CPU slot cards died or problem in the machine?

TCH · 18 December 2021, 13:19

I have two CPU slot turbocards (Harms Computertechnik Professional 030 Plus 500 and Daniel Instruments Stormbringer H530), both of them went down at the same time a few years ago. I have no idea what is the problem with them. The A500+ itself works and sidecar extensions (RocTec RocHard or ACA500 with Blizzard 1230) are also working.

How can i determine if the problem is in the machine or the cards themselves are died? The CPU-s are also confirmed to work, removing the ZIP RAMs did not helped. The machine itself was recapped a few years ago, when the cards stopped working and that did not help either. The Stormbringer had one 16V 220uF capacitor, that has been replaced, to no avail. Changing oscillators also did not help.

Unfortunately i have no third card or another A500 to test if the cards would work with a different machine or that the machine would work with a different card...

Jpor · 18 December 2021, 16:25

Quote:

Originally Posted by TCH

I have two CPU slot turbocards (Harms Computertechnik Professional 030 Plus 500 and Daniel Instruments Stormbringer H530), both of them went down at the same time a few years ago. I have no idea what is the problem with them. The A500+ itself works and sidecar extensions (RocTec RocHard or ACA500 with Blizzard 1230) are also working.

How can i determine if the problem is in the machine or the cards themselves are died? The CPU-s are also confirmed to work, removing the ZIP RAMs did not helped. The machine itself was recapped a few years ago, when the cards stopped working and that did not help either. The Stormbringer had one 16V 220uF capacitor, that has been replaced, to no avail. Changing oscillators also did not help.

Unfortunately i have no third card or another A500 to test if the cards would work with a different machine or that the machine would work with a different card...

Power supply good?

TCH · 18 December 2021, 18:28

I have two AT PC supplies, modified for Amigas and the A500+'s supply have an additional power cord for the RocTec RocHard. My dad made them in the nineties.
The A500's supply has been recapped, when the cards stopped working.

I've done a quick measure: the A1200's supply provides 4.70V and 11.70V, the A500+'s provides 4.68V and 12.28V. Both supply can drive the A500+, plus the ACA 500 and the attached Blizzard 1230 IV with 68882 and 64 MB RAM or the RocHard if it's the A500+'s supply. The cards did not work with either.

UberFreak · 18 December 2021, 19:07

4.7V on the 5V line is too low.

TCH · 18 December 2021, 21:07

Are you sure? The machine and the sidecar extensions are working with it. The A1200 with a Blizzard 1260 works with it.
If it is really low, then what can cause it? The A500+'s supply was recapped in 2015 december.

TCH · 18 December 2021, 21:49

Okay, i tried to measure it again with another supply. It turned out that the molex connectors of the supplies which are not connected, are providing 5V (+/- 0.02-0.03V). So, the supplies are actually give the 5V.
But if i measure the voltages on the connected Amiga connector (as in: i measure the voltages inside the machine), then i get this around 4.7V value. Does this mean something, or is this natural?

Shadowfire · 18 December 2021, 22:03

In ancient PCB's like in the Amiga, it is normal to have some drop as they don't use power planes for power distribution. The resistance of the copper traces produces a loss in voltage (you can verify this by using GND on the input power connector pin, and probing +5V at the connector, and +5V on the opposite side of the board while the machine is on.)

These drops were well-known by the hardware engineers of the time. It wasn't uncommon of them to turn up the power supply to 5.1-5.2V at the factory so that even at the lowest voltage levels of the board it didn't drop below 4.75V.

Is your multimeter calibrated? Because if it isn't, your measurements may not be as accurate as displayed, and I'd hesitate to turn up the power supply before confirming it with something that you know is pretty close to reality.

Generally speaking you want all of the IC's being between 4.75V and 5.25V. If its 4.7V at the connector, its certainly lower elsewhere on the board, and yes, it can certainly be a problem that "only" shows up when you have the accelerator on it.

Even the wire coming from the power supply has resistance; this is why you measure 4.7V with the machine connected, and 5.0V with it disconnected. The power supply is still regulating at 5.0V, but there is a drop in the PSU's wire (and the PSU connectors).

TCH · 18 December 2021, 22:28

And why only the CPU slot accelerators affected? The RocHard works and also the ACA 500 with the Blizzard 1230. Also the A1200 with the Blizzard 1260.
And why they worked before with the same supplies? I cannot say that the levels were the same before, but i doubt that all three supplies would have the same problem, especially since the A500+'s supply is recapped.

Or the problem what caused that amount of voltage drop occurred in the machine when the cards stopped? Perhaps the sidecar has less drop than the CPU slot?

I don't know for sure if my multimeter is calibrated, but i guess so.

TCH · 18 December 2021, 23:02

Okay, i've found my fourth supply. I measured the voltage without connecting it. It was above 5.2V, so i have one supply with that voltage level.
Then i plugged it in and measured on the molex. It was 5.0V. Then i measured inside the machine. It was 4.10V on the floppy connector and 3.97V on the back of the connector! And the machine along with the ACA 500 + Blizzard 1230 still worked, although the screen was noticeably darker than usually. The cards of course did not boot up.

TCH · 20 December 2021, 16:52

I've demoted my fourth supply (it was a modified PC supply anyway) and took the cable with Amiga connector from it and soldered a molex on to the other end of the cable, plus a pincher onto the shielding wire. I plugged this makeshift "adapter" to two other PC supply and with one of them (it was from a 286 PC) the floppy did some weird things. Sometimes it sounded like it tries to access a floppy (the keyboard LED became lit and stayed that way), but two times it begun to beep in a very high pitch.

Another idea: maybe the two batteries' leakage did some damage to the MB and CPU-slot cards cannot get enough power?
Or any other idea?

TCH · 20 December 2021, 19:50

My brother lent me an ATX power supply. 500W, i measured above 5.3V from directly the supply and measured 3.93V inside the machine. The cards are dead, but the machine + ACA 500 + Blizzard 1230 works without any problems.

Does anyone have any ideas?

18 December 2021, 13:19	#1
TCH Newbie Amiga programmer Join Date: Jun 2012 Location: Front of my A500+ Age: 38 Posts: 372	CPU slot cards died or problem in the machine? I have two CPU slot turbocards (Harms Computertechnik Professional 030 Plus 500 and Daniel Instruments Stormbringer H530), both of them went down at the same time a few years ago. I have no idea what is the problem with them. The A500+ itself works and sidecar extensions (RocTec RocHard or ACA500 with Blizzard 1230) are also working. How can i determine if the problem is in the machine or the cards themselves are died? The CPU-s are also confirmed to work, removing the ZIP RAMs did not helped. The machine itself was recapped a few years ago, when the cards stopped working and that did not help either. The Stormbringer had one 16V 220uF capacitor, that has been replaced, to no avail. Changing oscillators also did not help. Unfortunately i have no third card or another A500 to test if the cards would work with a different machine or that the machine would work with a different card... Last edited by TCH; 18 December 2021 at 13:23. Reason: cap, osc

18 December 2021, 22:03	#7
Shadowfire Registered User Join Date: Aug 2001 Location: Connecticut USA Posts: 617	In ancient PCB's like in the Amiga, it is normal to have some drop as they don't use power planes for power distribution. The resistance of the copper traces produces a loss in voltage (you can verify this by using GND on the input power connector pin, and probing +5V at the connector, and +5V on the opposite side of the board while the machine is on.) These drops were well-known by the hardware engineers of the time. It wasn't uncommon of them to turn up the power supply to 5.1-5.2V at the factory so that even at the lowest voltage levels of the board it didn't drop below 4.75V. Is your multimeter calibrated? Because if it isn't, your measurements may not be as accurate as displayed, and I'd hesitate to turn up the power supply before confirming it with something that you know is pretty close to reality. Generally speaking you want all of the IC's being between 4.75V and 5.25V. If its 4.7V at the connector, its certainly lower elsewhere on the board, and yes, it can certainly be a problem that "only" shows up when you have the accelerator on it. Even the wire coming from the power supply has resistance; this is why you measure 4.7V with the machine connected, and 5.0V with it disconnected. The power supply is still regulating at 5.0V, but there is a drop in the PSU's wire (and the PSU connectors). Last edited by Shadowfire; 18 December 2021 at 22:12.

18 December 2021, 22:28	#8
TCH Newbie Amiga programmer Join Date: Jun 2012 Location: Front of my A500+ Age: 38 Posts: 372	And why only the CPU slot accelerators affected? The RocHard works and also the ACA 500 with the Blizzard 1230. Also the A1200 with the Blizzard 1260. And why they worked before with the same supplies? I cannot say that the levels were the same before, but i doubt that all three supplies would have the same problem, especially since the A500+'s supply is recapped. Or the problem what caused that amount of voltage drop occurred in the machine when the cards stopped? Perhaps the sidecar has less drop than the CPU slot? I don't know for sure if my multimeter is calibrated, but i guess so. Last edited by TCH; 18 December 2021 at 22:43.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Is this Amiga 4000 CPU slot faulty?	reflex	support.Hardware	13	07 September 2022 05:20
Mediator (4 slot) graphics cards and 3D question	chocsplease	Amiga scene	17	02 November 2017 16:24
The CPU slot memory card for A2000A	TommyE	MarketPlace	8	30 January 2017 14:48
Would an A1200 cpu slot to A4000 cpu slot work ?	SimonV	Amiga scene	2	23 April 2014 08:53
Rare Collectable Amiga Slot Machine Boards	Twistin'Ghost	MarketPlace	5	02 August 2002 11:38

18 December 2021, 18:28	#3
TCH Newbie Amiga programmer Join Date: Jun 2012 Location: Front of my A500+ Age: 38 Posts: 372	I have two AT PC supplies, modified for Amigas and the A500+'s supply have an additional power cord for the RocTec RocHard. My dad made them in the nineties. The A500's supply has been recapped, when the cards stopped working. I've done a quick measure: the A1200's supply provides 4.70V and 11.70V, the A500+'s provides 4.68V and 12.28V. Both supply can drive the A500+, plus the ACA 500 and the attached Blizzard 1230 IV with 68882 and 64 MB RAM or the RocHard if it's the A500+'s supply. The cards did not work with either.

18 December 2021, 19:07	#4
UberFreak Registered User Join Date: Sep 2009 Location: the world Posts: 439	4.7V on the 5V line is too low.

18 December 2021, 21:07	#5
TCH Newbie Amiga programmer Join Date: Jun 2012 Location: Front of my A500+ Age: 38 Posts: 372	Are you sure? The machine and the sidecar extensions are working with it. The A1200 with a Blizzard 1260 works with it. If it is really low, then what can cause it? The A500+'s supply was recapped in 2015 december.

18 December 2021, 21:49	#6
TCH Newbie Amiga programmer Join Date: Jun 2012 Location: Front of my A500+ Age: 38 Posts: 372	Okay, i tried to measure it again with another supply. It turned out that the molex connectors of the supplies which are not connected, are providing 5V (+/- 0.02-0.03V). So, the supplies are actually give the 5V. But if i measure the voltages on the connected Amiga connector (as in: i measure the voltages inside the machine), then i get this around 4.7V value. Does this mean something, or is this natural?

18 December 2021, 23:02	#9
TCH Newbie Amiga programmer Join Date: Jun 2012 Location: Front of my A500+ Age: 38 Posts: 372	Okay, i've found my fourth supply. I measured the voltage without connecting it. It was above 5.2V, so i have one supply with that voltage level. Then i plugged it in and measured on the molex. It was 5.0V. Then i measured inside the machine. It was 4.10V on the floppy connector and 3.97V on the back of the connector! And the machine along with the ACA 500 + Blizzard 1230 still worked, although the screen was noticeably darker than usually. The cards of course did not boot up.

20 December 2021, 16:52	#10
TCH Newbie Amiga programmer Join Date: Jun 2012 Location: Front of my A500+ Age: 38 Posts: 372	I've demoted my fourth supply (it was a modified PC supply anyway) and took the cable with Amiga connector from it and soldered a molex on to the other end of the cable, plus a pincher onto the shielding wire. I plugged this makeshift "adapter" to two other PC supply and with one of them (it was from a 286 PC) the floppy did some weird things. Sometimes it sounded like it tries to access a floppy (the keyboard LED became lit and stayed that way), but two times it begun to beep in a very high pitch. Another idea: maybe the two batteries' leakage did some damage to the MB and CPU-slot cards cannot get enough power? Or any other idea?

20 December 2021, 19:50	#11
TCH Newbie Amiga programmer Join Date: Jun 2012 Location: Front of my A500+ Age: 38 Posts: 372	My brother lent me an ATX power supply. 500W, i measured above 5.3V from directly the supply and measured 3.93V inside the machine. The cards are dead, but the machine + ACA 500 + Blizzard 1230 works without any problems. Does anyone have any ideas?

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)