PDA

View Full Version : Power Supply dead



Vifa
December 28th, 2008, 07:33 AM
Power supply is dead, what to do?

Can it be repaired?

Dwight Elvey
December 28th, 2008, 07:40 AM
Power supply is dead, what to do?

Can it be repaired?

yes
dwight

patscc
December 28th, 2008, 08:11 AM
Power supply to what ?
patscc

Vifa
December 28th, 2008, 08:17 AM
The power supply to an Apple II. I think the computer is from 1978. Nothing happens at all when I power it up, so I assume the power supply is dead.

tezza
December 28th, 2008, 09:03 AM
Check the voltages, when the power supply is plugged into the motherboard and under load. See if they are correct.

Details are in redbook.pdf (page 110), which can be downloaded from http://apple2history.org/dl/dl.html

Tez

channelmaniac
December 28th, 2008, 09:53 AM
There are 2 versions of the Apple II+/IIe power supply - an Astec and non-Astec model.

Please post which one you have.

The non-Astec tends to need a cap in the feedback section and a 22 ohm 1/2w resistor replaced.

The Astec one tends to need a 220uf @10v 85 deg C cap replaced in the feedback section as well... (C7)... Replace it with a 220uf @16v 105 dec C cap.

BEWARE: The non-Astec supply DOES NOT have bleeder resistors across the main filter caps. If they are charged it takes a LONG time to bleed off and represents a major shock hazard. It's DC voltage which will cause your muscles to clamp down, making it hard to release that board.

Yes, I had 2 scars on my hand that took about 9 years to fade out from one of those non-Astec ones.

If those aren't the fix, let me know. I've repaired a lot of those over the years when I used to work for a grey market Apple repair shop.

RJ

Vifa
December 28th, 2008, 10:00 AM
A little label on the side says "ASTEC".

channelmaniac
December 28th, 2008, 10:01 AM
I almost forgot...

Try this thread:

http://newlifegames.net/nlg/index.php?topic=23.0

It's a thread in my repair logs on repairing switching power supplies.

RJ

channelmaniac
December 28th, 2008, 10:12 AM
OK.

Turn it on and listen carefully. Put your ear up to it. Do you hear it making a soft chirping sound? If yes, then you have a bad cap in the regulation part or you have a shorted diode on the output section.

If not, you'll have to troubleshoot it further.

You'll need to drill out two small rivets on the supply to open it up. If they are gone then someone else has already worked on it.

Open it and check the fuse. If the fuse is blown then you probably have a shorted bridge rectifier that needs replacing. Check it for shorts with a digital multimeter on the continuity setting.

It's pretty rare for the TO-3 case (big!) transistor to short out. Generally it's the bridge rectifier.

If you don't want to repair it, I can do it for you for a nominal price.

RJ

Vifa
December 28th, 2008, 10:14 AM
Hey, thanks - I will start reading it now.

Btw, I just spotted that the power supply gets warm.

Druid6900
December 28th, 2008, 08:45 PM
Hey, thanks - I will start reading it now.

Btw, I just spotted that the power supply gets warm.

Warm is OK, there is no fan. It's planned that way because it keeps the ROMs from icing up.

Supernovae, on the other hand, is very, very bad.

patscc
December 28th, 2008, 09:25 PM
Vifa said...Btw, I just spotted that the power supply gets warm.

Sure wish I could see infrared.
patscc

(Sorry, bad jole)

Terry Yager
December 29th, 2008, 03:18 PM
Sure wish I could see infrared.
patscc

(Sorry, bad jole)

Bad spelling, too...shouldn't that say 'bad joule'?

--T

patscc
December 29th, 2008, 08:12 PM
That's almost as smelly as Lutfisk.
Good one. :)
patscc

Vifa
January 4th, 2009, 05:32 AM
OK.

Turn it on and listen carefully. Put your ear up to it. Do you hear it making a soft chirping sound?



I don't think I can hear anything when I turn it on/off with the switch; but I can for sure hear something when I plug the power cord into it. Hard to explain what it sound like, not something big, but something is making a sound.




If you don't want to repair it, I can do it for you for a nominal price.



I really appreciate that, but unfortunately I live in Denmark and was hoping to find someone nearby me, it will be easier that way.

Anyway, I can't seem to find anyone who can fix this, so I will keep you in mind. In case I don't find anyone at all, can I then send the power supply to you by mail?

channelmaniac
January 4th, 2009, 06:05 AM
Yep.

If yours is the model with rivets on the side, drill or cut them out, take the screws out, and disassemble it. That way you can send just the board to save on shipping costs.

Before you do that, can you post a picture of the inside of the power supply? There are a couple of parts you can easily replace to try to fix it first.

RJ

Vifa
January 4th, 2009, 11:28 AM
Yep.

If yours is the model with rivets on the side, drill or cut them out, take the screws out, and disassemble it. That way you can send just the board to save on shipping costs.

Before you do that, can you post a picture of the inside of the power supply? There are a couple of parts you can easily replace to try to fix it first.

RJ

Hey there again. I have just removed the power supply, and tried to open it. Here comes some pics.

Vifa
January 4th, 2009, 11:36 AM
Funny thing, I found a date on the bottom of it - it reads "MAR 9 1979". That means it will soon turn 30, that just settle a deadline for my project; my Apple has to be up running again before so I can celebrate the birthday in a proper way...:)

Anyway, I have removed all the screws in both sides, it looked like the bottom of the power supply then would come off. I can move the bottom panal a very little bit up and down in one of the sides, but I can't seem to remove it completly? Do I have to remove the rivets too? - even though the rivets don't seem to hold the bottom.

Do you have a picture of a power supply which has been opened?

Vifa

channelmaniac
January 4th, 2009, 11:43 AM
Drill out the rivets or use a small set of side cutters to cut the heads off so you can open the supply.

Vifa
January 4th, 2009, 11:46 AM
Thanks, I will keep on the work tomorrow, and do a litte update with pics here when I get so far.

tezza
January 4th, 2009, 12:31 PM
Funny thing, I found a date on the bottom of it - it reads "MAR 9 1979". That means it will soon turn 30, that just settle a deadline for my project; my Apple has to be up running again before so I can celebrate the birthday in a proper way...:)


That's about the same vintage as my Apple II Europlus.

Those power supplies can be difficult to get into. I recently got hold of a lot of old apples and apple clones. Thankfully all the geniune Apple power supplies were working. Some of the clone PSUs were not however. I tried to disassemble them, but it proved no easy task. I had to destroy the rivets.

IIn the end I decided I had enough working PSUs so I didn't need to fix the broken ones. Besides that they were very rusty and corroded.

Good luck with the repair.

Tez

channelmaniac
January 4th, 2009, 04:32 PM
Tez,

Generally they are easy repairs...

Astec unit: Replace 1 cap in the feedback section
Non-Astec unit: Replace 1 cap and 1 resistor in the feedback section

Beyond that, it's a matter of simply looking for shorted diodes - both the bridge rectifier or the low voltage side diodes. If the fuse is blown, it's the bridge rectifier.

Those tidbits will fix 95+% of the Apple II+/IIe power supplies.

RJ

tezza
January 4th, 2009, 05:31 PM
Tez,

Generally they are easy repairs...

Astec unit: Replace 1 cap in the feedback section
Non-Astec unit: Replace 1 cap and 1 resistor in the feedback section

Beyond that, it's a matter of simply looking for shorted diodes - both the bridge rectifier or the low voltage side diodes. If the fuse is blown, it's the bridge rectifier.

Those tidbits will fix 95+% of the Apple II+/IIe power supplies.

RJ

Thanks. Good to know if I ever come across any more, or if my existing ones start to give me trouble.

Tez

Ole Juul
January 4th, 2009, 05:34 PM
channelmaniac: Generally they are easy repairs...
Astec unit: Replace 1 cap in the feedback section
Non-Astec unit: Replace 1 cap and 1 resistor in the feedback section ... RJ
Now those are indeed interesting tidbits. I haven't really groked switching supplies. Do you have a link or general diagram to help in finding or identifying the feedback section and those specific parts if they aren't obvious? There's probably others here that would be interested too. :)

patscc
January 4th, 2009, 05:44 PM
Ole Juule asked...groked switching supplies
Try
http://www.national.com/an/AN/AN-556.pdf

It's a good overview/intro, and then, once you grok how they work & the different types, you can just dig through app notes/datasheets on the National site for circuits.

patscc

Vifa
January 5th, 2009, 02:51 AM
I tried to disassemble them, but it proved no easy task. I had to destroy the rivets.


Good luck with the repair.

Tez

Thank you. I will continue on my work tonight. Yes, I am affraid I also have to drill out the rivets too - does anyone know if I can destory anything inside the box when I drill those out?

channelmaniac
January 5th, 2009, 05:48 AM
There are components inside that could be damaged if you drill too far in. I like to use a small set of electronics type diagonal cutters to slide under the rivet head and cut it. It'll leave some scarring around where the rivets were but is much safer on the components.

RJ

Vifa
January 5th, 2009, 07:42 AM
Hey there again.

I just drilled out the rivets, turned out it was not even nessasary to remove the rivets at all. Nevertheless, the box is open now!..:)

I have just taken som pictures of the inside:

channelmaniac
January 5th, 2009, 07:52 AM
Wow... I've worked on > 100 of those supplies and I've never seen that layout before.

Can you take the board out and take pictures of the bottom so I can see where the traces go?

Vifa
January 5th, 2009, 08:27 AM
Yes, I will do that.

Remember, this Apple was bought in Europe, there might be some differences beetween layout Europe and US between.

Vifa
January 5th, 2009, 09:20 AM
Wtf, it seems like the box has been opened before.

Try to take a look at the pictures I just have taken. The on/off switch on the back, is not connected to anything at all inside the box. We have had this computer since new, except for 5-10 years where a family member used it; or actually, I think she used it very little. Anyway, since we got it back again it has not been working.

I maybe think the switch may has been bad, and someone has tried to fix this so the power was always on. Anyway, try to look at pictures.

Vifa
January 5th, 2009, 11:27 AM
Wow... I've worked on > 100 of those supplies and I've never seen that layout before.

Can you take the board out and take pictures of the bottom so I can see where the traces go?

I tried to remove the board, but it is impossible, or at least it is without cutting anything at first. But I tried to disassemble it as much as I could. Here are some final pictures of it. I tried to take pictures of how the power was running from the power input.

I don't know what to do now?

channelmaniac
January 5th, 2009, 02:24 PM
Should be simple to remove. Unplug the power cables from the board, crimp the big black plastic insert holding the wires with a pair of pliers, and pull it out from the hole. Once it's out of the hole you should be able to remove 6 screws holding the board in place and lift it right out.

Raymond

mwillegal
January 6th, 2009, 04:13 AM
The usually way to remove rivets, is to estimate size of the hole that the rivet passes through. Drill straight through center of the head of the rivet until the head is separated from the shaft. You will end up with the flange of the rivet head on the drill bit shaft and the shaft of the rivet can then be pushed or will fall through the hole. There should be no damage to the hole itself or surrounding surface and it can be easily restored with a new rivet to like new status. If I remember right, the one Apple power supply I drilled the rivets out of, had a 1/8 inch hole.

Regards,
Mike Willegal

Vifa
January 6th, 2009, 06:50 AM
The usually way to remove rivets, is to estimate size of the hole that the rivet passes through. Drill straight through center of the head of the rivet until the head is separated from the shaft. You will end up with the flange of the rivet head on the drill bit shaft and the shaft of the rivet can then be pushed or will fall through the hole. There should be no damage to the hole itself or surrounding surface and it can be easily restored with a new rivet to like new status. If I remember right, the one Apple power supply I drilled the rivets out of, had a 1/8 inch hole.

Regards,
Mike Willegal

Hey, thanks. I drilled the rivets out yesterday. As mentioned in one of my earlier post, somebody has repaired the power supply before me. But I assume there have been 5 rivets on each side of the box to hold the bottom; in my case there was no rivets, but screws instead. Nevertheless, there was still 4 rivets, 2 placed on each side. Unfortunately it was not necessary to remove these 4 rivets, but I first figured that out when I had done it.You can see at the picture where the screws and rivets was.

tezza
January 6th, 2009, 09:53 AM
Yes, rivets are a pain in the ****. I don't know why Apple (or whoever built the power supplies) didn't use screws. Obviously they didn't expect people to attempt a repair even though, as has been mentioned, it's generally quite straightforward.

Getting into the box is often the hardest part!

Tez

nige the hippy
January 6th, 2009, 10:10 AM
Wow... I've worked on > 100 of those supplies and I've never seen that layout before.


I have - probably a European variation -

will read further but do the inlet fuse, the bridge rectifire (;)) and the cap next to it look a bit crispy. EDIT - no sorry, must have been shadows on the photo.

channelmaniac
January 6th, 2009, 02:34 PM
Well,

If you took that pic before you opened it then someone else has already been in it.

There are 4 screws along each side of the bottom to remove the cover. In the middle there are empty holes. That is where the rivets were that you were to remove.

RJ

mfortuna
January 6th, 2009, 05:53 PM
This picture illustrates what was done to bypass the switch. The black wire and red wire are spliced together. They should go to the switch instead. But with the switch bypassed, the power supply should still work as long as it is plugged in.

http://www.vintage-computer.com/vcforum/attachment.php?attachmentid=1136&d=1231179509

Mike

Vifa
January 14th, 2009, 06:17 AM
Can you take the board out and take pictures of the bottom so I can see where the traces go?

There you go. I have just taken some pictures of the bottom.

Vifa
January 14th, 2009, 06:21 AM
I found somebody nearby me who maybe could fix the power supply. I turned in the power supply last week, and I got it back today. I was told the power supply wasn't dead, and it seemed like it worked fine.
Nevertheless, the 'light button' on the keyborad doesn't light up, and the two floppy disks I have doesn't spin either.

So what is then wrong?


Vifa

channelmaniac
January 14th, 2009, 07:09 AM
Well, now...

It's time for you to break out the multimeter and check the power supply when it's plugged into the computer and turned on.

Set the multimeter to DC volts. If it isn't autoranging then choose the 20volt setting.

On the 4116 DRAM chips pin 16 is ground. Put your black lead on it. Put the red lead on pin 1 and you should read -5v. Put it on pin 8 and you should read +12v. Finally, on pin 9 you should read +5v.

If one of these is missing then the power supply or power jack is messed up.

If they ALL are missing then you may have a short on the motherboard or an expansion card. Unplug all cards one by one and see if the system powers up. When you find the bad card, report back here.

If all the cards are out and it's still dead, you may have a short on the motherboard.

To check for shorts... use a DIGITAL multimeter. DO NOT USE ANALOG. Put it on the continuity test function.

Go back to the 4116 DRAM chip. Put the black lead on pin 16. Put the red lead on pin 1 then 8 then 9. If you have less than 10 ohms to ground then you'll have a shorted part on your hands. Most likely it will be one of those tiny capacitors on the +12v line and your 12v line will show a short. I've seen that before.

RJ

Vifa
January 15th, 2009, 11:39 AM
Hey there. I finally got a voltmeter.

I assume the 4116 DRAM chips are the ones inside the red box I have drawn in the attached picture, right?

I am not sure which pin is number 16 and which one is number 1, I tried to find a datasheet but with no luck. How can I tell which pin is 16 and which one is 1?

I tried to put the black lead and the red lead on what I thought was pin 16 and pin 1, but I am not sure it was, something strange 'clicking' sounds/noise then came from the power supply. Not loud, but you could easily hear them.

Before I go further on, please help me find out where pin 16 and 1 is.

Vifa

per
January 15th, 2009, 11:53 AM
Hey there. I finally got a voltmeter.

I assume the 4116 DRAM chips are the ones inside the red box I have drawn in the attached picture, right?

I am not sure which pin is number 16 and which one is number 1, I tried to find a datasheet but with no luck. How can I tell which pin is 16 and which one is 1?

I tried to put the black lead and the red lead on what I thought was pin 16 and pin 1, but I am not sure it was, something strange 'clicking' sounds/noise then came from the power supply. Not loud, but you could easily hear them.

Before I go further on, please help me find out where pin 16 and 1 is.

Vifa



[--U--]
Pin1 -[ ]- Pin16
-[ ]-
-[ ]-
-[ ]-
-[ ]-
-[ ]-
-[ ]-
Pin8 -[ ]- Pin9
[-----]

channelmaniac
January 15th, 2009, 11:56 AM
There's a notch on the chip to help you identify pin 1.

If you are hearing a clicking sound coming from the power supply then there is a short somewhere on the board. Skip checking the voltages and jump right into testing continuity to see which power line has the short.

Do the continuity tests with the computer OFF.


16 9
--------
) |
--------
1 8

Vifa
January 15th, 2009, 12:15 PM
All right then.

Now when I know where pin 1 is located, I have tried to do what you (channelmaniac) told me to do - and the situation is as:

---------------------------------------------------------

Black lead on pin 16 and red lead on pin 1: Nothing

Black lead on pin 16 and red lead on pin 8: it sparks, and the clicking noise from the power supply appears again.

Black lead on pin 16 and red lead on pin 9: it sparks, and the clicking noise from the power supply appears again.

---------------------------------------------------------

I have tried with other of the chips and the situation is the same.

patscc
January 15th, 2009, 12:21 PM
You're sure you're checking continuity (resistance) and not current, right ? And with the power supply unplugged from the board ?
Sparks are not good when testing with a multimeter.
patscc

per
January 15th, 2009, 12:21 PM
All right then.

Now when I know where pin 1 is located, I have tried to do what you (channelmaniac) told me to do - and the situation is as:

---------------------------------------------------------

Black lead on pin 16 and red lead on pin 1: Nothing

Black lead on pin 16 and red lead on pin 8: it sparks, and the clicking noise from the power supply appears again.

Black lead on pin 16 and red lead on pin 9: it sparks, and the clicking noise from the power supply appears again.

---------------------------------------------------------

I have tried with other of the chips and the situation is the same.
Note that not all of the IC's are using pin 8 and 9 as power sources. However, I'm not sure of the pinouts of the rest of them.

Vifa
January 15th, 2009, 12:26 PM
You're sure you're checking continuity (resistance) and not current, right ? And with the power supply unplugged from the board ?
Sparks are not good when testing with a multimeter.
patscc

No, so far the multimeter has been on 20v.

But I will do the continuity testing right now with the power supply unplugged.

patscc
January 15th, 2009, 12:42 PM
pin 1: -5 V
pin 8: 12 V
pin 9: +5 V
pin 16: Ground.

You should measure any voltages with respect to ground. Preferably by clipping the black lead with an aligator clip to system ground somewhere, this way you don't have to balance two probe tips against the ic.

The rest of the pins:

pin 2: Data in
pin 14: Data out
(remember, it's organized 16k x 1)
pin 5: Address 0 (lsb)
pin 7: A1
pin 6: A2
pin 12: A3
pin 11: A4
pin 10: A5
pin 13: A6

pin 3: Write enable
pin 4: Row address strobe
pin 15: Column address strobe

Where they hefty sparks, or real tiny ones ?

post if you have troubles tracking down pinouts.

patsc

Vifa
January 15th, 2009, 01:04 PM
pin 1: -5 V

Where they hefty sparks, or real tiny ones ?

patsc

Most of them were tiny ones, but there were also some sparks that you might call a little bit hefty. Is this serious? I mean, can any damage have been done?

Anyway, there is something wrong with my multimeter...:( I will borrow another one tomorrow; but the question is then if I shall start all over with putting the multimeter on 20v (even though there were sparks from pin 8 and 9 when I just tried with the bad working multimeter) and start all the testing again?

So far, I have had the multimeter on 20v and tested pin 1,8 and 9 with the black lead on pin 16 and the power supply plugged in. Then I have had the multimeter on the continuity where I couldn't read anything at all, probably because of the bad multimeter I have borrowed.


I really appreciate you effort to help me, so far thanks to all!...:)

Vifa

channelmaniac
January 15th, 2009, 01:33 PM
Are you sure you are using your meter correctly?

If you are on the 20A scale instead of 20V you'll short things and get sparks.

Also, if you have 3 holes for leads on your meter, make sure you are using the +DC volt one and not the DC Amp one or you'll short things there too. (My Radio Shack meter is like this)

RJ

channelmaniac
January 15th, 2009, 01:35 PM
Also, it sounds like your power supply is working if you are seeing sparks. ;)

Do you have a logic probe? If you have nothing on the screen but yet have power, I'd suspect a bad crystal. It's not too uncommon on the old Apples. In the USA it's a 14.31818MHz crystal.

RJ

patscc
January 15th, 2009, 02:19 PM
Can you post a picture of your multimeter, the way you had it set up ?
It really does sound like you had it on a current seeting (uA , mA or A ), or the leads connected incorrectly. That it suddenly has a problem supports this thought, since current inputs are typically fused, and it sounds like you might have blwon the fuse.
patsc

Vifa
January 16th, 2009, 07:54 AM
The multimeter I have borrowed is one of those with three holes, and I had probably plugged the leads wrong (it is working now ;)). The picture shows how it is plugged now.

I just did some testing on the resistence, I had the multimeter on '2K'.


The black lead on pin 16 and the red lead on...

---------------
pin 1: 0,725
pin 8: 1,1
pin 9: 0,178
---------------



Vifa

patscc
January 16th, 2009, 08:26 AM
Ok, now that you've go it set up, change the range to 20V and plug in the pwr supply and measure the pin voltages. Remember, use a clip or something to clip the black lead to ground.

patscc

Vifa
January 16th, 2009, 11:57 AM
Ok, now that you've go it set up, change the range to 20V and plug in the pwr supply and measure the pin voltages. Remember, use a clip or something to clip the black lead to ground.

patscc


I have just done it with the 20V and power supply plugged in:

With the black lead on pin 16 and the read lead on...

---------------
Pin 1: (-5,7) - (-6,3)
Pin 8: 11,8 - 12
Pin 9: 4,95
---------------


Vifa

channelmaniac
January 16th, 2009, 12:12 PM
Hmmmm... that -5v is way out of whack.

So you have power, but you have nothing on the screen? And the computer doesn't beep when you turn it on?

What are you using for a monitor?

RJ

patscc
January 16th, 2009, 12:33 PM
Pin 1 is actually above the spec for the part.
Ok, now, set your multimeter to AC( 20 V range ), and let's see if perhaps you've got excess ripple coming through on one of the supply legs.
When you post the results, make sure you specify if it reads mV or V

patscc

Vifa
January 16th, 2009, 12:34 PM
My monitor is an Apple /// Monitor.

But nothing seems to happen when the power is on. No light in the 'light button', the floppy disks don't spin and the monitor shows nothing.

channelmaniac
January 16th, 2009, 12:38 PM
Pin 1 is actually above the spec for the part.
Ok, now, set your multimeter to AC( 20 V range ), and let's see if perhaps you've got excess ripple coming through on one of the supply legs.
When you post the results, make sure you specify if it reads mV or V

patscc

Problem is that it's a switching supply and his voltmeter may not even read the ripple as the frequency will be very high.

channelmaniac
January 16th, 2009, 12:40 PM
My monitor is an Apple /// Monitor.

But nothing seems to happen when the power is on. No light in the 'light button', the floppy disks don't spin and the monitor shows nothing.

Well, you have voltage to the board. Here's where you need a logic probe.

Replace the crystal and see what you get. You should at LEAST have a bunch of garbage on the screen at this point. The clock circuit is probably down.

Don't go by the light on the keyboard as they can burn out. If you have no clock circuit then I doubt the floppy drives would even light up at this point.

Vifa
January 16th, 2009, 12:54 PM
Well, you have voltage to the board. Here's where you need a logic probe.

Replace the crystal and see what you get. You should at LEAST have a bunch of garbage on the screen at this point. The clock circuit is probably down.

Don't go by the light on the keyboard as they can burn out. If you have no clock circuit then I doubt the floppy drives would even light up at this point.

I don't have a logic probe at home, but I can get one tomorrow. I am not sure where the crystal is located and what it looks like - can you point it out in a picture?

patscc
January 16th, 2009, 01:01 PM
Does the -5 leg actually come from a switcher, or from a linear regulator in the power supply since it's such a low-current leg ? Common configuration in a lot of older power supplies, HOWEVER, I don't know about the Apple, so if someone could point me to a schematic...
Anyway, Vifa's got a multimeter, so that's what we'll use 'til we run out of ideas.

You can take the multimeter to the lowest address pin of a DRAM, or of the processor, and you should get a reasonably constant AC signal around 2.25 or so VAC (multimeter's averaging of the pulse train over time, try it on known good one)

patscc

channelmaniac
January 16th, 2009, 01:25 PM
It's all from the switcher on the Apple.

The crystal looks like this:

http://www.allproducts.com.tw/manufacture99/crystal/product1.jpg

Vifa
January 16th, 2009, 02:04 PM
I will continue my work tomorrow.


I count on getting a logic probe tomorrow, can you describe what I shall do when I get it?

I think I have found the crystal too, it is placed in the row on the motherboard which is called 'A', right? Anyway, how can I get to replace this? and with what?


Vifa

patscc
January 16th, 2009, 03:33 PM
I should think that before any crystals get checked or replaced, an effort should be made to see if the -5 V leg is indeed functioning correctly, since the crystal has no effect on the PSU, but the PSU has an effect on the rest of the computer.
Does anyone have a schematic or something with what the voltages should be, and what range the allowed fluctuation is ?

patscc

channelmaniac
January 16th, 2009, 05:53 PM
The 4116 DRAM spec sheet calls for -4.5 to -5.5 so the power supply will need some repair work, however, even with it out of spec there still should be some sign of life on the monitor.

Since there is absolutely nothing, I think the crystal or the chip driving it is bad.

patscc
January 16th, 2009, 06:10 PM
That's why I suggested the measurements with the AC range of the multimeter on an A0 line, if there's activity (i.e. if the clock's running) there's be an AC voltage around 2.25 or so.
patscc

Vifa
January 17th, 2009, 05:00 AM
That's why I suggested the measurements with the AC range of the multimeter on an A0 line, if there's activity (i.e. if the clock's running) there's be an AC voltage around 2.25 or so.
patscc

I will try to do this, but I am not sure how my multimeter shall be set. The picture shows how I assume my multimeter shall be.

Then, where shall I measure?

patscc
January 17th, 2009, 07:06 AM
You want it on 200 V~. You have it on DC.
Measure pin 1 & pin 5. Make shure you notice if the readings say mV or V

patscc

Vifa
January 17th, 2009, 07:47 AM
Right, I have attached a picture of how the multimeter now is set. With this setting I have done what you told me to, and if the black lead is on pin 1 and the red lead on pin 5 I read 11,5; if the red lead is on pin 1 and the black lead on pin 5 I read 4,2.

I am not sure whether not it says mV or V, nothing appears to indicate it.

patscc
January 17th, 2009, 08:26 AM
Your meter appears to have a rather coarse resolution on AC.
You want the black lead against ground, or pin 16 on the 4116, and then measure pins 1 & 5 with the red lead.

patscc

Vifa
January 17th, 2009, 08:46 AM
Of course, just did the testing again.

Pin 1: (-00.1) - (-0.00)
Pin 5: 00.0 - 00.1

patscc
January 17th, 2009, 09:35 AM
Well, pin 5 doesn't look good.
Can you check pin 4 & pin 15 ?
patscc

Vifa
January 17th, 2009, 10:16 AM
Pin 4: (-00.8 ) - (-01.0)
Pin 15: 08.7

The first row of DRAM chips (row C) are all reading 04.6 on pin 15. But it also looks like these are a bit different.

Vifa
January 20th, 2009, 08:26 AM
What can I do now?

channelmaniac
January 20th, 2009, 10:43 AM
Well, without having any real test gear or knowledge of troubleshooting... the only thing left to do is either send it to someone that can repair it or take a last ditch effort of replacing the crystal to see if you can end up getting something, anything showing on the screen.

Raymond

patscc
January 20th, 2009, 10:52 AM
Can you borrow an oscilloscope from someone for a while, or take the computer to somewhere that has an oscilloscope ?

patscc

scommstech
January 21st, 2009, 04:20 PM
What can I do now?

Hi
Been following your progress.
Really you should be checking your Power supply properly.
The Astec PSU is the same for Europe as the US. The
110V/220V link decides. Your PSU has been worked on previously as the
switch has failed and is bypassed.
Apple clones coppied the same or similar circuit and board layout.
Early 8088 clones even used this same circuit. It is a good one.
These PSU do not work "off" the machines. They need to
be loaded. The clicking you hear is the switch mode ramping up but
then dying as the transformer saturates killing the cycle.
It will continue trying. Open circuit (no load) or shorting will
cause this to happen. Some PSU have adjustable +5V o/p
If this is set too high the +12v crowbars and shuts the PSU down.
There is a 6 pin connector onto the motherboard.
The 2 black leads are Earth.
There is a +5v.+12v, -5v, and -12v.
These are the safest places to test for voltage. Use a paper clip
pushed down the hole with the wire. Use 2 clips. Use one clip in
one of the black earth wire holes and use the other paperclip wire to
check the other wires for voltage. The colours are usually
Orange = +5V, Yellow = +12V Green - 12V Blue -5V.
If it is the PSU you will have to desolder/resolder components.
The clicking and sparking shows that the main fuse is ok
and there is some PSU activity.
Be very careful testing the components on the motherboard
as if you slip you can short bits and damage them. The arcing you mention
is not good.
The PSU electrolytic capacitor mentioned as a prime suspect is
if I remember right C7 (220uf/10v)
Best of luck
Scommstech

patscc
January 21st, 2009, 04:42 PM
scommstech said...Really you should be checking your Power supply properly.

I think Vifa's already done that:

Vifa said...
Pin 1: (-5,7) - (-6,3)
Pin 8: 11,8 - 12
Pin 9: 4,95

There's some more measurements in the thread if you dig for them.

The clicking/shutdown of the switcher was due to Vifa attempting to measure the voltage legs with the multimeter set to current, thus causing a short.

patscc

scommstech
January 22nd, 2009, 05:53 AM
I was in a hurry, didn't read all the threads.
As the volts are reasonable it does look as though the prob
is with the motherboard.
I repaired about a dozen apples last year but I did need my scope.
Regards
Scommstech.

patscc
January 22nd, 2009, 05:59 AM
scommstech said...volts are reasonable
The only one that was a bit of a concern was the -5V leg. The max spec for the 4116's -5 leg is -5.5 V, and Vifa's power supply seems to be slightly out of that range.
It seems there's no clock, so I think right now we're all waiting to see if Vifa can get his hands on a scope or probe or logic analyzer, or take the Apple to a location that does.
patscc

scommstech
January 22nd, 2009, 09:20 AM
No mention of the +12v

However here are some scope results of the clock generation cctry
taken last year on a motherboard.

IC B2 pin8 (14M clock) 1.5v peak to peak sign wave 1:1 ratio .075us
B1 pin3 (col ref) 4V p-p square wave 1:1 .3us
B1 pin15 (7m) 5V p-p sign wave 1:1 .5us
B1 pin14 (-7m) 5V p-p sign wave 1:1 .15us
C2 pin15 -ras 5V p-p squarish 1-3 app .5us
C2 pin14 AX 5V p-p square 1:1 .5us
C2 pin13 -cas 5V p-p square 1:1 .5us
C2 pin12 03 5v P-P square 1:1 .5us
B1 pin7 00 5V P-P square 1:1 1us
B1 pin6 01 5V P-P square 1:1 1us
B12 pin8 LD194 5V P-P +ve pulse 1 or .1us
A2 pin3 -LDPS 5V P-P -ve pulse 1 or .1us
B6 pin15 -RAMSEL 5V P-P 1:1 1 or .1us

Last 3 measurements, I changed the decimal point on my notes and can't
remember which is right.
Hope these help
regards
scommstech

patscc
January 22nd, 2009, 10:21 AM
scommstech mentioned...No mention of the +12v
That's the voltage on pin 8 of the 4116's, it was measured (see a couple of posts back).
What I can't remember is if the -12V supply ever got mentioned.

Vifa, have you managed to gain access to an oscilloscope yet ?

patscc

scommstech
January 22nd, 2009, 11:54 AM
That's the voltage on pin 8 of the 4116's, it was measured (see a couple of posts back).
What I can't remember is if the -12V supply ever got mentioned.

Vifa, have you managed to gain access to an oscilloscope yet ?

patscc

I knew only 3 levels had been mentioned. miss read the - sign as -ve
The -12v winding sits on the -5v. If the -5v is out the -12 probably is also.
Don't think -12V is used on the motherboard though.
Scommstech

channelmaniac
January 22nd, 2009, 04:47 PM
Even if all the RAM is fried (which I doubt... the -5 was used for the charge pump), there still should be some glimmer of partial life showing up on the monitor.

I've replaced the 14.318MHz crystal in a number of II+/IIe machines back in the day. If you can't find the test gear, just try to replace that crystal and see what you get.

patscc
January 22nd, 2009, 06:09 PM
channelmaniac said...just try to replace that crystal and see what you get
The fact that when Vifa tried to get an AC signal off a couple of pins to see if there was activity and drew a blank makes channelmaniac's suggestion a real good one.
patscc

mwillegal
January 23rd, 2009, 03:39 AM
I wouldn't recommend swapping out soldered in parts without some more concrete reason. Something IS possibly wrong with timing here. However the crystal is the last place I would look. Could be system timing, which includes the following ICs: B1, C1, C2, B12, A2, D2, B2, B12 or sync counter D11-D14. Just as likely to be a bad connection to a socket as a bad component. There could also be something wrong with the last stage of the video generation, which involves a bunch more chips and some discrete components. You might try swapping some of the chips around, before getting out a soldering iron. If you don't have a donor board to use for a source of ICs. many of the chips in the Apple II are used in multiple places and like chips can be swapped with each other to see if the behavior of the board changes. Your best bet is to debug with a logic probe or oscilloscope, rather than by trial and error.

PS I believe that video generation should only require +5 and +12 volts. +12 volts is only needed for oscillator circuit. Note that the screen will not have a useful display of text or graphics without DRAM and associated voltages also working.

patscc
January 23rd, 2009, 06:23 AM
mwillegal said... logic probe or oscilloscope, rather than by trial and error
I think we all agree on that, unfortunatly it seems Vifa doesn't have one, and thus far hasn't managed to get his hands on one. I wouldn't suggest randomly swapping out IC's, but replacing a crystal is pretty straightforward, and following channelmaniac's suggestion certainly would be something quick to do, and wouldn't do any harm, or would it ?
patscc

channelmaniac
January 23rd, 2009, 07:24 AM
Ah, the joy of Monday Morning Quarterbacking. :rolleyes:

I've replaced way more than one bad crystal on Apple II series machines back in the day.

(Back in the day = working for Contemporary Concepts in Paris, TX back in the mid 90s. We did grey market component level Apple II and Mac repairs)

It's the quickest and easiest thing to replace when you have power but absolutely NO signs of life on the monitor. And, it's cheap. You can find those (14.318 ) crystals in many different pieces of gear if you want to rob one to try.

RJ

Vifa
January 26th, 2009, 06:34 AM
Vifa, have you managed to gain access to an oscilloscope yet ?



So far no luck.