Jump to content

CRT Issue with Macintosh TV


Recommended Posts

This is the first time I have ever powered this up since I bought it a few months back. The original logic board needs recapping, so I put an LC575 board in it to see if the chassis was okay. It does boot, but there is a clicking accompanied by flickering of the CRT. Is this a flyback transformer issue? The board itself looks normal (from the top side - I haven't taken it out yet). Any ideas on common failure points, or am I looking at buying a new board?

Link to post
Share on other sites

Was this machine shipped?  It could be cracked solder joints on the analog board.  The last two CRT monitors I had shipped all had cracked solder joints on the flyback since that's the heaviest component on the board.

Link to post
Share on other sites
  • 68kMLA Supporter

Looks like there's an open circuit somewhere, reflowing solder joints for capacitors and main connectors would be the first thing I'd try. I don't think this is a flyback issue.

Link to post
Share on other sites
  • 68020

Sadly, my Mac TV has similar issues with the video.  I suspect it to be the flyback transformer which no longer exists.  My LC5x0 have similar issues.  I've recapped everything and checked for dry solder connections but nothing.  I've looked into making new flyback but given the cost it is unlikely that I'll be able to achieve the MOQ :(

 

Link to post
Share on other sites
7 minutes ago, maceffects said:

 I've looked into making new flyback but given the cost it is unlikely that I'll be able to achieve the MOQ :(

 

 

Flyback failure is sadly only going to become more and more prevalent over the next decade. These parts were never intended to last 30-40 years. We really need to figure out how we can actually save these old machines from a currently-inevitable, permanent death. Is it possible to create an equivalent to a flyback transformer from more modern parts, for example? What's actually inside a flyback transformer besides a pair of coils? I recently expressed my frustration at just how sparse the information is on these parts and what they actually do. I'm no dummy when it comes to electronics but the flyback has always mystified me.

 

I've tried reaching out to HR Diemen to see if they'd be willing to release/sell their old designs into the public domain but their web presence is near-zero and the forms on their decrepit site do not work. I'm not sure the company even still exists - the phone number on their site is not connected any more.

 

What we really need is a scenario like that which has played out over in the BUG Pickles thread, where the original designer of the cards was tracked down, contacted, and has now gotten involved with trying to revive the extant cards. Finding the person or persons responsible for the engineering of these transformers would be a huge leap for the community. If schematics for these parts cannot be sourced then there is a huge amount of work to be done on reverse-engineering dozens of different models. It is sad to think that the information is out there yet so far from the right hands. 

Link to post
Share on other sites
28 minutes ago, PowerMac_G4 said:

 

Flyback failure is sadly only going to become more and more prevalent over the next decade. These parts were never intended to last 30-40 years. We really need to figure out how we can actually save these old machines from a currently-inevitable, permanent death. Is it possible to create an equivalent to a flyback transformer from more modern parts, for example? What's actually inside a flyback transformer besides a pair of coils? I recently expressed my frustration at just how sparse the information is on these parts and what they actually do. I'm no dummy when it comes to electronics but the flyback has always mystified me.

 

I've tried reaching out to HR Diemen to see if they'd be willing to release/sell their old designs into the public domain but their web presence is near-zero and the forms on their decrepit site do not work. I'm not sure the company even still exists - the phone number on their site is not connected any more.

 

What we really need is a scenario like that which has played out over in the BUG Pickles thread, where the original designer of the cards was tracked down, contacted, and has now gotten involved with trying to revive the extant cards. Finding the person or persons responsible for the engineering of these transformers would be a huge leap for the community. If schematics for these parts cannot be sourced then there is a huge amount of work to be done on reverse-engineering dozens of different models. It is sad to think that the information is out there yet so far from the right hands. 

 

The flyback transformer is a critical component in the circuit. you cant just substitute it either. it requires intimate knowledge of the circuit design that only the engineer knows, or a damn good reverse-engineer. 

 

Why? because the flyback not only produces a high voltage, it also produces all the secondary scan voltages for the other circuits in the set. Vertical Sweep, B+ Boost, etc.... Including focus/screen. Also, its part of a tuned-circuit with the horizontal output stage, and all the other support components are selected around that tuning. And if that weren't enough, different flyback designs have different frequency responses and losses over frequencies. 

 

This doesn't include the x-ray protect, and high voltage regulation feedback circuitry that is connected to a tap on the flyback as well. 

 

So its not as easy as swap-n-go. I wish it were. 

 

@apm can certainly elaborate more on this subject too. 

Edited by techknight
Link to post
Share on other sites
  • 68kMLA Supporter
56 minutes ago, maceffects said:

I've looked into making new flyback but given the cost it is unlikely that I'll be able to achieve the MOQ :(

 

Just out of interest, what was the MOQ/cost, approximately?  I seem to have missed this discussion.

 

3 minutes ago, techknight said:

it requires intimate knowledge of the circuit design that only the engineer knows

 

I've always conceptualised the flyback as a box full of analogue magic and danger in approximately equal doses...

Link to post
Share on other sites
3 minutes ago, cheesestraws said:

 

I've always conceptualised the flyback as a box full of analogue magic and danger in approximately equal doses...

 

Yeah, it can be. it was even worse in the vacuum-tube type TV era. those flybacks were all specific, and they were all more intimately tuned with the horizontal output than they are "today". The horizontal output pulse only affected from center-right of the CRT. The flyback collapse/response affected center-left. Thats why this is important. 

 

Newer/more expensive designs. (like Philips RPTVs from the late aughts) had HVGs instead of Flybacks, which they were very similar but they were a self contained unit that was self-oscillating at its own frequency and was in no-way tied to the horizontal output stage, except for phase locking. These are muuuuch easier to substitute if it ever came down to it. 

 

But, flybacks that are tied into and intimately involved in the horizontal output network? nope. 

Edited by techknight
Link to post
Share on other sites

that reminds me, I have one of those big Sony trinitron (B&W G3) style monitors here that someone brought to me while back. It has a bad flyback as well. it draws too much current in higher resolutions which trigger an x-ray protect shutdown. 

 

Only getting worse. 

Link to post
Share on other sites
  • 68020
Posted (edited)
6 hours ago, cheesestraws said:

 

Just out of interest, what was the MOQ/cost, approximately?  I seem to have missed this discussion.

 

Sadly, it was not just the MOQ but some of the parts had to be custom produced outside of the flyback transformer vendor.  In terms of dollars we'd be look at $4,000-$6,000 in tooling plus $15 per unit variable cost, plus freight and tariffs, the MOQ is 500 units.  The problem is that most of the no start Perofrma/LC 5x0 series issues can be traced to flyback transformer and/or caps.  There is actually an area you can jump to force it to display video but usually it only reveals flyback issues like yours.  This is very prevalent with age and use.  I'd estimate 50% of units actually have issues.  However, given that people love to gut the 550/575 logic boards for Color Classics and that at this point people are getting most of the 5x0 machines destroyed in shipping - I can't imagine this working out.  Seems like with flyback transformers and cases we'd need to raise $50,000-$60,000 to save all future machines, sadly.  And that is with me contributing...  Seems to lose practicality as fewer and fewer are extant.  It might actually simply be too late because of too little demand now that so many are lost.  

Edited by maceffects
Link to post
Share on other sites

@maceffects I was wondering about this too just this week.  I have a Mac TV and 5x0 Macs that I would love to save and that flyback is obviously a critical component. Between the ABS getting extremely brittle and the flyback issue I really hope we can find a way to save these Macs for the future.

 

On a sidenote, maybe I'm wrong but I can't help but feel like running these Mac with the original caps may well be contributing to the demise of the flybacks.  I mean, besides some leaking and others totally drying out they aren't operating at their intended capacitance and I imagine there has to be a lot of extra noise in the circuit.

Link to post
Share on other sites

I've had a bit of a revelation.  Since I am not an EE maybe @techknight can sanity check?

 

@maceffects are you certain it's the flyback transformer?  Here is what I realized (I'll try to be as concise as possible):

 

1. Years ago I was really busy and made the mistake of sending some of my old Mac hardware off to someone (shall remain nameless) to recap and one of those boards happened to be my Mac TV analog board.  It had gone from working properly when I shipped it off to having video that looked just like what @Garrett_B posted when it returned.  I was furious and depressed and put it on a shelf until I had more time to diagnose it.

 

2. During my pandemic isolation I started doing my own recapping, and in the process was able to get multiple systems with this same exact Sony Trinitron tube (including Apple AV14 monitor, Mac TV, and an LC575) working with crisp and stable video.  In the process I noticed that all of these Apple designs use a very similar main analog board and even the component labels match.  They also share another very interesting similarity ...

 

3.  A few days ago I noticed that @maceffects had posted a list of caps for the Mac TV analog board but there was, what I thought, a huge missing piece of data: the type of capacitor.  Then it hit me.  The nameless individual I referenced above [1] had used low-quality general purpose (GP) caps for everything on the logic board.  But in my work with all of these analog boards I've been soldering on Apple specifically used low impedance (LI) caps for a number of the ones in the CRT circuit.  Consistently from board to board specific ones, even through changes in vendor (Nichicon, UCC, Rubycon), all of these specific caps were LI designs.  One of those LI caps in particular, CF8, is known to cause unstable video and distortion when it drifts off spec due to leakage or other issue.  Replacing it with a correct one restores proper video.  

 

I am not an electronics expert (I know exactly enough to be dangerous) but I do know that CRT circuits have a lot of complicated inter-related feedback mechanisms that expect the correct resistance and capacitance in a circuit in order to work properly.  Is it not reasonable to think that maybe caps of the wrong impedance value (or ESR) - either because they have failed due to age, or picking the wrong replacement cap - are causing or at least contributing to this problem?  

 

Maybe I am totally wrong and the flybacks really are the universal cause of this problem, but my anecdotal experience with 1, 2, and 3 make me think that maybe there is a far more simple solution we are overlooking here.  At a minimum maybe the flyback transformer is responsible for fewer of the problems than we blame them for.

 

Edited by eraser
Link to post
Share on other sites

Today I restored yet another LC 5x0 Mac to properly working video.  No flyback swap needed.  I also found references to some TV repair information that discusses the importance of proper impedance on circuits in the system:

 

https://www.industrial-electronics.com/electronic-srvc_mag_1979-04_bad.html

 

It's also worth pointing out that the rated capacity of LI capacitors to suppress ripple is typically considerably higher than that of the GP counterparts.  If the circuit running these CRTs is subject to significant ripple then that would suggest that a LI cap would really be necessary for stable video.  I've also noticed that most of the caps I am referencing here were covered in adhesive by Apple during production.  This implies that as designed they are indeed likely intended to experience significant ripple during normal operation, enough that they are likely to physically vibrate.  The adhesive arrests the vibration.

 

If anyone wants to experiment then I will even give you the exact part numbers I've had success with.  This is the product of much research.  These same caps should work for all LC5x0 Macs, including the TV.  Most of them will even work for a Color Classic (except CL14).

 

CF8 - Nichicon UHW Series / 3300uF 16V 105*C / Part Number UHW1C332MHD6

CL14 - Nichicon UPM Series / 47uF 100V 105*C / Part Number UPM2A470MPD1TD (LC5x0 only)

CL21 - Nichicon UHW Series / 100uF 100V 105*C / Part Number UHW2A101MPD1TD 

CP12 - Nichicon UHW Series / 100uF 100V 105*C / Part Number UHW2A101MPD1TD

CP39 - Panasonic FM Series / 3300uF 16V 105*C / Part Number EEU-FM1C332L (Needs to lay horizontally)

CP40 - Panasonic FM Series / 5600uF 10V 105*C / Part Number EEU-FM1A562B 

CP41 - Nichicon UHW Series / 1000uF 16V 105C / Part Number UHW1C102MPD

CP42 - Panasonic FM Series / 5600uF 10V 105*C / Part Number EEU-FM1A562B

CP46 - Nichicon UHE Series / 1000uF 10V 105*C / Part Number UHE1A102MPD

CP56 - Panasonic FM Series / 3300uF 16V 105*C / Part Number EEU-FM1C332L (Needs to lay horizontally)

 

Link to post
Share on other sites

Thanks everyone for the great discussion and tips. I'm going to try a few more troubleshooting steps later and see if I can figure anything else out. This computer has never been shipped - bought it locally, but I suppose cracked joints could still exist. I tried to take the analog board out yesterday, but it seems the edge connector is really stuck. I can move it back and forth a little, but it's clearly hung up on something. I'll report back when I can see the condition of the joints under the transformer.

 

As for the logic board, I was testing using a 575 board, not the original TV logic board (haven't had time to recap it yet). Could this be the issue? If it's picky about the 575 board, I do have a freshly recapped Color Classic board I can throw in (tested working fine). Are they compatible?

 

On an unrelated note, buying capacitors recently has been terrible. A couple months ago I should have just doubled my order quantities. Thousands in stock at DigiKey previously and now 50 week lead times? Yikes!

Link to post
Share on other sites

Interesting thoughts about the impedance of capacitors.  Do keep us posted if you learn more!  I always wonder if I’m putting in capacitors that are too high or low impedance.  I know @JDW found that there is a capacitor inside the Sony SE/30 power supply that will cause power output with higher amounts of ripple if the impedance of that one capacitor is not correct.  It was in one of his YouTube videos... don’t remember which one.

 

FWIW, I use Nichicon PW almost exclusively now for my Mac recaps.  It comes in sizes and dimensions very similar to the original caps I am replacing, has good temperature and hour ratings, and is designated as “Low Impedance for Switching Power Supplies” on the Nichicon data sheet.  I have done my LC5xx analog board, several Compact Mac analog boards, and an iMac G3 analog board with mostly Nichicon PW and haven’t had any problems yet.  Based off my experiences, it seems to be a good series for us Mac restorers.

Edited by 68krazy
Link to post
Share on other sites
  • 68020

@eraser I'll have to validate that, however, it had the same issues before recapping.  I cannot speak for all machines, but merely my fairly large sample.  Given the similarities, I'd be this is an issue for the Color Classic as well but I only had a few of those and 2 I gave up on a couple years ago.  

Link to post
Share on other sites
2 hours ago, maceffects said:

I'll have to validate that, however, it had the same issues before recapping.  I cannot speak for all machines, but merely my fairly large sample.  Given the similarities, I'd be this is an issue for the Color Classic as well but I only had a few of those and 2 I gave up on a couple years ago.

 

Yeah, but that still fits my theory.  In this case new caps and the old and failed caps would show roughly the same behavior. Whether they are new or old they are not suppressing enough ripple current for the video to be stable.  I really hope I'm right about this.  This just doesn't seem like a flyback issue to me.

 

7 hours ago, 68krazy said:

 I know @JDW found that there is a capacitor inside the Sony SE/30 power supply that will cause power output with higher amounts of ripple if the impedance of that one capacitor is not correct.  It was in one of his YouTube videos... don’t remember which one.

 

Right.  You absolutely always want to match LI caps in a switching power supply.  In this case though, there should be a considerable amount more ripple in the CRT circuit than that in a power supply.  The difference being that in a power supply the ripple is higher (mains) voltage.  

 

8 hours ago, Garrett_B said:

As for the logic board, I was testing using a 575 board, not the original TV logic board (haven't had time to recap it yet). Could this be the issue? If it's picky about the 575 board, I do have a freshly recapped Color Classic board I can throw in (tested working fine). Are they compatible?

 

On an unrelated note, buying capacitors recently has been terrible. A couple months ago I should have just doubled my order quantities. Thousands in stock at DigiKey previously and now 50 week lead times? Yikes!

 

The logic board won't change this behavior at all, so no, that's not the issue.  You can swap all of those logic boards you mentioned.  The only one you can't swap is an LC580 - despite its name the design is very different than the other LC5xxs.

 

Yeah.  I use Mouser for my caps and have been running into stock shortages there too - as well as long lead times.  Luckily they have a giant selection so I can usually find a suitable alternative even if the ones I really want aren't available. 

Link to post
Share on other sites
On 6/8/2021 at 12:12 AM, 68krazy said:

I know @JDW found that there is a capacitor inside the Sony SE/30 power supply that will cause power output with higher amounts of ripple if the impedance of that one capacitor is not correct.  It was in one of his YouTube videos... don’t remember which one.

 

Actually, the problematic capacitor is found in the Sony PSU used in the Apple HDSC-series external hard drive enclosures, as described in my video here at 39:01:

 

 

And here is the detailed spreadsheet I made comparing the capacitor data:

https://docs.google.com/spreadsheets/d/1ep3kz2229ORI6euaM0qNeN_txZiJhhvH7oje3TEY_dI/edit?usp=sharing

 

Link to post
Share on other sites

I had no idea there were this many potential differences in capacitors of the same uF and voltage rating - crazy. I used to think that any replacement would be better than the existing one simply because it was new, but that doesn't seem like it's the case.

 

At any rate, I was able to remove the AB from the Mac after releasing the two tabs near the front (didn't see them right off the bat). The board is dusty, but it's very clean with no signs of cap leakage anywhere. Obviously this doesn't mean all the caps are good, but it's a start. I really can't find any smoking gun, other than what looks like a crack on the flyback? What do you guys think - normal or not? Still debating on whether I should just blindly replace all the caps or go about repairing another way.

b.jpg

c.jpg

d.jpg

Link to post
Share on other sites
On 6/5/2021 at 12:48 AM, eraser said:

I've had a bit of a revelation.  Since I am not an EE maybe @techknight can sanity check?

 

@maceffects are you certain it's the flyback transformer?  Here is what I realized (I'll try to be as concise as possible):

 

1. Years ago I was really busy and made the mistake of sending some of my old Mac hardware off to someone (shall remain nameless) to recap and one of those boards happened to be my Mac TV analog board.  It had gone from working properly when I shipped it off to having video that looked just like what @Garrett_B posted when it returned.  I was furious and depressed and put it on a shelf until I had more time to diagnose it.

 

2. During my pandemic isolation I started doing my own recapping, and in the process was able to get multiple systems with this same exact Sony Trinitron tube (including Apple AV14 monitor, Mac TV, and an LC575) working with crisp and stable video.  In the process I noticed that all of these Apple designs use a very similar main analog board and even the component labels match.  They also share another very interesting similarity ...

 

3.  A few days ago I noticed that @maceffects had posted a list of caps for the Mac TV analog board but there was, what I thought, a huge missing piece of data: the type of capacitor.  Then it hit me.  The nameless individual I referenced above [1] had used low-quality general purpose (GP) caps for everything on the logic board.  But in my work with all of these analog boards I've been soldering on Apple specifically used low impedance (LI) caps for a number of the ones in the CRT circuit.  Consistently from board to board specific ones, even through changes in vendor (Nichicon, UCC, Rubycon), all of these specific caps were LI designs.  One of those LI caps in particular, CF8, is known to cause unstable video and distortion when it drifts off spec due to leakage or other issue.  Replacing it with a correct one restores proper video.  

 

I am not an electronics expert (I know exactly enough to be dangerous) but I do know that CRT circuits have a lot of complicated inter-related feedback mechanisms that expect the correct resistance and capacitance in a circuit in order to work properly.  Is it not reasonable to think that maybe caps of the wrong impedance value (or ESR) - either because they have failed due to age, or picking the wrong replacement cap - are causing or at least contributing to this problem?  

 

Maybe I am totally wrong and the flybacks really are the universal cause of this problem, but my anecdotal experience with 1, 2, and 3 make me think that maybe there is a far more simple solution we are overlooking here.  At a minimum maybe the flyback transformer is responsible for fewer of the problems than we blame them for.

 

 

Any time i recap an SMPS circuit or anything with high frequencies, i automatically order low impedence/ESR capacitors, its just habit for me. So this is something i never thought of during the event of NOT having the right caps. 

Link to post
Share on other sites
8 hours ago, techknight said:

Any time i recap an SMPS circuit or anything with high frequencies, i automatically order low impedence/ESR capacitors, its just habit for me. So this is something i never thought of during the event of NOT having the right caps. 

 

Right!  That's exactly why I was shocked by these cases where the impedance characteristics of the original cap wasn't addressed with the replacement.

 

9 hours ago, Garrett_B said:

The board is dusty, but it's very clean with no signs of cap leakage anywhere.

 

No visible signs.  I can tell you that while I was recapping one of these exact boards I was in the middle of desoldering a cap and was totally surprised by a loud hiss.  It was startling enough that I almost dropped the board.  When I investigated I found that the cap I was working on was sitting in a pool of its own goo and the hissing sound was that electrolyte boiling and the gas escaping from under the cap.  I hadn't visibly noticed any leak before I started my work.  I would not be surprised if you find that once you start your work you find some similar "pools."

 

9 hours ago, Garrett_B said:

I really can't find any smoking gun, other than what looks like a crack on the flyback? What do you guys think - normal or not? Still debating on whether I should just blindly replace all the caps or go about repairing another way.

 

I can't really tell by that photo.  Do you have a magnifying glass that you can use to inspect it?  Does it look like an actual crack in the housing?  

Link to post
Share on other sites
9 minutes ago, eraser said:

I was shocked by these cases where the impedance characteristics of the original cap wasn't addressed with the replacement.

 

You don't always know the exact ESR/impedance characteristics of the old caps because you can't always find a datasheet for very old caps, and measuring the current failing cap (with a great LCR meter like the DE-5000) only tells you the current characteristics, which are no doubt way off from what those were at the time it was new.  We therefore must often do cap replacements with trial and error.  The good news is that mostly applies to caps being replaced on switching power supplies, rather than caps being replaced on a motherboard.

 

In some switching power supply designs, especially in the early days of switchers, it was not uncommon for the design to require a certain minimum capacitor ESR on the output stage.  It seems counterintuitive because normally you want less ESR on the output of a power supply to keep ripple to a minimum, but it's sometimes necessary.  Even some switcher ICs today require it.  That's you will will sometimes you will find a ceramic cap in series with a small resistance value on a power supply output.  I considered that when experimenting with the troublesome cap on the Sony PSU for the HD20SC drive enclosure, but ultimately a single replacement capacitor with a satisfactory ESR value (which I figured out by testing with different caps) is the best approach because it's a convenient drop-in component -- no mod required.

Link to post
Share on other sites
17 hours ago, JDW said:

You don't always know the exact ESR/impedance characteristics of the old caps because you can't always find a datasheet for very old caps, and measuring the current failing cap (with a great LCR meter like the DE-5000) only tells you the current characteristics, which are no doubt way off from what those were at the time it was new.  We therefore must often do cap replacements with trial and error.  The good news is that mostly applies to caps being replaced on switching power supplies, rather than caps being replaced on a motherboard.

 

I totally agree.  For this particular system many of the analog boards were manufactured around 1993-1995 and although the capacitor series have been discontinued the data sheets are still available online.  Before choosing a replacement I looked up each cap to try and figure out what the original design intent seemed to be.  It's very time intensive but a bit rewarding.  

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...