caused this to happen).
It appears to be unavailable as far as I can see -does anyone know where I can buy a new one?

I see from the schematic that it supplies power to U608 (LMC1992) which I understand handles stereo audio volume/tone control. So this could explain why I haven't been getting the usual "click" sound when pressing the keyboard keys. I guess it also explains another issues I've been having: if I go to the "sound" CPX control panel I've been getting an error message saying "Stereo sound not available" which led me thinking that the DIP switch block (S701) near the TOS ROMs was having contact issues with DIP #8. Does all of this sound reasonable?
But it still doesn't explain why there's no reaction when I press the keys when booting (the TOS 2.06 memory test should quit and the computer should continue booting).
Is it likely that U304 (68850) is damaged? That's the first IC which the keyboard signals goes into after the jack connector/line filter.
Last edited by Fujiyama on Mon Jun 26, 2017 8:19 pm, edited 1 time in total.
Mega STe | MonSTer with dual IDE-CF memory card adapter | STe | SM-144 |NEC Multisync 1990SXi | IDE doubler | ST_ESSC | RSVE | ICD Link II | Link '97 | HD floppy drive/AJAX | HD floppy module | Minolta PCMCIA card-drive | Realtime Clock module | Discovery cartridge | Unitor-2 | Export | Combiner | Steady Eye | Human Touch | Unicorn USB

User avatar
Fujiyama
Atari Super Hero
Atari Super Hero
Posts: 580
Joined: Thu Jul 12, 2007 8:21 am
Location: Norway

Re: MegaSTe hardware problems -help locate it please!

Postby Fujiyama » Thu Jun 08, 2017 12:26 pm

UPDATE:
I've made some new discoveries:

a) having soldered the pins of U609 (the 78L82 voltage regulator) back on to the motherboard (a temporary fix, I still need to get a new one) I can now confirm that U608 (LMC1992) is getting +8.2V (I measured +8.4V) on pin 28.

b) there's still "No stereo sound available" even though the above has been fixed and pin #8 is confirmed on (I checked that pin 11 of U705 (74LS244) indeed goes to GND).

c) having put an IC socket in place of the old U301 (7406) and inserting two different 7406 ICs I'm getting different results when booting/rebooting:

- U301 (7406N): boot OK, keyboard/mouse response OK. Makes no difference if floppy drive is connected
- U301 (7406N -same IC as above), but with MIDI IN connected to a synth's MIDI OUT (powered on): boots OK, but no keyboard/mouse response and no key click. Makes no difference if floppy drive is connected.
- U301 (HD7406): won't boot (the Atari boot screen pops up but there's no memory test appearing) when floppy drive is attached. When floppy drive is removed the memory test appears but there's no keyboard response or "click" keypress sound.

These were two 7406 chips I had in my spare parts box, so they could very well be malfunctioning, but I currently don't have any other ones to test. Still it appears U301 (7406) is doing several important things.... thoughts on this anyone?
Mega STe | MonSTer with dual IDE-CF memory card adapter | STe | SM-144 |NEC Multisync 1990SXi | IDE doubler | ST_ESSC | RSVE | ICD Link II | Link '97 | HD floppy drive/AJAX | HD floppy module | Minolta PCMCIA card-drive | Realtime Clock module | Discovery cartridge | Unitor-2 | Export | Combiner | Steady Eye | Human Touch | Unicorn USB

siriushardware
Captain Atari
Captain Atari
Posts: 347
Joined: Thu Aug 21, 2014 7:55 pm
Location: UK

Re: MegaSTe hardware problems -help locate it please!

Postby siriushardware » Mon Jun 26, 2017 10:20 pm

Fujiyama: Decided to continue in your thread here, because it invites suggestions from anyone else who may have good ideas.

I've looked at the wider area around the sound section and I see that it's not only the LMC1992 which runs on the 8.2V supply.

All four elements of IC 605, LF357 (Two immediately to the right of the LMC1992, and one to the right of each of the two DACs on the circuit diagram) run from a +8.2V supply, so we can guess where that must come from. (The little 78L82).

You should check to see if there is 8.2V not only on LMC1992 pin 28 where you already know there is, but also on on pin 4 of IC605.

While you are looking in that area, check that the positive supply to IC607 (also LF357) pin 4 is +5V.

Finally check that the minus supply for IC605 and IC607, on pin 11 of both of those ICs, is minus 5 volts. The diagram set I found online does not include the power supply but the pin markings on the power connector J201 (sheet 2 of 11, centre) show that the minus 5 volt supply comes from the main power supply itself, unlike the +8.2V supply which is derived locally using that 8.2V regulator.

With respect to the 7406, you seem pretty unlucky to have three plain 7406 devices which all appear to be faulty in different ways. According to my databook the main difference between the 'original' and LS versions of the 7406 is the current consumption, about 32mA for the original device and 6mA for the LS version. The propagation delay for both types is given as 12.5uS, so there's no apparent speed difference. Both are inverting buffers with open-collector outputs rated at 30V maximum. It would do no harm to try a 74LS06 if you have one.

As to exactly what the 7406 (IC301) is involved in, as well as the four elements used in the MIDI out / Midi thru buffers I see one element of the same 7406 (IC301) in the output from the 'power-on-reset' circuit (upper left area, sheet 2 of 11) which then goes to U801, the TTSCU? IC. (Sheet 3 of 11). If one or more of your suspect 7406 ICs have the 'A' element (pins 1 and 2) damaged that might explain why the MSTe will not even boot. I can't find the missing sixth ('B') element of IC301 - you might have better luck if you try. It may not be used.

I notice there is another 7406 in the machine (IC 406) which is mainly concerned with buffering signals between the FDD controller IC and the FDD, where five of its elements are used for that purpose. (Sheet 4 of 11). If you put that IC in a socket as well it will give you the option to swap IC301 and IC406. However, I haven't yet found the sixth element of IC406 either, so I'm not sure if that IC is involved in anything else.

With reference to IC303 (Midi UART) and IC304 (Keyboard UART) these are apparently (EDIT) identical 68B50 ICs, so if they are socketed or if you are prepared to put them in sockets you could try swapping those over to see what happens.

With respect to the PC900 MIDI input optocoupler which you were discussing elsewhere, there is no harm in trying a modern 'improved' equivalent but if you do that and it still does not work then you will be left in doubt as to whether the new version is really as compatible as it is supposed to be. If you fit the original 'legacy' type PC900, then, provided the ones you buy are not fakes, it should 'just work'. If they are cheap enough, just buy one example of each and fit the 'legacy' version while you are trying to get it up and running, and then once it works, fit the 'improved' type to see if it really does improve matters.
Last edited by siriushardware on Sun Jul 02, 2017 9:31 am, edited 1 time in total.

User avatar
Fujiyama
Atari Super Hero
Atari Super Hero
Posts: 580
Joined: Thu Jul 12, 2007 8:21 am
Location: Norway

Re: MegaSTe hardware problems -help locate it please!

Postby Fujiyama » Tue Jun 27, 2017 7:40 am

Thanks for your comments, Siriushardware.
I'm going to check the voltage on those pins, but you talk about fake semiconductors, something I want to avoid of course.
Some of the components are hard to come by, but someone suggested utsource which I did and found several components which didn't pop up elsewhere including the 8.2V regulator, 74LS06 and the PC900V opto-isolator (also an improved version: PC900V0NSZXF).

Do you reckon I'll be getting fakes from there? Perhaps someone else reading has ordered from utsource and can tell their experiences.
Mega STe | MonSTer with dual IDE-CF memory card adapter | STe | SM-144 |NEC Multisync 1990SXi | IDE doubler | ST_ESSC | RSVE | ICD Link II | Link '97 | HD floppy drive/AJAX | HD floppy module | Minolta PCMCIA card-drive | Realtime Clock module | Discovery cartridge | Unitor-2 | Export | Combiner | Steady Eye | Human Touch | Unicorn USB

AtariZoll
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2759
Joined: Mon Feb 20, 2012 4:42 pm
Contact:

Re: MegaSTe hardware problems -help locate it please!

Postby AtariZoll » Tue Jun 27, 2017 7:54 am

Do I see it right, that no mention of main power supply, main power voltages ? Replacing electrolyte capacitors in PSU, those filtering +5V, +12V, on mbo.
Especially important when it was not used over years.
English language is like bad boss on workplace: it expecting from you to strictly follow all, numerous rules, but self bending rules as much likes :mrgreen:

siriushardware
Captain Atari
Captain Atari
Posts: 347
Joined: Thu Aug 21, 2014 7:55 pm
Location: UK

Re: MegaSTe hardware problems -help locate it please!

Postby siriushardware » Tue Jun 27, 2017 5:28 pm

AtariZoll: Of course that's true, but Fujiyama has already demonstrated that he can trace broken regulators, read circuit diagrams and measure voltages. His most recent questions were specifically about the areas we were talking about, in particular no stereo sound.

Fujiyama: I did not mean to imply that UTsource can not be trusted to sell you genuine parts, indeed it would be remiss of me to suggest so when I know absolutely nothing about them.

There is a problem with fake components coming onto the market generally - RF output transistors used in old communications equipment have been a particular problem, and I've also heard that there has been a problem with some fake 2N3055 power transistors which have the correct external package but a much smaller / weaker device inside - so they will pass on a basic transistor test, but collapse when subjected to a full load.

Basically any 'classic' semiconductor which has been out of production for many years and is still sought after by people trying to keep retro or vintage equipment going can potentially be a target for counterfeiters.

Order your optocouplers from where ever you can find them, and the odds are they will probably be OK. Word of mouth / peer review is certainly one way to get reassurance about a particular supplier or product.

User avatar
Fujiyama
Atari Super Hero
Atari Super Hero
Posts: 580
Joined: Thu Jul 12, 2007 8:21 am
Location: Norway

Re: MegaSTe hardware problems -help locate it please!

Postby Fujiyama » Thu Jun 29, 2017 10:51 am

Good suggestion about the power supply AtariZoll. I will look into that later to measure the physical size as well as check their values. You mean all electrolytic capacitors in the PSU? Panasonic is a good brand for capacitors I've been told, but they have several series. Which ones do you recommend, or another brand perhaps?

Siriushardware: I'm become wary of ordering from UTsource mainly because of all the postings/feedback that came up after searching the web for "UTsource fake". Then again people some people are rather quick to shout "scam! fake! alert! warning!" almost immediately before checking if there are other reasons things don't work, so it's hard to say. But so many "fake" postings aren't exactly reassuring, the company being from China and all...

I take it ordering from the large electronics dealers (Farnell, RS components, Mouser, Digikey etc.) will be relatively safe, the only problem being that with their rather high prices (some have minimum orders and all of them operative with rather high shipping costs (DHL, Fedex etc.) even though I personally don't care if I'd have to wait a week instead of till the next day) you really have to put together a big order for it to be worth it. And of course one company doesn't have all the parts you need, which is why I'm trying to find equivalents for the obsolete ones (such as the PC900 opto isolator).

UPDATE: I've just learnt that the (currently still available) H11N1 opto-isolator is fully compatible with the PC900, so I'm going to order that together with a couple 74LS06N and capacitors for the PSU (and I suppose the big ones for the main board as well -same for my STe). I've already ordered a 78L82 voltage regulator from the UK eBay seller Siriushardware suggested (although more expensive I don't want to take my chances with UTsource. It'll be more expensive/difficult to replace the stereo tone/volume chip in the computer I'm sure if it blows up because of a fake/low spec regulator).
Oh, all this after checking the voltage of those ICs suggested a couple of posts back in case I need those replaced as well.
Mega STe | MonSTer with dual IDE-CF memory card adapter | STe | SM-144 |NEC Multisync 1990SXi | IDE doubler | ST_ESSC | RSVE | ICD Link II | Link '97 | HD floppy drive/AJAX | HD floppy module | Minolta PCMCIA card-drive | Realtime Clock module | Discovery cartridge | Unitor-2 | Export | Combiner | Steady Eye | Human Touch | Unicorn USB

siriushardware
Captain Atari
Captain Atari
Posts: 347
Joined: Thu Aug 21, 2014 7:55 pm
Location: UK

Re: MegaSTe hardware problems -help locate it please!

Postby siriushardware » Thu Jun 29, 2017 6:35 pm

When you get your regulator, we'll suggest a way to test it for basic functionality before you put it in the machine. Do you have a bench power supply which can provide either a variable DC voltage or one or more DC voltages? Do you have an experimenter's 'breadboard'?

To be honest, I would say that the 78L82 is not the sort of exotic high value semi which people are likely to bother to fake.

I don't want to burst your bubble re: The H11N1 but I have just looked at the datasheets for the H11N1 and the PC900V, and although both incorporate a schmitt trigger (the effect of which is to transform slowly rising or falling input signals into sharp, fast output signals, which is a desirable characteristic for MIDI) the one in the PC900 is then followed by an open-collector transistor which provides the actual output.

In the H11N1, the output is directly from the schmitt trigger itself, that is, the output is a logic-level output instead of an open-collector output. The two devices are therefore NOT compatible.

Here are the internal diagrams of the PC900 and H11N1 respectively: Inside the PC900 You can clearly see the open collector output transistor, along with the internal voltage regulator and amplifier which not many other optocouplers will have. You can also see that the output from the
schmitt trigger in the H11N1 is a logic output: There is no open collector output, as there is on the PC900. In fact, I never realised until just now how specialised the PC900 is.

I would urge you, if at all possible, just to obtain and fit the correct PC900 device if you can, and then only if you have established that the original opto is actually faulty.

What were your results from measuring

(a) The main power supply voltages
(b) The voltages on the supply pins of IC605 and IC607?

PC900.jpg
H11N1.jpg
You do not have the required permissions to view the files attached to this post.

siriushardware
Captain Atari
Captain Atari
Posts: 347
Joined: Thu Aug 21, 2014 7:55 pm
Location: UK

Re: MegaSTe hardware problems -help locate it please!

Postby siriushardware » Thu Jun 29, 2017 7:47 pm

Been thinking further about this: Assuming you don't have an oscilloscope or a logic probe (stop me if you do have either of those things), there is a way you can test the PC900 in-circuit and the MSTe does not need to be fully working, as long as the +5V supply is OK, and assuming that at least some of the 7406 (U301) is OK. You can make sure of that by replacing U301.

You will need:

-A MIDI keyboard or synth which has an optional 'LOCAL OFF' mode.

This is where the synth effectively splits into two completely separate units in one box: A MIDI keyboard which sends MIDI information out from the synth's MIDI OUT port, and a MIDI tone module which plays sounds in response to MIDI information arriving at MIDI IN. When LOCAL OFF is turned on, playing on the keys won't produce any sounds from the synth. But if you connect the synth's MIDI OUT to the synth's MIDI IN with a MIDI cable, the notes you play are transmitted out from the keyboard, along the MIDI cable and back in to the tone module and you hear sounds. For future reference, this is quite a useful way to test a MIDI cable when you only have one keyboard or synth handy.

Assuming your synth can do this, the second thing you need for the test is a modified MIDI cable wired like this: Either make one, or temporarily rewire an existing one.

MSTe_midi_thru_cable..jpg


Now:
-Connect synth MIDI OUT to MSTe MIDI IN with a normally wired MIDI cable
-Connect MSTe MIDI OUT to synth MIDI IN using the modified cable
-Turn on the synth and put it into LOCAL OFF mode.
-Turn on the MSTe power.

Play something on the synth keyboard. If you hear sounds coming out on the synth, the PC900, U301F, and U301E are all working, and you don't need to replace the PC900. The test does not prove the other elements of U301 are working, however. It is quite possible for only one element in a logic IC to be blown, and all the rest working.

This test takes advantage of the fact that the MSTe has hardware MIDI THRU. Any MIDI data received on the MSTe's MIDI input through the PC900 is retransmitted through U301F and U301E to pins 3 and 1 on the MSTe's MIDI OUT port, from where (with a suitably rewired cable) it can be sent back to the synth / keyboard.
You do not have the required permissions to view the files attached to this post.

User avatar
Greenious
Hardware Guru
Hardware Guru
Posts: 1160
Joined: Sat Apr 24, 2004 5:39 pm
Location: Sweden

Re: MegaSTe hardware problems -help locate it please!

Postby Greenious » Sat Jul 01, 2017 3:48 pm

Actually the ACIA is MC6850, the particular model used is 68B50, not 68850, which is why you can't find the datasheet.

Anyway, I'm inclined to think this either is a broken ACIA, or MFP problem. After reading through your posts in the thread, I suspect that either one of the ACIAs, or the MFP, is constantly triggering the kbd/midi interrupt (they share). And I believe that the interrupt first checks with the IKBD to see if it got data, and if not, moves on to the MIDI, which is the one left that has to have triggered the interrupt...

4 bombs when booting is trickier, but usually comes down to internal HDD, or floppy, connected but not powered. It can also happen with external ACSI drives connected but not powered on.
Updated my guides as of june 28th, 2016. Check'em out and feedback!
viewtopic.php?t=5040

siriushardware
Captain Atari
Captain Atari
Posts: 347
Joined: Thu Aug 21, 2014 7:55 pm
Location: UK

Re: MegaSTe hardware problems -help locate it please!

Postby siriushardware » Sun Jul 02, 2017 9:22 am

Greenious wrote:Actually the ACIA is MC6850, the particular model used is 68B50, not 68850, which is why you can't find the datasheet.


Sorry, that was entirely down to the rough quality of the diagrams, where the "B" looked like an '*8" to me.

User avatar
Fujiyama
Atari Super Hero
Atari Super Hero
Posts: 580
Joined: Thu Jul 12, 2007 8:21 am
Location: Norway

Re: MegaSTe hardware problems -help locate it please!

Postby Fujiyama » Mon Jul 03, 2017 12:01 am

That's a lot of great advice and comments!

I've been away for a few days but will get round to checking those voltages hopefully tomorrow and will get back here with my findings.
In the meantime I'm looking into replacement capacitors for the Mega STe power supply: for Panasonic capacitors, which of their many series would be suitable for this type of application?
Mega STe | MonSTer with dual IDE-CF memory card adapter | STe | SM-144 |NEC Multisync 1990SXi | IDE doubler | ST_ESSC | RSVE | ICD Link II | Link '97 | HD floppy drive/AJAX | HD floppy module | Minolta PCMCIA card-drive | Realtime Clock module | Discovery cartridge | Unitor-2 | Export | Combiner | Steady Eye | Human Touch | Unicorn USB

User avatar
Fujiyama
Atari Super Hero
Atari Super Hero
Posts: 580
Joined: Thu Jul 12, 2007 8:21 am
Location: Norway

Re: MegaSTe hardware problems -help locate it please!

Postby Fujiyama » Mon Jul 03, 2017 1:05 pm

Never mind about the recapping questions as I found this thread (psu.why & how to replace the dreaded capacitors in ATARI ST!).
I've opened up the PSU and am in the midst of measuring the cap sizes and will check those voltages when done.

Yes, I do have a small pocket oscilloscope which I made from a kit (JYE tech DSO Shell DSO-150) but have limited experience in using it. I also have a good digital multimeter and a solderless breadboard but no lab bench PSU unfortunately.
Mega STe | MonSTer with dual IDE-CF memory card adapter | STe | SM-144 |NEC Multisync 1990SXi | IDE doubler | ST_ESSC | RSVE | ICD Link II | Link '97 | HD floppy drive/AJAX | HD floppy module | Minolta PCMCIA card-drive | Realtime Clock module | Discovery cartridge | Unitor-2 | Export | Combiner | Steady Eye | Human Touch | Unicorn USB

User avatar
troed
Atari God
Atari God
Posts: 1182
Joined: Mon Apr 30, 2012 6:20 pm
Location: Sweden

Re: MegaSTe hardware problems -help locate it please!

Postby troed » Mon Jul 03, 2017 2:50 pm

Fujiyama wrote:I've opened up the PSU and am in the midst of measuring the cap sizes and will check those voltages when done.


I recapped a Mega STE this weekend, using a kit from exxos. He lists the values he uses in his kits here:

http://exxoshost.co.uk/atari/last/psu/ (scroll down to Mega STE)

His store: https://www.exxoshost.co.uk/atari/last/storenew/

/Troed

User avatar
Fujiyama
Atari Super Hero
Atari Super Hero
Posts: 580
Joined: Thu Jul 12, 2007 8:21 am
Location: Norway

Re: MegaSTe hardware problems -help locate it please!

Postby Fujiyama » Mon Jul 03, 2017 4:14 pm

Cool!
May I ask if the MOSFET transistor listed there was also included, or just the caps?

Did you notice any difference before/after?
Mega STe | MonSTer with dual IDE-CF memory card adapter | STe | SM-144 |NEC Multisync 1990SXi | IDE doubler | ST_ESSC | RSVE | ICD Link II | Link '97 | HD floppy drive/AJAX | HD floppy module | Minolta PCMCIA card-drive | Realtime Clock module | Discovery cartridge | Unitor-2 | Export | Combiner | Steady Eye | Human Touch | Unicorn USB

User avatar
troed
Atari God
Atari God
Posts: 1182
Joined: Mon Apr 30, 2012 6:20 pm
Location: Sweden

Re: MegaSTe hardware problems -help locate it please!

Postby troed » Mon Jul 03, 2017 5:13 pm

Fujiyama wrote:Cool!
May I ask if the MOSFET transistor listed there was also included, or just the caps?

Did you notice any difference before/after?


Only the caps. Also, I recapped for someone else so I have no knowledge on workings before/after.

User avatar
Fujiyama
Atari Super Hero
Atari Super Hero
Posts: 580
Joined: Thu Jul 12, 2007 8:21 am
Location: Norway

Re: MegaSTe hardware problems -help locate it please!

Postby Fujiyama » Mon Jul 03, 2017 11:24 pm

siriushardware wrote:I've looked at the wider area around the sound section and I see that it's not only the LMC1992 which runs on the 8.2V supply.

All four elements of IC 605, LF357 (Two immediately to the right of the LMC1992, and one to the right of each of the two DACs on the circuit diagram) run from a +8.2V supply, so we can guess where that must come from. (The little 78L82).

You should check to see if there is 8.2V not only on LMC1992 pin 28 where you already know there is, but also on on pin 4 of IC605.


Finally gotten round to doing that now.
The problem is that the 78L82 (8.2V regulator) is very flaky. I did resolder the broken pins, but they're broken right up to its body and keeps getting loose, and it also seems the voltage fluctuates ever so slightly (0.01V) between some 5V readings., depending on if it's attached or not. I assume this might be normal, but we'll find out more when the new 78L82 arrives. Until then, take the following readings with an ever so slight grain of salt.
U605 and U607 are LF347 devices by the way, not LF357.

U508 (LMC1992) pin 28 shows +8.43V
U605 (LF347) pin 4 shows +8.43V

While you are looking in that area, check that the positive supply to IC607 (also LF357) pin 4 is +5V.


U607 (LF347) pin 4 shows +4.83V

Checking the PSU could also be useful when troubleshooting. Using the Mega STe PSU (Phihong PSM-5341) schematics I've been able to determine what the wire colours of its connector correspond to. With the negative probe of my digital multimeter attached to one of the four black wires (GND) I've used the positive probe to measure each of those other pins in turn as follows:

+5V line (three red wires): +5.025V
-5V line (white wire): -5.036V
-12V line (blue wire): -12.572V
+12V line (yellow wire): +12.120V

There's also an additional (orange) wire which the PSU schematic has labelled "Power good". I have no idea what this is for but it shows +4.892V
IMHO the PSU output appears to be within specs but I suppose there's more to it than that, like spiky/noisy power, right? Is this something I can check simply by attaching my oscilloscope to GND and one of the PSU power lines? With limited knowledge of using a scope and the knowledge that you should keep away from switched PSU units unless you know exactly what you're doing I've refrained from a "trial and error" type approach here, not wanting to risk blowing up the scope.

Finally check that the minus supply for IC605 and IC607, on pin 11 of both of those ICs, is minus 5 volts. The diagram set I found online does not include the power supply but the pin markings on the power connector J201 (sheet 2 of 11, centre) show that the minus 5 volt supply comes from the main power supply itself, unlike the +8.2V supply which is derived locally using that 8.2V regulator.


U605 (LF347) pin 11 shows +4.83V
U607 LF347) pin 11 shows +4.83V

With respect to the 7406, you seem pretty unlucky to have three plain 7406 devices which all appear to be faulty in different ways. According to my databook the main difference between the 'original' and LS versions of the 7406 is the current consumption, about 32mA for the original device and 6mA for the LS version. The propagation delay for both types is given as 12.5uS, so there's no apparent speed difference. Both are inverting buffers with open-collector outputs rated at 30V maximum. It would do no harm to try a 74LS06 if you have one.


Unfortunately I've tried the ones I have. Once I determine which other parts I need/may need I'll be ordering a few 74LS06 devices from Farnell.
I'm disappointed but not surprised that neither of the two '06 chips I had worked correctly as they were found in my parts box and likely already used. With brand new components from a reputable dealer it should be easier to focus on which parts actually need to be replaced.

As to exactly what the 7406 (IC301) is involved in, as well as the four elements used in the MIDI out / Midi thru buffers I see one element of the same 7406 (IC301) in the output from the 'power-on-reset' circuit (upper left area, sheet 2 of 11) which then goes to U801, the TTSCU? IC. (Sheet 3 of 11).


Right. I see it, marked U301A.
You mean sheet 8 (not 3) of 11 for the TTSCU (U801), right? I believe it's the XPOR line from U301A (pin 2). I haven't been able to identify what kind of IC U801 is (the schematics don't say, it appears to be underneath the rightmost VME IDC-cable connector which I haven't photographed unfortunately, and the computer is powered on now (after finally getting a successful boot), so I'll check this later if necessary by removing the cable.
I could only find U301C, U301D, U301E and U301F (sheet 3) and U301A (sheet 2) like you did. No sign of any U301B in the schematics. The PCB is mounted in the bottom part of the Mega STe casing while testing, but I'll try to remember to check where each pin goes of U301 so we can determine if there is a "B" or if it's unused.

If one or more of your suspect 7406 ICs have the 'A' element (pins 1 and 2) damaged that might explain why the MSTe will not even boot. I can't find the missing sixth ('B') element of IC301 - you might have better luck if you try. It may not be used.


Could be. Part of the reason for the 4 bombs I'm getting is likely due to the missing floppy drive. The strange thing is that it sometimes boots fine anyway. I've just received a new floppy cable which I'll try, but I don't think that's the source for all these issues. In addition to the obviously problematics 78L82 I wouldn't at all be surprised if things start working again after getting a new 74LS06.


I notice there is another 7406 in the machine (IC 406) which is mainly concerned with buffering signals between the FDD controller IC and the FDD, where five of its elements are used for that purpose. (Sheet 4 of 11). If you put that IC in a socket as well it will give you the option to swap IC301 and IC406. However, I haven't yet found the sixth element of IC406 either, so I'm not sure if that IC is involved in anything else.


I couldn't find it but suspect it might be underneath the hard drive controller board.
However, I prefer not to desolder anything unless I really need to, just in case something goes wrong. If it (U406) works, don't fix it :wink:
I'll wait for my new replacement instead.


With reference to IC303 (Midi UART) and IC304 (Keyboard UART) these are apparently (EDIT) identical 68B50 ICs, so if they are socketed or if you are prepared to put them in sockets you could try swapping those over to see what happens.


I'm not sure if they're socketed or not. I believe they might be underneath the MonSTer board, so I'll have to look into that later.
That's all for now. Time to get some shuteye.
Mega STe | MonSTer with dual IDE-CF memory card adapter | STe | SM-144 |NEC Multisync 1990SXi | IDE doubler | ST_ESSC | RSVE | ICD Link II | Link '97 | HD floppy drive/AJAX | HD floppy module | Minolta PCMCIA card-drive | Realtime Clock module | Discovery cartridge | Unitor-2 | Export | Combiner | Steady Eye | Human Touch | Unicorn USB

User avatar
Greenious
Hardware Guru
Hardware Guru
Posts: 1160
Joined: Sat Apr 24, 2004 5:39 pm
Location: Sweden

Re: MegaSTe hardware problems -help locate it please!

Postby Greenious » Mon Jul 03, 2017 11:47 pm

Fujiyama wrote: Checking the PSU could also be useful when troubleshooting. Using the Mega STe PSU (Phihong PSM-5341) schematics I've been able to determine what the wire colours of its connector correspond to. With the negative probe of my digital multimeter attached to one of the four black wires (GND) I've used the positive probe to measure each of those other pins in turn as follows:

+5V line (three red wires): +5.025V
-5V line (white wire): -5.036V
-12V line (blue wire): -12.572V
+12V line (yellow wire): +12.120V

There's also an additional (orange) wire which the PSU schematic has labelled "Power good". I have no idea what this is for but it shows +4.892V
IMHO the PSU output appears to be within specs but I suppose there's more to it than that, like spiky/noisy power, right? Is this something I can check simply by attaching my oscilloscope to GND and one of the PSU power lines? With limited knowledge of using a scope and the knowledge that you should keep away from switched PSU units unless you know exactly what you're doing I've refrained from a "trial and error" type approach here, not wanting to risk blowing up the scope.


With the appropriate probe, you hook up the oscilloscope to ground and power. What you see is the "ripple" in voltage from the PSU.

This ripple on the powerline is normal and should correspond to the frequency of your mains power (50/60Hz), and is unavoidable when you run off AC mains. (atleast with reasonably sized power supplies...)

What you want to look for is the size of the ripple, typically a good PSU should have 2% variation or less, and should not change significantly under varying load (ie floppy disk access). I reckon below 5% is still acceptable for an old PSU, but if it is more than that a recap is likely a good idea. The main problem with aging caps is increasing ripple and the inability to compensate for sudden changes in load, when recapping is absolutely necessary is a judgement call, but seeing the ripple and how it performs under varying loads helps a lot.

The "power good" signal is the PSU telling the computer that it is up and running, and the power is good and stable. This is also a common measuring point for PSUs, since caps going bad typically also needs longer to charge up, and PSU taking too long from power on until power good is signalled are generally considered failing. https://en.wikipedia.org/wiki/Power_good_signal
Updated my guides as of june 28th, 2016. Check'em out and feedback!
viewtopic.php?t=5040

siriushardware
Captain Atari
Captain Atari
Posts: 347
Joined: Thu Aug 21, 2014 7:55 pm
Location: UK

Re: MegaSTe hardware problems -help locate it please!

Postby siriushardware » Tue Jul 04, 2017 12:10 am

Sorry about all the mistakes. My copy of the diagram is seriously difficult to read so '8', 'B', and '3' are all easily confused.

If pin 11 on both LF347s (not LF357, as you say) is really positive, that's a definite problem.

Those pins (and pin 3 of the two DAC ICs, U603 and U604) are all supposed to be at minus 5 volts. You have the -5V coming out of the PSU, so where it it going missing? Can you check the voltage on both ends of R604?

Locate R604 on the diagram, (Audio section, upper left). According to me, the feed into the left hand side of R604 is minus five volts. Or is that just my terrible diagram again?

Op-Amp ICs like the LF347s, when configured the way they are in this circuit, usually need a + supply and a -supply. It looks as though the minus supply for the LF347s and the DACs should be coming from the PSU -5V through R604.

User avatar
Fujiyama
Atari Super Hero
Atari Super Hero
Posts: 580
Joined: Thu Jul 12, 2007 8:21 am
Location: Norway

Re: MegaSTe hardware problems -help locate it please!

Postby Fujiyama » Tue Jul 04, 2017 12:32 am

siriushardware wrote:Been thinking further about this: Assuming you don't have an oscilloscope or a logic probe (stop me if you do have either of those things), there is a way you can test the PC900 in-circuit and the MSTe does not need to be fully working, as long as the +5V supply is OK, and assuming that at least some of the 7406 (U301) is OK. You can make sure of that by replacing U301.

[..... snip]

This test takes advantage of the fact that the MSTe has hardware MIDI THRU. Any MIDI data received on the MSTe's MIDI input through the PC900 is retransmitted through U301F and U301E to pins 3 and 1 on the MSTe's MIDI OUT port, from where (with a suitably rewired cable) it can be sent back to the synth / keyboard.


This reminds me of a "MIDI output selector" box I built many years ago. I purely mechanical device used to select which MIDI synth should be connected to the ST's IN (for Sysex dumping) instead of reconnecting MIDI cables. I also added a software/hardware MIDI THRU switch for allowing me to control MIDI devices from a master keyboard without turning on the ST. This utilized the rather unusual THRU pin assignment in the ST and a quick read leads me to believe I already have that cable you describe with the switch in the box I made.

Tests so far lead me to believe that at least the 7406 (U301) is at fault, but it might very well be the opto-isolator as well, or maybe even the U303 (MC68850). One thing at a time and a lot of patience....

I see you've replied while I wrote this.
I'll get back to you on that but will make a second attempt to get some sleep now :lol:
Mega STe | MonSTer with dual IDE-CF memory card adapter | STe | SM-144 |NEC Multisync 1990SXi | IDE doubler | ST_ESSC | RSVE | ICD Link II | Link '97 | HD floppy drive/AJAX | HD floppy module | Minolta PCMCIA card-drive | Realtime Clock module | Discovery cartridge | Unitor-2 | Export | Combiner | Steady Eye | Human Touch | Unicorn USB

User avatar
Greenious
Hardware Guru
Hardware Guru
Posts: 1160
Joined: Sat Apr 24, 2004 5:39 pm
Location: Sweden

Re: MegaSTe hardware problems -help locate it please!

Postby Greenious » Tue Jul 04, 2017 12:33 am

I found a pic of what you should be seeing on the scope:

dc ripple.jpg


The dotted lines is the power coming out from the rectifier, the filled line what you see after your caps are done smoothing the voltage.
You do not have the required permissions to view the files attached to this post.
Updated my guides as of june 28th, 2016. Check'em out and feedback!
viewtopic.php?t=5040

User avatar
Greenious
Hardware Guru
Hardware Guru
Posts: 1160
Joined: Sat Apr 24, 2004 5:39 pm
Location: Sweden

Re: MegaSTe hardware problems -help locate it please!

Postby Greenious » Tue Jul 04, 2017 12:39 am

siriushardware wrote:Sorry about all the mistakes. My copy of the diagram is seriously difficult to read so '8', 'B', and '3' are all easily confused.


Don't worry too much about that, we have all struggled with the schematics available at one point or another. (and still do every once in a while)
Updated my guides as of june 28th, 2016. Check'em out and feedback!
viewtopic.php?t=5040


Social Media

     

Return to “Hardware”

Who is online

Users browsing this forum: No registered users and 5 guests