I'm preparing my Apple II Rev 0 and related documentation for sale soon, as mentioned in a earlier post on this forum. I last powered it on successfully in September 2014 and let it run a few days, with on/off cycling every so often. Turned it off, backed away in its Apple carrying case and put it on a shelf.
I'm the original owner, purchased it in late 1978 all with its original PSU and chips, Serial # A2S1-3465, PSU A2M001-3615 (toggle switch), and used it extensively for generating vector graphics used in professional laser shows. I graduated to the IIe around 83-84. It has been periodically turned on (booted) to maintain its health. I have to admit that I was surprised that it powered up to all the correct Hex mode, FP Basic prompts in 2014 since it was last turned on and exercised about 8 years earlier.
I've been taking photographs of it and its accessories for the Ebay posting. When it came time to take current photos of operational display screens...it would not turn on. The PSU was totally dead. I had purchased a full set of replacement caps for this eventuality a few years ago and intend to repair the PSU. In the meantime I'll use a spare IIe PSU to make sure the MB is still viable. It has the FP ROM card all the original cassette tapes and more.
I've read many Rev 0 and PSU post on this forum over the years and noticed a lot a questions about whether there were Rev 0 PSU schematics and component lists, etc. The Rev 0 PSU schematic in the Apple II Reference Manual appears to represent one of the earliest incarnations and is not accurate for my Apple II and PSU serial numbers. The bottom of the PSU PCB is "PWR SUP 4A", so it is probably a 4th generation to what is shown in the Reference Manual.
So I've taken extensive photographs of my PSU and its components, including the single sided PCB and created a component list.
Thought I would first see whether there is any interest in my posting furthing information before continuing (because maybe its already on the forum and I've just not seen it).
Cheers,
Just post what you have. It doesn't matter if there is interest right now or not. Someone will definitely find your work useful in the next 100 years.
Thanks CVT, will do so.
I did just find this post from 2018 of an almost identical PSU, https://www.applefritter.com/content/help-recognizing-early-apple-ii-psu. The serial number is much later than mine but the PCB traces are identical. The key differences are shown by these photos, first mine, then his.
20230419_213303-cs.jpg
Drakepirate's PCB below:
IMG_9255ss.jpg
I just found the key difference between these two early PSU PCBs is found in the very lower right-hand corner of the boards. There is a trace widening and rerouting in the later version in the 10 o'clock position of the "ETCO" lable, AND the addition of what appears to be a 39 Ohm resistor just above the Apple logo on the later version. Its hard to know if this 39 Ohm resistor is original.
It is not "ETCO" but rather "FETCO", a circuit board assembly service company, and may still be in business https://www.fetco.com/parts-warranty, or maybe it is a derivative from the Apple era FETCO.
C1 and C2 measure good with the Altas ESR70
The 470uF/6.3v radial electolytic was questionable, noticable bugle on the "can's" top surface, Atlas ESR70 measured 1027uF ESR=1.22
Of the three axial 10uF/15v electrolytics, C6, C7 and C22, one was "open" the other two very high ESR > 40 but measured 10uF.
Progress being made on component list and their layout location on PCB that conforms to the labeled schematic found in the Apple II Reference Manual circa 1977/78 shown here:
A2PowerSupply-2-c.jpg
After electrolytic replacement
I tried reforming some of the caps but in the end decided it made more sense to put new , low ESR ones in. Most I got from Mouser.com, one came from Newark because Mouser only sold in quantities of 200 (but at 80 cents apiece).
Most all of my A2M001-3615 PSU schematic components, shown above, have their values or part numbers associated to their Identifiers (Rx, Cx, Qx, CRx, etc.) and I will post the list soon. There were only a few glass diodes I did not confirm but could be 1N914 or 1N4148 or similar. I chose not to desolder them in order to see if a part number was on their underside.
Original
A2M001-3615 Original
Does anyone have some idea how that SMPS works? Four transistors. It is the ultimate in simplicity!
It looks like Q3 is the main driver transistor switching the supply, and the top winding on the left side the main driver from the mains power.
R4/C4 looks like a snubber to protect Q3 from voltage spikes.
The base of Q3 is connected to the bottom coil on the left side, which looks like the primary feedback winding.
It looks like Q1 shuts off Q3 through Cr7. Q2 I think is there to limit the current through Q1, because as the current in Q1 increases, the voltage across R10 increases. As it approaches 0.7 V, Q2 turns on and raises the voltage of the emitter of Q1, lowering the emitter-base voltage of Q1 and thus limiting the current through it.
AR1 I think is to limit the time Q1 shuts off, because once AR1 is on, the capacitor C8 discharges and eventually the current through the base of AR1 drops to zero and AR1 shuts off. It could be that the AR1 transistor is there to ensure there Q1 does not prevent startup?
CR5 looks like its a zener diode that provides the regulation by conducting to the positive supply when the voltage reaches a certain level, turning on Q1 and shutting off Q3.
It looks like CR11 turns on when the voltage in the winding its connected to (third from the top on the left) is a high enough voltage so that CR11 starts to conduct and draws current away from the emitter or Q1 enabling it to turn on and shut off Q3, which is another feedback winding.
C6 looks like it there to provide the voltage to turn on AR1 and the diode CR9 prevent C6 from having a reverse voltage.
Anyways that must have been fun to design in the mid 1970s. There were probably some transistors burned creating that.
Yes, it is amazing someone figured out how to make the early SMPS work. There are actually 4 transistors, one optoisolator transistor and 1 SCR
I believe the AR1* transistor on the primary side (part of an opto-isolator) is turned on by the AR1* LED tied to R19 and Q4 on the 12v secondary side.
I put an EIDE hard drive as a load on the +5v and +12v supplies, turn on the PSU and the hard drive's 12v motor spun up, ran for a while and shut down. I metered the voltages on these two supplies and saw that both were slowly oscillating between about 1v and 2.25v (on the +5v) and between 1v and 3-4v on the +12v supply.
Apparently an EIDE drive doesn't put much of a load on the +5v rail because when testing the drive connected as load to bench power supplies the current draw on each power rail of the two bench power supplies, there was virtually no significant current draw off the +5v test supply and about 250mA on the +12v supply.
Without any load on these two SMPS's supply voltages, the re-cap'd PSU does the click-click-click thing, with load, no clicking but slow oscillating low voltage outputs. Its going to be a pain to troubleshoot. :(
After my last response, I re-read the attached PDF document I found on-line some months ago (after much searching) that documents many of the ASTEC SMPS'. The first 40 pages show schematics and part lists for ASTEC's various MSPS, including the +AA11040-B version used in Apple II's PSU 605-5703. The last 30 pages describe SMPS Basics. Pages 47 and 54 show block diagrams of the Buck Flyback Circuit. From this I found it easier to see how the early A2M001 in this thread correlates in similar operation.
Astec_Power_Supplies_Aug82.pdf
Woohoo! It is working after all.
While digging in my power resistor stash, looking for suitable load resistors with appropriate power ratings for my Apple II PSU's, I found that...a long time ago I made a +5v and +12v load resistor set providing a 4.6 ohm load for +5v and 34.2 ohm load for +12v.
With these heavier resistive loads the PSU works as designed! I'm running a load test on it now.
Before I tried this heavier load on this early A2 PSU, I dug out an Astec 605-5703, Serial # 986354 I had purchased on Ebay a few years back just to have a backup. It was advertised as "untested". Turned out it did not produce any output. This afternoon I opened it up and removed the circuit board. It appeared to have been re-capped with low ESR caps. All looked correct, except one cap was soldered in backwards (220uF/10v). I desoldered it and put in with the correct polarity. Now this PSU came back to life as well! I did a load test on it just now with my newly found Apple II PSU load resistor set and it passed.
I found a good circuit operational description in this PDF publication I had on file.
See pages 22 & 23 (in the manual), page 27 of the PDF file.
Apple-II-Circuit-Description-Processed-OPT.pdf
Hurray! What a relief! My Apple II Rev 0 is alive. Some keys are funky, they output several characters when pressed, probably need cleaning, so that's my next task.Ran memory test and need to reseat some memory chips, as I'm getting some errors from 07CF to 07F7 and at 0265.
But it is good to see it displaying on a screen again.
Here is my rough draft of the early Apple II PSU discussed in this thread. Let me know if anyone finds any errors. I'll add the text list of component numbers and component value fairly soon along with an update to this drawing.
CamBam+ A2M001-3615-PCB-2.jpg
I hope others find this helpful.
Circuit trace paths are based on the actual A2M001 PSU. I've represented them by thin polylines in this CAD drawing for simplicity since I posted photos of the actual traces earlier in this thread. This first drawing is primarily for component identification and location. I'll post a version that focuses on component drill holes, pads and traces tomorrow, hopefully.
Circuit trace lines were drawn as centerlines for all given circuit traces. The centerline circuit trace polylines were drawn by tracing over the photo circuit traces at full scale. The black dots prepresent drill hole centers, red circles are drill hole diameters (.050" dia. mostly, larger for screw mounts), black circles are pad sizes (.13", .16, .17, .18" dia., mounting holes .125" dia., transformer large pads .25" dia., smaller transformer pads .18" dia. (Pad dimensions were measured with a digital micrometer so it surprised me that there was so much subtle variation between .16, .17 and .18" diameter pads. Quite a few pads were trimmed to allow for pad clearance or for circuit trace path clearances. When pad circles overlap they are all connected. This information is for reference only.
CamBam+ A2M001-3615-PCB-SolderSide.jpg
A bit more to follow...
Is it me or the polariy of C10..C12 are inverted?
They are inverted - the plus should be on the side of the transformer. Good catch!
C10, c11 and c12 were all inverted. Here is the corrected version:
CamBam+ A2M001-3615-PCB-ComponentSide.jpg
Here is the Excel spreadsheet I created of the parts used in this A2M001-3614 PSU used in my Apple II Rev. 0 S/N 3465. Please let me know if any errors are found and corrections needed.
I created part labels based on type, e.g. R1 thru Rn, C1 thru Cn, etc. I noted when a part number was not present in the schematic or the PCB. As I believe I mentioned before, I was unable to identify the glass diodes that are similar in appearance to a 1N914 or 1N4148. There was no identifying marks on the SCR or the power diodes on the secondary side, CR12, CR13, CR14. If anyone knows what these are please let me know and I will update the spreadsheet.
I did see two or three pairs of pads and drill holes as if for components that were never populated on this version of the PSU PCB.
I believe most of the resistors are metal film although I'm not certain, but that's what they look like to me.
If there is any interest I will post the replacement part numbers for the low ESR Electrolytics I purchased from Mouser and Newark Electronics. Only one cap was purchased from Newark because Mouser had a 200 minimum order quatity.
Too bad that the transformers used in this supply are probably unobtanium these days because with this work it would probably be pretty simple for someone to create Gerbers to make PCBs to build reproductions of this.
How true. I started to do this in KiCAD and because of the unubtanium transformer decided it wasn't worth the effort.