• Please review our updated Terms and Rules here

ASTEC AA11040C repair

I

iz8dwf

Veteran Member
Joined
Mar 20, 2017
Messages
812
Location
Italy
Hi All,
I'm in a apple II repair month :)
I have this Astec AA11040C supply that had the usual smoked rifa X2 capacitor. Other than that, it's outputting about 5.5V on the +5V line and 13V on the +12V line. However, with a 7 ohm load on +5V, the rail goes to 5.0V. I'll check the ESR of all electrolytics, but I was curious to know what's the minimum load (if any) to start regulating correctly the +5V rail?
Is there a schematic of this power supply floating around?
Thanks in advance.
Frank IZ8DWF
 
Al Kossow said:
http://bitsavers.org/pdf/apple/power_supply/Astec_Power_Supplies_Aug82.pdf
Click to expand...

Hi Al!
Of course I had already downloaded that, but it covers the AA11040 and AA11040-B only. I don't know yet if the -C that I'm repairing is too much different from -B (so far it looks almost identical), that's why I asked if this 1983 version is documented separately maybe.
Did anyone try that minimum regulation load? By a quick look at the schematic, it shouldn't have regulation problems with light loads, but I may be wrong on this.

Frank IZ8DWF
 
But you'd expect it to drop to the regulated level once you put some sort of load on it wouldn't you? So long as it doesn't suddenly start putting out 3.75 and 10 instead of 5 and 12 it should be pretty happy I would've thought.
 
Anthony Adverse said:
But you'd expect it to drop to the regulated level once you put some sort of load on it wouldn't you? So long as it doesn't suddenly start putting out 3.75 and 10 instead of 5 and 12 it should be pretty happy I would've thought.
Click to expand...

I don't expect it to regulate that "poorly". With about 10mA load, its output voltage is 5.5V, with 160mA load, it's 5.2V, with 730mA load, it goes at 5.0V.
My question since the beginning is: is this supply supposed to need any sort of minimum loading on the 5V rail, and if yes, what is that load, in everyone elses experience?
It is my first encounter with Apple II SMPS, so maybe it's just poor design from 1980's era, I just don't know.
Thanks

Frank IZ8DWF
 
I don't have a direct answer to the question at hand. In my experience with IBM "AT" power supplies, the +5VDC rail required a minimum of 2 to 4 amps. And I anticipate the same goes with the Apple II power supplies, since both it and the AT switching power supplies are around the same era (i.e. pre-ATX.)

You might be complicating things when the answer could be staring right at you. I'd simply connect the PS to the Apple II logic board, measure the current draw, and the result should be your answer. Run the test with no cards and keyboard. This may not be the exact calculated minimum you maybe shooting for. But, in the words of the great ozzie bloak Dave Jones, physical measurements are better than theoretical.
 
groink said:
I don't have a direct answer to the question at hand. In my experience with IBM "AT" power supplies, the +5VDC rail required a minimum of 2 to 4 amps. And I anticipate the same goes with the Apple II power supplies, since both it and the AT switching power supplies are around the same era (i.e. pre-ATX.)

You might be complicating things when the answer could be staring right at you. I'd simply connect the PS to the Apple II logic board, measure the current draw, and the result should be your answer. Run the test with no cards and keyboard. This may not be the exact calculated minimum you maybe shooting for. But, in the words of the great ozzie bloak Dave Jones, physical measurements are better than theoretical.
Click to expand...

Well, surely I'll try the "real" load, once I make some more tests.
However in my experience, "AT" type SMPS have a fairly good regulation circuit. I've never seen a 10% variation of the +5V rail even at zero load on all the ones I've repaired and used in many years.

Frank
 
I know that some claim the PSU will switch off without sufficient load, but mine runs fine no load to full load, but yes, the regulation is quite 'loose'. Regulation from 5.5 down to 5.0 seems fine to me. There are no IC comparitors in this PSU, its an analog switching circuit.
 
Gary C said:
I know that some claim the PSU will switch off without sufficient load, but mine runs fine no load to full load, but yes, the regulation is quite 'loose'. Regulation from 5.5 down to 5.0 seems fine to me. There are no IC comparitors in this PSU, its an analog switching circuit.
Click to expand...

Well, unfortunately 5.5V is a bit on the edge. This particular PSU was sent to me because it seems the only thing that could have damaged two "FloppyEMU" boards. Unfortunately I haven't yet found a schematic of this FloppyEMU, but it uses a Xilinx XC9572XL CPLD directly connected to 5V I/O. This is usually fine since this part is rated up to 5.5V :)
You see the problem here?
Both FloppyEMU boards have a shorted CPLD :(

I think I better tweak somehow the voltage regulation of this PSU if I can... I still can't believe they made a thing that regulate so poorly, even with all analog regulation, it should be better than 5% easily.
Frank
 
groink said:
Seeing you mentioned the Floppy Emu, did you run across this article: https://www.bigmessowires.com/2014/03/28/troubleshooting-damaged-chips/
Click to expand...

Yes I've read it and sincerely it doesn't add any clue at least to what I know.
My totally unproven theory is that those CPLDs have a bit of "datasheet confusion". I mean, if absolute maximum rating for inputs is 5.5V, how would it be safe to list the same 5.5V as the allowed range? If you check any good old TTL/CMOS datasheet, you'll notice something like absolute maximum rating of, say, 6V and allowed max input of, say, 5.5V. These differences exist to account for production tolerances etc.

Frank IZ8DWF
 
iz8dwf said:
Well, unfortunately 5.5V is a bit on the edge. This particular PSU was sent to me because it seems the only thing that could have damaged two "FloppyEMU" boards. Unfortunately I haven't yet found a schematic of this FloppyEMU, but it uses a Xilinx XC9572XL CPLD directly connected to 5V I/O. This is usually fine since this part is rated up to 5.5V :)
You see the problem here?
Both FloppyEMU boards have a shorted CPLD :(

I think I better tweak somehow the voltage regulation of this PSU if I can... I still can't believe they made a thing that regulate so poorly, even with all analog regulation, it should be better than 5% easily.


Frank
Click to expand...

Different times with different requirements.

I believe a C supply is basically a B, the +5V can be adjusted by altering R32 & R25 ratio.

Failing that, an old style PC PSU with the -5V can still be bought 'new' and runs an Apple II easily, so you could always try one of these (I have one on my spare motherboard for testing and it works well, with the added advantage of having a soft switch to turn the PSU on & off)
 
iz8dwf said:
Yes I've read it and sincerely it doesn't add any clue at least to what I know.
My totally unproven theory is that those CPLDs have a bit of "datasheet confusion". I mean, if absolute maximum rating for inputs is 5.5V, how would it be safe to list the same 5.5V as the allowed range? If you check any good old TTL/CMOS datasheet, you'll notice something like absolute maximum rating of, say, 6V and allowed max input of, say, 5.5V. These differences exist to account for production tolerances etc.

Frank IZ8DWF
Click to expand...

One more "hint" from the datasheet: The 3.3V VCCINT power supply must be at least 1.5V before 5V signals are applied to the I/Os.
Now, imagine a connected device to an Apple II where the 5V is raising, and the CPLD supply depends on the local 3V3 regulator capacitance decoupling (both input and output decoupling...). Do you believe the designer paid attention to this?
That's one of the reasons I never use "tolerant 5V" devices in my retro-designs. Real 5V parts are still around (mostly Atmel and Altera), so why take these risks?

Frank
 
You must log in or register to reply here.
Back
Top