Hi Jan, thanks for your nice repair videos :) I also have an early A500 with the serial number 00229 in my collection. This looks like yours (US chicken lips keyboard and non embossed Commodore logo in the case). The invoice and warranty card with matching serial number are also included. This one was sold on 29. April 1987 by COMPUSERV, (Commodore approved service center) in Belgium, Antwerpen. Greetings from Germany, Frank
Hi Jan. Less than 100 Ohm impedance to ground on the 5V rail is concerning. It may mean that some ICs are marginal, leaking current to ground. More modern power supplies may be able to provide the extra current, but the longer you use the Amiga this way, the higher the chance that the ICs in question will eventually short out. One way to check this is to remove the socketed ICs one at a time, while measuring the impedance from 5V to ground on the multimeter. At some point you will pull out an IC that will lead to a jump in the impedance, beyond 100 Ohm. That is then the guilty party. Another test is to hook up the oscilloscope, with AC coupling, to the 5V rail of a modern Amiga with the computer on. Check the ripple with the old and newer power supplies. If the old one has higher ripple, you will know it's less powerful than the new ones, which may explain why it fails to deal with the 70 Ohm impedance from 5V to ground in the old Amiga. Either way, I doubt 70 Ohm is normal.
It's also entirely possible there's a capacitor about to let go, I've had a few of the the little yellow bead decoupling capacitors fail and go leaky /short on A500s
70 ohms is only 75mA at 5V. Some of the resistor packs connect between 5V and ground as well (I have no idea why they did that). What reading you get on your multimeter generally depends on the test voltage it applies.. it can be enough to turn some gates on at least partially
@@jaycee1980 That's not how you calculate current consumption in active circuits (i.e. circuits containing more than just passive elements). When the computer is actually on and the gates/flip-flops are switching, there will be a lot more current consumption than your static calculation. Every time a flip-flop or gate switches from one state to another, there is current inrush beyond your simple I = V/R. If only everything were as simple as Ohm's law... Unfortunately, computers are not resistors.
@@thebyteattic Yes, I know that... but when you are just measuring across the rails with a multimeter, you are not probing an active circuit as the meter is not supplying enough voltage to make the circuit active. That wasn't the point I was trying to make. What I was saying is that 70 ohms across the 5V supply rails of the board doesnt indicate any sort of short - in pure resistive terms its only a low current leakage.
@@jaycee1980That low current leak is indicative of a marginal semiconductor, which is the entire point I originally made. The leak may be considerably higher when the device is actually subjected to Vcc and switching. Semiconductors should not have low impedance from Vcc to ground.
Thanks Jan! Great video. Really appreciate all the work you put into this Amiga. Is a part of our history and past as Amiga users. Always a joy to see these machines still surviving after these many years. I personally have an A2500, A3000 and a CDTV that need TLC. None of them are working at this point and hoping I can bring them back to life.
Having been there on many occasions I can agree this is an excellent choice. Once in the collection, anyone can 'adopt' this machine so there is money available specifically for (extra) caring for this piece of history.
1. Plug the power into a known working board and Che k voltages. Not sure but maybe more modern systems don't use the 5v, but use a voltage regards off the 12v. Not sure, don't work on amiga 2. Cross check the power outputs with a working power supply C Keep up the great work Cheers. Keith
You really are guesing in the dark with that setup. Having a decent modern PSU with stable voltages and measurements to show you current draw on each voltage rail would be very helpful to debug systems which draw incorrect amounts of current etc. Also don't be shy to use diode mode when testing continuity etc. It's often a more telling test than just ohms.
Early A500 were power hungry and had a beefy PSU, highly sought after by the early 90s because the accelerators needed those extra amps. It might be that the PSU is just too weak to power up the old A500. Later revisions cut power consumption considerably, from 35W PSUs to 24W at the end. Should be easy to check as the values are embossed on the PSU case.
There will be a voltage drop over the connector when under load. It will be that the connector has been used with the thicker pin connectors. It will measure fine without a load a quick fix is just to swap on a different cable from one of the other power supplies
On the PSU: I would measure the voltage at the connector on the power supply side, if that is at normal voltage when it shows 0.2v on the other side you will know the issue is with the cable or connectors. P.S: Many multimeters continuity tests "beep" with a very high ohm value like 50ohms which would obviously not be ok for a power line. But also realize that tests is done with almost no current passing, so a voltage test as mentioned first would give more useful info.
I was thinking the same, not trusting continuity test on a wire that could pull half an amp, I found it interesting that -12v read high and +12v read low, as if the ground wire was raising a volt or two. I would measure the volt drop from the ground pin at the power supply compared to ground on the computer, and same with the 5v pins at the psu versus at the computer. But I fully understand not wanting to spend more time on this.
Interesting all those minor changes in the board. I have a rev 3 chicken lips A500, the board is labeled sn038042, week 20 of 87. Y1 is populates, X1 isn't. It's at least a rather early version (being rev3 and hi-tek keyboard). It has a Chinon drive. But, my case has an embossed logo. And from the Open Wide post on amigalove I actually read that embossed logo was key to find an old Amiga. For me the chickenlips keys gave it away and made me buy it. I'm 100% sure the rev5 A500 I had, from 91 or so, had a commodore sticker on the case rather than an embossed logo. All my friends had that too, except for one who got it on launch. And that one had an embossed logo. So I'm not sure what the type of Commodore logo actually gives away. Great series, I really enjoyed it and learned a lot again.
i would send the machine to the "RMC Cave" over in the UK. his channel is amazing. he has lots of old machines which he takes care of and show cases them for people to look at and play with. would definitely give that one a thought!
Not sure about this PSU, but I think some of the older Amiga PSU come with a sense line for the +5V supply, which is used to keep the 5V voltage stable, regardless of the current being drawn by the computer, compensating cable losses (i.e. voltage drop by the cable's resistance, according to Ohm's law). If it has a sense line, could something be wrong with it? 🤔
Hi Jan. Love your channel! Maybe you could try get a thermal image camera on the board while its powered up with another power supply and see if anything stands out as getting warmer than it should.
Just try to measure voltage inside power supply during powering this old amiga rev 3. For make shure that this is not a cable or socket issue under load.
When you use that anti static wristband you are grounded, so beware of that when you are working with high voltage. I don't know if you have a GFCI or not, but if you should touch the live connection with your right hand the current could go through your heart because your left hand is grounded.
Don`t they have a small resistor in the wristband circuit? I seem to remember this from the 1980`s when I worked in a microchip factory. I must of been bored and measured it.
since all three rails were low, trace it back to a common component. That'll be your weak link. Checking the voltages with an o-scope may also show waveforms on the DC lines that shouldn't be there. Good luck!
Okay.. so on the jack side we have 2 . something volt.... and on the plug side at that same time??..... if the power supply is holding up the 5v.. then the voltage is being lost in the cable or the jack/plug connector.. 🤷♀ ....... if the power supply is lowering the voltage... Well down here somebody said that is a feedback (power-good) on one of the cables..🤔
Hi Jan. Could the power supply issue be the pin size. A circuit can behave different when loaded. It can show continuity without load and break down under load. As a sanity check, can you solder another lead with the bigger pins to the underside of the power supply board to test ?
70 ohm is not that bad, it is only around 70mA with 5V (if my math is right). Perhaps an isolation issue in the Amiga, I mean between 5V and the other rails?
I had the same Problem with my Rev3 Board. And the Problem was the 1 Ohm Resistor EMI 406 on the A500 Rev3 Board. Maybe it's the same Problem? Check the Resistor😉
I dig didn't know about the email thing, but a few years ago, I bought an A500 with an American layout chicken lips keyboard and the Commodore logo badge on the top and a Rev 5 mainboard from a man in Italy named Federico. He had a different last name, though. It has a very unusual label on the bottom for some reason. It currently doesn't work, though. Coincidence or could I have gotten the first Federico's A500?
@@JanBeta .. Okay.. so on the jack side we have 2 . something volt.... and on the plug side at that same time??..... if the power supply is holding up the 5v.. then the voltage is being lost in the cable or the jack/plug connector.. 🤷♀ ....... if the power supply is lowering the voltage... Well down here somebody said that is a feedback (power-good) on one of the cables..🤔
I'd like to think I run a computer museum in my home, so I'd like to put my hat in the ring to get this Amiga. Just kidding. Though I do have a pretty decent collection, so ...
This particular ISMET supply often suffers from fractured solder joints around that main transformer, but I dont think that's the problem here. From memory, an SG3525 is driving a transistor and that produces the regulated +5V. The +12V is produced by a 7812, and the -12 produced by a 7912. This is one of the supplies where failure in the 5V circuit (specifically the pass transistor failing) can cause unregulated voltage output, killing the Amiga. The SG3525 is a bit of an odd choice here, as it's usually used in "totem pole" arrangement for driving a transformer.
i believe that the "broken psu" is the problem. check the wattage that each power supply can handle on 12v and 5v lines. the psu seems quite simple. u have the rectifier diode and the transformer coil. compare the two psu internals to see what is different like the resistance on the transformer etc. good luck 🙂
It doesn't have the original top case, it has the top case from the later models before the lasrt model that had the A500+ plus style badge on the case
because of 4 digit. many people trying to avoid numbers like 4, 13 etc. some of them likes 7,777, 5, 60, 69. btw, i'm working on the zx spectrum emulation on rp2040 board. it is almost done
Very strange that power supply, definitely sounds like there's more to it that's gone wrong than is obvious, but it's definitely one of those things where it's easier to cut your losses and replace it, cos it could just drive you nuts chasing a fault that is just not willing to be cleared... :\
I have even heard that they did different serial numbers for different countrys. So 2000 can be nr 1 in usa and 4000 in UK so number 1 can be what ever :) what i heard. Follow the chip date
Hi Jan, thanks for your nice repair videos :)
I also have an early A500 with the serial number 00229 in my collection. This looks like yours (US chicken lips keyboard and non embossed Commodore logo in the case). The invoice and warranty card with matching serial number are also included. This one was sold on 29. April 1987 by COMPUSERV, (Commodore approved service center) in Belgium, Antwerpen.
Greetings from Germany,
Frank
Hi Jan. Less than 100 Ohm impedance to ground on the 5V rail is concerning. It may mean that some ICs are marginal, leaking current to ground. More modern power supplies may be able to provide the extra current, but the longer you use the Amiga this way, the higher the chance that the ICs in question will eventually short out. One way to check this is to remove the socketed ICs one at a time, while measuring the impedance from 5V to ground on the multimeter. At some point you will pull out an IC that will lead to a jump in the impedance, beyond 100 Ohm. That is then the guilty party. Another test is to hook up the oscilloscope, with AC coupling, to the 5V rail of a modern Amiga with the computer on. Check the ripple with the old and newer power supplies. If the old one has higher ripple, you will know it's less powerful than the new ones, which may explain why it fails to deal with the 70 Ohm impedance from 5V to ground in the old Amiga. Either way, I doubt 70 Ohm is normal.
It's also entirely possible there's a capacitor about to let go, I've had a few of the the little yellow bead decoupling capacitors fail and go leaky /short on A500s
70 ohms is only 75mA at 5V. Some of the resistor packs connect between 5V and ground as well (I have no idea why they did that). What reading you get on your multimeter generally depends on the test voltage it applies.. it can be enough to turn some gates on at least partially
@@jaycee1980 That's not how you calculate current consumption in active circuits (i.e. circuits containing more than just passive elements). When the computer is actually on and the gates/flip-flops are switching, there will be a lot more current consumption than your static calculation. Every time a flip-flop or gate switches from one state to another, there is current inrush beyond your simple I = V/R. If only everything were as simple as Ohm's law... Unfortunately, computers are not resistors.
@@thebyteattic Yes, I know that... but when you are just measuring across the rails with a multimeter, you are not probing an active circuit as the meter is not supplying enough voltage to make the circuit active.
That wasn't the point I was trying to make. What I was saying is that 70 ohms across the 5V supply rails of the board doesnt indicate any sort of short - in pure resistive terms its only a low current leakage.
@@jaycee1980That low current leak is indicative of a marginal semiconductor, which is the entire point I originally made. The leak may be considerably higher when the device is actually subjected to Vcc and switching. Semiconductors should not have low impedance from Vcc to ground.
Power supply: The multi meter you are using can measure current, so you can use that to see how much current the Amiga draws.
"That belongs in a museum" - Indiana Jones
Thanks Jan! Great video. Really appreciate all the work you put into this Amiga. Is a part of our history and past as Amiga users. Always a joy to see these machines still surviving after these many years. I personally have an A2500, A3000 and a CDTV that need TLC. None of them are working at this point and hoping I can bring them back to life.
Great work Jan.. Amazing to keep this in a the Retro Community.. RMC May be all over this one :)
Neil from RMC - The Cave would be a good home for this!
Well done Jan, top work and really interesting to go through the history of this machine.
deine videos sind einfach klasse. danke ❤
RMC Cave is probably one very good candidate for putting this Amiga in a proper museum where it can be appreciated and used.
As for your question at the end, think of the Dutch Home Computer Museum!
Tbh, that's a great idea. Would be great to visit and see an orginal Jan Beta repaired Amiga.
It would also be a great roadtrip for Jan! btw the museum is located in the city of Helmond in the Netherlands.
Having been there on many occasions I can agree this is an excellent choice. Once in the collection, anyone can 'adopt' this machine so there is money available specifically for (extra) caring for this piece of history.
1. Plug the power into a known working board and Che k voltages.
Not sure but maybe more modern systems don't use the 5v, but use a voltage regards off the 12v. Not sure, don't work on amiga
2. Cross check the power outputs with a working power supply
C
Keep up the great work
Cheers. Keith
You really are guesing in the dark with that setup. Having a decent modern PSU with stable voltages and measurements to show you current draw on each voltage rail would be very helpful to debug systems which draw incorrect amounts of current etc.
Also don't be shy to use diode mode when testing continuity etc. It's often a more telling test than just ohms.
Early A500 were power hungry and had a beefy PSU, highly sought after by the early 90s because the accelerators needed those extra amps. It might be that the PSU is just too weak to power up the old A500. Later revisions cut power consumption considerably, from 35W PSUs to 24W at the end. Should be easy to check as the values are embossed on the PSU case.
There will be a voltage drop over the connector when under load. It will be that the connector has been used with the thicker pin connectors. It will measure fine without a load a quick fix is just to swap on a different cable from one of the other power supplies
On the PSU: I would measure the voltage at the connector on the power supply side, if that is at normal voltage when it shows 0.2v on the other side you will know the issue is with the cable or connectors. P.S: Many multimeters continuity tests "beep" with a very high ohm value like 50ohms which would obviously not be ok for a power line. But also realize that tests is done with almost no current passing, so a voltage test as mentioned first would give more useful info.
I was thinking the same, not trusting continuity test on a wire that could pull half an amp, I found it interesting that -12v read high and +12v read low, as if the ground wire was raising a volt or two. I would measure the volt drop from the ground pin at the power supply compared to ground on the computer, and same with the 5v pins at the psu versus at the computer. But I fully understand not wanting to spend more time on this.
Great to see you back at it!
LOVE the Transistor Transistor shirt!
Interesting all those minor changes in the board. I have a rev 3 chicken lips A500, the board is labeled sn038042, week 20 of 87. Y1 is populates, X1 isn't. It's at least a rather early version (being rev3 and hi-tek keyboard). It has a Chinon drive. But, my case has an embossed logo. And from the Open Wide post on amigalove I actually read that embossed logo was key to find an old Amiga. For me the chickenlips keys gave it away and made me buy it. I'm 100% sure the rev5 A500 I had, from 91 or so, had a commodore sticker on the case rather than an embossed logo. All my friends had that too, except for one who got it on launch. And that one had an embossed logo. So I'm not sure what the type of Commodore logo actually gives away.
Great series, I really enjoyed it and learned a lot again.
i would send the machine to the "RMC Cave" over in the UK. his channel is amazing. he has lots of old machines which he takes care of and show cases them for people to look at and play with. would definitely give that one a thought!
Not sure about this PSU, but I think some of the older Amiga PSU come with a sense line for the +5V supply, which is used to keep the 5V voltage stable, regardless of the current being drawn by the computer, compensating cable losses (i.e. voltage drop by the cable's resistance, according to Ohm's law). If it has a sense line, could something be wrong with it? 🤔
None of the brick PSU's had remote sense
Hi Jan. Love your channel! Maybe you could try get a thermal image camera on the board while its powered up with another power supply and see if anything stands out as getting warmer than it should.
Just try to measure voltage inside power supply during powering this old amiga rev 3. For make shure that this is not a cable or socket issue under load.
Maybe check out the Home Computer Museum in Helmond, Netherlands.
When you use that anti static wristband you are grounded, so beware of that when you are working with high voltage. I don't know if you have a GFCI or not, but if you should touch the live connection with your right hand the current could go through your heart because your left hand is grounded.
Don`t they have a small resistor in the wristband circuit? I seem to remember this from the 1980`s when I worked in a microchip factory. I must of been bored and measured it.
A 1 MegaOhm resistor. Pretty safe.
since all three rails were low, trace it back to a common component. That'll be your weak link. Checking the voltages with an o-scope may also show waveforms on the DC lines that shouldn't be there. Good luck!
Yes !!! Jan beta, amiga, restoration ! Too bad it's Sunday night 😢
Okay.. so on the jack side we have 2 . something volt.... and on the plug side at that same time??..... if the power supply is holding up the 5v.. then the voltage is being lost in the cable or the jack/plug connector.. 🤷♀ ....... if the power supply is lowering the voltage... Well down here somebody said that is a feedback (power-good) on one of the cables..🤔
Hi Jan. Could the power supply issue be the pin size. A circuit can behave different when loaded. It can show continuity without load and break down under load. As a sanity check, can you solder another lead with the bigger pins to the underside of the power supply board to test ?
70 ohm is not that bad, it is only around 70mA with 5V (if my math is right). Perhaps an isolation issue in the Amiga, I mean between 5V and the other rails?
Have you tried to change both PSU cable and male and female conectors?
I had the same Problem with my Rev3 Board. And the Problem was the 1 Ohm Resistor EMI 406 on the A500 Rev3 Board. Maybe it's the same Problem? Check the Resistor😉
Ottimo video
Maybe the Homecomputer Museum in Helmond, the Netherlands is a good place for this machine.
GOOD OLD CHECKMATE KIT
Yes sometimes you must let go of stuff that dosent want to be fixed 😅 good work great channel
I dig didn't know about the email thing, but a few years ago, I bought an A500 with an American layout chicken lips keyboard and the Commodore logo badge on the top and a Rev 5 mainboard from a man in Italy named Federico. He had a different last name, though. It has a very unusual label on the bottom for some reason. It currently doesn't work, though. Coincidence or could I have gotten the first Federico's A500?
Unconnect RAM expansion and check everything again. :)
Is it the same with the fdd power unplugged?
Yes, pretty much.
@@JanBeta .. Okay.. so on the jack side we have 2 . something volt.... and on the plug side at that same time??..... if the power supply is holding up the 5v.. then the voltage is being lost in the cable or the jack/plug connector.. 🤷♀ ....... if the power supply is lowering the voltage... Well down here somebody said that is a feedback (power-good) on one of the cables..🤔
I'd like to think I run a computer museum in my home, so I'd like to put my hat in the ring to get this Amiga. Just kidding. Though I do have a pretty decent collection, so ...
❤
did you sign that one?
Here in Sweden it's common with A500 with "serial nr 1". They have probably been fixed in a service place.
This particular ISMET supply often suffers from fractured solder joints around that main transformer, but I dont think that's the problem here.
From memory, an SG3525 is driving a transistor and that produces the regulated +5V. The +12V is produced by a 7812, and the -12 produced by a 7912. This is one of the supplies where failure in the 5V circuit (specifically the pass transistor failing) can cause unregulated voltage output, killing the Amiga.
The SG3525 is a bit of an odd choice here, as it's usually used in "totem pole" arrangement for driving a transformer.
The first Amiga 500 to come out of the factory. I'm guessing came out in the boot of Jef Porters car.
Is that a 1.2 Kick Start on this Amiga? I bought my 500 just as 1.3 was released in the UK. I don't remember the date.
Answered my own question. Just saw a screen shot from Jan's earlier video.
Have you tried it without the memory expansion?
i believe that the "broken psu" is the problem. check the wattage that each power supply can handle on 12v and 5v lines. the psu seems quite simple. u have the rectifier diode and the transformer coil. compare the two psu internals to see what is different like the resistance on the transformer etc. good luck 🙂
Maybe the RMC Museum in Gloucestershire UK is a candidate.
It doesn't have the original top case, it has the top case from the later models before the lasrt model that had the A500+ plus style badge on the case
Guess I was wrong! Both my A500's with chickenlips keyboard has the case with just the molded logo. One of them has the rev 5 motherboard
take the board out with not aditional components as it may be leaking to the ground shield
THE SHIELD MAKES THE BOARD HEAT UP STRESSES THE CHIPS
HIGH RESISTANCE ON CONNECTOR CLEAN OR REPLACE 😊😊😊
🍪
Sounds like the ic is bad try replacing the SG3524 with a SG3524N.
Why did they skip rev. 4
because of 4 digit. many people trying to avoid numbers like 4, 13 etc. some of them likes 7,777, 5, 60, 69. btw, i'm working on the zx spectrum emulation on rp2040 board. it is almost done
You should show it at a local retro computer or expo, with all the documentation :)
Very strange that power supply, definitely sounds like there's more to it that's gone wrong than is obvious, but it's definitely one of those things where it's easier to cut your losses and replace it, cos it could just drive you nuts chasing a fault that is just not willing to be cleared... :\
I think the butler did it.
Ich bin mir sicher, das die Keramik-Chips nur im Amiga 1000 verbaut waren. Sie waren zu teuer für die "billigen" Amigas.
I have even heard that they did different serial numbers for different countrys. So 2000 can be nr 1 in usa and 4000 in UK so number 1 can be what ever :) what i heard. Follow the chip date
hey hi, Jan, maybe You can help me explain the 500 I have , and show in the video I put on TH-cam on my channel, Then I got the first also. LOL