OqtaDrive

[gozdniJezek123]

Hello thank you for your answer, I have the internal one with raspberry Pi zero W:

Code: Select all

https://www.pcbway.com/project/shareproject/OqtaDrive_QL_v1_2_5bca1c12.html

The connection is made via approx 10cm long breadboard protoboard jumper wires, which has me worried a bit, as they are thin and the speaker is close. I've removed both internal drives for the moment and to be honest I've never tried them out on their own, since I don't have any cartridges, but that might be a good idea, to hook up the second one to see what happens.

The issues with the pre-made image for me was, if i remember correctly that raspberry pi hanged, it never got to the point where it would connect to the wireless network (i did setup the wpa_supplicant.conf and all, but no luck). I don't remember at which exact step it stopped (I used the HDMI output to check what was happening, but I forgot the exact details).

If I set the baud rate to 1000000, I don't get time="2023-12-01T14:30:20+01:00" level=debug msg="ping from QL" in the logs, but everything else looks normal. I tested the 1000000 over USB and Arduino IDE and it also did not work, so my guess is that It must be something with this particular clone.

This is the newest part of the log I think may be relevant as this seems to be some error code:
time="2023-12-01T14:39:28+01:00" level=debug msg="PUT canceled" code=1 drive=3

And if i remove the flag OLD_NANO=y I get (log no2)

[xelalex]

Hello thank you for your answer, I have the internal one with raspberry Pi zero W:

Ok, got a clearer picture now. Did you connect OqtaDrive to the Microdrive connector of drive 1 on the QL PCB?

What was the action you did when you took the latest log? I can see that drive 3 gets activated and some write operation is tried, but timed out. If the action was only a dir, then this could mean that you got noise on the R/W line. Keeping the connection wires as short as possible is recommended.

I've removed both internal drives for the moment and to be honest I've never tried them out on their own, since I don't have any cartridges, but that might be a good idea, to hook up the second one to see what happens.

I would leave that for later. It's easier to troubleshoot with the h/w drives out of the way.

If I set the baud rate to 1000000, I don't get time="2023-12-01T14:30:20+01:00" level=debug msg="ping from QL" in the logs, but everything else looks normal. I tested the 1000000 over USB and Arduino IDE and it also did not work, so my guess is that It must be something with this particular clone.

OK, then this one really can't got that fast. But that's no problem. 500k works as well.

[gozdniJezek123]

I connected it to the left one, I think it's the correct one (the guys on youtube also used the left one).

With the latest logs I only tried to dir. Now that I think about it, the error is inconsistent and I saw it before but not always, so I'm thinking it may be the noise.
I'll shorten the wires next week (the kids get in the way over the weekend ) and report back what happens. Thank you for your pointers and help! Have a nice weekend!

[spkr]

Ahhh. I figured out what I did wrong, I consistently wired A<>B mirrored. So my upper lines were lower and vice versa.
Now that I have that corrected, all is well, and I have a working oqtadrive.

Now I can finally use the oqtadrive, hopefully in parallel with the QSound and move on to get some tracker connected!

Thanks!

[gozdniJezek123]

Hello,
I've shortened the lines to approx 5cm in my internal drive, but i still have no luck. I tried to beep out the connections in case there was any bad connections, but i don't seem to find any issues. Can just someone please confirm, that the connection points on this PCB way image are correct (I assume they are, since otherwise other people would also complain).

Code: Select all

https://pcbwayfile.s3.us-west-2.amazonaws.com/web/23/02/10/2224593646256.png

So currently i have two hypothesis, it's either the Arduino or my QL. Since other people also had the most issues with the Arduino, I would firstly try to explore that path. I would not like to give on the current one yet, but does anyone have a link to Aliexpress with one that they manage to get it working in case I can't make it to work? Are there any other extra debugging steps that I can take to figure what might be the issue? The current one I'm using had to be programmed with atmega328old (via CMD) on the PC, since the Arduino IDE was not able to flash it, could that have any effect on anything? I tried the flag OLD_NANO=y multiple times, but if there is a sequence of events/commands that is important, there might be a chance I missed it (for example when updating software the flag needs to be set etc...).

If we assume to blame Arduino I see three possible failure points:
a) there is some unknown hardware limitation, as on surface majority of Arduino functionalities look like they working fine.
b) I missconfigured something and the components are just not playing along nicely.
c) there is something in the actual code for the Arduino sketch, that is not playing nicely with this particular model.

What i think its also worth mentioning is that I don't get always the same results, as sometimes I see the timeout errors and sometimes I don't. This might be some timing issue?
Please find attached the latest logs.

[xelalex]

So currently i have two hypothesis, it's either the Arduino or my QL.

The PCB works fine, so no worries there. I'd start with the Nano. The logs show that your daemon installation works, and serial communication with the Nano is ok as well, but something between Nano and QL is wrong. Maybe one or several GPIOs of the Nano are damaged. The Nano is easy to replace, so I'd try that. I've never had problems with the cheap ones, just don't buy too cheap. I usually pick one in the 8 to 10 Euro range, and they have worked well so far. Also, even though I'm sure you've done this a dozen times already, it doesn't hurt to yet again check the connection to the QL PCB, to make sure it's not connected upside down or mirrored. Happens to the best

[gozdniJezek123]

Hello well this is awkward, of course you're right! I cleaned the socket with contact cleaner, bent the breadboard protoboard jumper wires end to make better contact, and while fiddling around i noticed that R/W socket was a bit looser than the others, so i bent the leg of a cable a bit more. And now it works! I'm so happy and embarrassed at the same time. In my defense i couldn't find the R100 on the board to check the contact on the second point, but this was really a rookie mistake. Thank you for your help, I really appreciate it! Lesson learned check the damn wires for the 14th time!

[xelalex]

Glad to hear it's working! Yeah, sometimes it's the really basic stuff that goes wrong, and all the time we think it must be something really complicated. Been there too... Have loads of fun with your OqtaDrive!

[gozdniJezek123]

Another update, I guess the contact cleaner dried out and it lost contact again, I'll have to add some solder to the points of the wire termination. But I think I'll manage. Just out of curiosity, what type of wires do other people use for the connection? And thank you once again, I'm looking forward to using it!

[gozdniJezek123]

Hey, It's me again. I'm really not sure what is happening, but there might also be a fault with my QL.

At the bottom left side there is a cut trace, but I'm not sure if this is some sort of mod or it is a fault. I'm still struggling with my damn wires and I'm not sure if they are to be blamed. Yesterday that I managed to get it to "work", I cleaned this part of the PCB with IPA and It would appear that i made contact on some broken trace or something, at least that is my hopium now. I think my board revision is Issue 5.

Also is this the only schema:

Code: Select all

https://www.sinclairql.net/srv/QLSchematicIssue5.gif

It's a bit difficult to read and I don't know why R98 is unpopulated, is this ok?