I'm having trouble sending audio files to the cassette interface in a IIe PAL version. There are clearly some ciruit changes e.g. no 741, but half an MC1458 opamp seems to be the zero crossing detector. Is the basic format (frequencies 1kHz/2kHz/ 770Hz) the same, or were there slight differences due to the different clock crystals in these machines? I guess I could adjust the speed of the recording wav file slightly.
Anonymous
User login
Please support the defense of Ukraine.
Direct or via Unclutter App
Active forum topics
Recent content
Navigation
No Ads.
No Trackers.
No Social Media.
All Content Locally Hosted.
Built on Free Software.
We have complied with zero government requests for information.
If you use https://asciiexpress.net/gameserver/, are both HiFi and LoFi failing, or just HiFi?
There is no difference in the format between Apple II+ and Apple IIe or between PAL and NTSC versions.
Use maximum volume on your smartphone. Some smartphones even at maximum volume level have insufficient signal amplitude. I am using passive transformer amplifier in this case with step-up transformer as a single component taken from old German wire/line phones.
I hadn't seen this server (was looking at the one with ProDOS), but I tried one randomly from here and both HiFi and LoFi fail.
Alan
Check the -5V rail. The RC741 OpAmp (or the MC1458 in the PAL version) for the AudioIn is the only chip that needs -5V, so if it was missing all along, you would not have noticed before:
AudioIn.png
If the -5V is there, I suggest replacing the OpAmp itself. You can also check C11 and the resistors, but they are unlikely to fail.
Always use a short stereo AUX cable with the playback device on maximum volume and make sure it makes good contact with the AudioIn jack.
Thanks for that schematic! I hadn't been able to find one of the PAL IIe (except the actual video portions shown in the Understanding the Apple IIe book). I've probably got TL072 in stock somewhere and I think they are similar, will check the -5V rail with the scope.
Actually that schematic doesn't match up with the board, no 741 in sight. I found a scope and put it on the op-amp output pin 1. It's not a very good op-amp is it looking at the slew rates.... I don't remember the MC1458 being great though.
widths (Medium).jpg
(This is not meant to be a comedy oscilloscope shot, as it happens it is a Tek 222 battery crt scope, but the other option is a Tek 7904 which wouldn't fit on the bench where the Apple is). The lead in tone counts as fairly stable 770.0Hz when played on VLC.
Clearly this is not a clean eye pattern (it's from the portion just after the 770Hz tone so it should be standard apple 1kHz/2kHz frequencies, not the higher ones used later). I do have an NE5532 low noise dual op-amp which has the same pinout but it may be "too much of an improvement" and lose the ability for volume adjustments to compensate for small frequency adjustments.
The CPU clock measures 1.0153MHz (on a good 12 digit counter using a 100ms gating time) so about 0.995 of the frequency of the NTSC version.
In post #7, rfrb2 wrote:
" Clearly this is not a clean eye pattern (it's from the portion just after the 770Hz tone so it should be standard apple 1kHz/2kHz frequencies, not the higher ones used later). I do have an NE5532 low noise dual op-amp which has the same pinout but it may be "too much of an improvement" and lose the ability for volume adjustments to compensate for small frequency adjustments.
. . .
The CPU clock measures 1.0153MHz (on a good 12 digit counter using a 100ms gating time) so about 0.995 of the frequency of the NTSC version."
Uncle Bernie comments:
I'm quite fond of Apple-1 and Apple II cassette interfaces (took me some months of my life to bring the former up to snuff) and so please allow me the comment that you seem to be on a wild goose chase here.
The Apple II cassette interface is very robust and should work fine with almost any signal source, and it should never fail / give load errors with "perfect" modern media players.
All the talks about "low noise" opamps is nonsense, you can use a n y opamp known to man in the Apple II cassette interface and it would be fast enough and low noise enough.
So something else is amiss in your case. I don't have supernatural powers to diagnose your cassette interface by "remote viewing" - I have only about 15 feet of books on occult "sciences" compared to 300 feet of real science - but all what I want to tell you is to avoid wild goose chases.
Just make sure you have enough signal on the TAPE IN to be enough above the hysteresis. If you want to put "science" in it, use a sinus signal of 2 kHz on TAPE IN and observe the output of the comparator / opamp. This requires NO programming. You should get a clean rectangular signal out (use CH2 of scope) and then you can see if there is any hysteresis present. If this looks OK then your problem is elsewhere, and not on the cassette input circuit.
CPU clock freqency is irrelevant as long as it comes from a crystal oscillator and as long as the Apple II itself works. This is because the difference between the 1 kHz and 2 kHz "bit cells" is certain to dominate regardless of the small deviations to be expected from a functional crystal oscillator. Blaming the crystal is chasing "ghosts" . . . unless it is dead, and then the Apple II would have no clock at all.
- Uncle Bernie
@afrb2 you never described exactly how your tape input fails. Do you hear an apple2 speaker beep after several seconds after LOAD command and starting the playing?
Ah. I realise I wasn't clear. I was trying disk images from the apple II disk server https://asciiexpress.net/diskserver and these are not encoded using the standard tones but have a bootstrap with standard tones and then use an 8000 baud pair 12kHz / 6kHz and I found comment that the decoding is very clock rate sensitive since it seems to rely on instruction timing. I clearly need to find a WAV file of a standard cassette image encoded using the 1kHz/2kHz tones since this is what Apple intended to work.
If I'm very lucky trying to load DOS 3.3 I can get a beep and actually get the first INSTA-DISK message but then I get a checksum failure. Most of the time I get ERR very shortly after the end of the 770Hz lead in tone and another ERR shortly thereafter.
There's definitely no need for a low noise op-amp; but I work more with RF instrumentation than old computers so that's what I have in the spares bin. The eye pattern is my usual starting point for problems with a digital encoding.
There is definitely enough input that the zero crossings are being detected and the hysteresis works (and not so much that distortion impacts).
I looked at the eye diagram of the input audio and it was all over the place. Source was either a very old iPhone or a Windows PC. Went and got a studio audio interface and used that instead as the source. Dead stable eye diagram and loaded up INSTA-DISK.
Thanks to @Uncle_Bernie for the reminder that the Apple IIe casette interface is reliable. Windows device drivers on the other hand....
Alan
Yep, old iPhones are tricky! I got an old iPhone 6 Plus and an even older iPhone 4 (the first one with a retina display). While the iPhone 4 can load games from https://asciiexpress.net/gameserver/ on both LoFi and HiFi, the iPhone 6 Plus produces an error on HiFi and can only load successfully on LoFi.
Since you got a scope, this is what a "good" cassette input signal looks like or at least one that works correctly.
This capture shows an ascii express transfer (on NTCS iie but 50/60Hz shouldn't matter). Here there's first load sync, the disk writer application transfer, then a delay and sync, finally the sector data to write starts on the far right. I'm sure Uncle Bernie may have a reason for that first small dip is from... I didn't look into it. I think the scope was 1V/division, but that would mean the sync tones are above normal "line level" which shouldn't exceed 2V, and then the data doesn't look like it got above 1.2V or so. Most importatnt would be data cannot be saturated. So even tho this is a pretty picture, it may be worthless. I'll try to get a better example.
Also no matter what source, almost all modern audio devices have audio ehnancements which need to be disabled. This includes equalizations, bassbost, dolby headphones, etc. Everything off. If you got a line level output (older PCs with sound cards had this) that would be best. Otherwise make sure signal voltages within the +/-2V line level range.
ascii-express audio.jpg
I've actually found somthing more confusing about the two versions. I discovered that on some systems the Lo-fi version works, but the hi-fi doesn't and then on other systems it's the exact opposite hi-fi works but lo-fi doesn't!
Everything else in the path the same, only thing different is the II being used.
I may be hung up on the word "chip" but did PAL video otuput use -5V too?
Where can I find the PAL schematic?
In post #15, jeff d wrote:
I may be hung up on the word "chip" but did PAL video output use -5V too?
Where can I find the PAL schematic ?
Uncle Bernie answers:
Apple IIe video circuits use -5V. The PAL schematic can be found on one of the Apple II repositories, there is a pdf named
"apple_IIe_euro_schematic", for instance, here:
https://archive.org/details/apple_iie_euro_schematic
The real trick is to know the search term. "euro", not "pal".
BTW, just from the circuit topology of the video driver (a PNP and NPN combo) it might work without the -5V being in spec.
- Uncle Bernie
I've also posted them here previously: https://www.applefritter.com/comment/102216#comment-102216
Thanks, that's what I thought.
CVT's statement in post 5 had me confused because it wasn't clear if they were saying there was only one consumer of -5V or if there was only one chip (the opamp) and most wouldn't say a transistor is a chip. in that description. I expected video out wouldn't work without -5V, but have also never tested that. Wonder who will check first....
I had looked asimov for the schematic but will look again for euro.
I only call ICs chips (short for microchip), which I believe is the proper use. As far as the PAL video amp is concerned, it will probably still work with the -5V rail shorted to ground or at some other negative voltage a malfunctioning power supply can produce.