and another CD32 Riser! - Page 4

kolla · 09 March 2018, 12:52

Quote:

Originally Posted by zerohp

I think you only need to shift 3.3V to 5V to connect a RPi to the CD32 TX/RX. RPi doesn't have +12/-12 rs232 signaling, and neither does the CD32.

Exactly - if I was to make "full blown" solution, it would be max232 on the CD32 side and max3232 (3.3V version of max232) on the Pi, but that seems overkill. Electronics is not my field, so if anyone has suggestion to what parts would be needed to do signal level shifting between 3.3V and 5V, I would be keen to know.

Bprepared · 09 March 2018, 12:58

How about one of these?
https://www.rapidonline.com/Adafruit...RoCrUcQAvD_BwE
or
https://www.sparkfun.com/products/12009

plasmab · 09 March 2018, 13:18

Quote:

Originally Posted by kolla

Exactly - if I was to make "full blown" solution, it would be max232 on the CD32 side and max3232 (3.3V version of max232) on the Pi, but that seems overkill. Electronics is not my field, so if anyone has suggestion to what parts would be needed to do signal level shifting between 3.3V and 5V, I would be keen to know.

Frankly I’d just plug a $2 usb TTL level serial port into the pi. 3 wires and you are done

kolla · 09 March 2018, 13:21

@BPrepared

Hah, funny, I just dropped by a colleague who has more electronics foo, and he suggested exactly that

He also had some other, simpler suggestions, like letting TX from pi go directly to RX on CD32, and just level down the TX from CD32 with a resistor, sort of like this...

Code:

CD32 RX -------------------------- TX Pi
     TX -----[2k ohm]------------- RX
                           |
                       [4k ohm]
                           |
    GND -------------------------- GND

kolla · 09 March 2018, 13:24

Quote:

Originally Posted by plasmab

Frankly I’d just plug a $2 usb TTL level serial port into the pi. 3 wires and you are done

Not what I aim for. I want the serial _console_ of the Pi on this link. The purpose is not just to give the CD32 an uplink, but to also have console for the companion Pi.

Bprepared · 09 March 2018, 13:28

Quote:

Originally Posted by kolla

@BPrepared

Hah, funny, I just dropped by a colleague who has more electronics foo, and he suggested exactly that

He also had some other, simpler suggestions, like letting TX from pi go directly to RX on CD32, and just level down the TX from CD32 with a resistor, sort of like this...

Code:

CD32 RX -------------------------- TX Pi
     TX -----[3k ohm]------------- RX
                           |
                       [1k ohm]
                           |
    GND -------------------------- GND

Lol, yes at the current and load values here a simple potential divider should work in theory too
Edit: although I think those resistors are round the wrong way as that shown would give you 1.25V levels, but even with them swapped you would get 3.75V not 3.3V

plasmab · 09 March 2018, 14:31

Quote:

Originally Posted by kolla

Not what I aim for. I want the serial _console_ of the Pi on this link. The purpose is not just to give the CD32 an uplink, but to also have console for the companion Pi.

That’s just a software thing on the pi though. You can choose what TTY the console lives on. Fair enough though.

kolla · 10 March 2018, 16:11

Quote:

Originally Posted by Bprepared

Edit: although I think those resistors are round the wrong way as that shown would give you 1.25V levels, but even with them swapped you would get 3.75V not 3.3V

Yes, hence "something like this", a wee bit of math to find the exact values is needed

How about 4k ohm and 2k ohm? Or 2k ohm and 1k ohm? 5V * 4/(2+4) = 5V * 2/(1+2) = 3,33333V (updated my "drawing").

Bprepared · 10 March 2018, 20:42

Looks spot in to me

Bprepared · 14 March 2018, 15:16

Well, after several hours of head scratcing last night I've come to the conclusion that a K150 PIC programmer doesn't cut it when it comes to the ICSP of the SSOP version of 16F628A's (it works fine on the DIPs I've tried!) I have therefore ordered a PICKit 3 to try that instead, if that doesn't work I'm completely at a loss as to why I can't program the PIC using ICSP and will switch to DIPs instead

Stedy · 14 March 2018, 21:59

Quote:

Originally Posted by Bprepared

Well, after several hours of head scratcing last night I've come to the conclusion that a K150 PIC programmer doesn't cut it when it comes to the ICSP of the SSOP version of 16F628A's (it works fine on the DIPs I've tried!) I have therefore ordered a PICKit 3 to try that instead, if that doesn't work I'm completely at a loss as to why I can't program the PIC using ICSP and will switch to DIPs instead

Hi,

Connect pin 10 of the PIC16F628 to ground and try again.
Factory fresh, the PIC16F628 supports Low Voltage Programming (LVP) entry to which is controlled by the LVP/PGM pin, which happens to be pin 10 of that device. If you are using conventional high voltage programming, >10V on MCLR, you must pull this to ground.

Probably when you put the DIP parts in a programmer, it drives PGM/LVP.

Ian

Bprepared · 15 March 2018, 00:21

Thanks Ian, guess I didn't read that part of the documentation

I think you are on to something here, unfortunately I can't find my jumper wires to test this at the moment, but the K150 doesn't do LVP. The PICkit should arrive pretty soon anyway and that does support LPV.

solidcore · 15 March 2018, 11:24

@bprepared, I assume it's one of the Chinese knock-off K150 PIC programmers? : what o/s do you use it on, I managed to find working windows 10 drivers for mine which went from killing chips to programming them perfectly each time, although regarding LVP I'm unsure BUT the k150 docs say it does support the 16F628. So perhaps this is a driver issue?

I use a combo of:

Windows 10:
+ Prolific Comm Driver v3.3.11.152 ( http://www.magicdriver.com/driver158...3.3.2.102.html )
+ DIyPAcK25EP2.zip for the programming software. ( http://www.ozitronics.com/download/DIYpack25EP2.zip )

Any other combination ended up with dead chips.

Bprepared · 15 March 2018, 13:14

Quote:

Originally Posted by solidcore

@bprepared, I assume it's one of the Chinese knock-off K150 PIC programmers? : what o/s do you use it on, I managed to find working windows 10 drivers for mine which went from killing chips to programming them perfectly each time, although regarding LVP I'm unsure BUT the k150 docs say it does support the 16F628. So perhaps this is a driver issue?

I use a combo of:

Windows 10:
+ Prolific Comm Driver v3.3.11.152 ( http://www.magicdriver.com/driver158...3.3.2.102.html )
+ DIyPAcK25EP2.zip for the programming software. ( http://www.ozitronics.com/download/DIYpack25EP2.zip )

Any other combination ended up with dead chips.

Pretty much the same combo as me then

I got mine from Hobbykit.com, which seems cut above the standard Chinese knock-off I previously used (I had taken the chips tried with that one out of circulation by default), and they actually provide a link to the know working windows 10 drivers and MicroBrn along with the program delay tweak needed to get the 16F628A working. As I say it does not seem to be a 16F628 compatability issue (DIPs program just fine), its that the K150 doesn't support LVP and I didn't realise that the default config for the chips was LVP unless PGM was grounded - this was my first venture into PIC ICSP.

solidcore · 15 March 2018, 13:34

Fair point mate. Good luck with the other programmer or if you get round to trying Ian's suggestion let us know how it fairs .

Bprepared · 15 March 2018, 14:29

Oh I do intend to test Ian's solution as compatability with as many different programmers as possible is always desirable. If it works there will be one final spin of the board with a resistor footprint for shorting pin 10 to gnd when nessecary. I just can't find my 'solidcore' breadboard jumper wires, only stranded ones which I don't trust on a surface mount board and are a pain to solder to find pitch SMD IC's without potentially shorting other pins by accident.

Bprepared · 16 March 2018, 15:22

SUCCESS!!!!
The PIC is now programmed using the PICKit 3 so once I get back to my cave on Tuesday I will solder down the remaining components (I only soldered down the PIC to check the programming circuit initially) and all being well and tested I'll get the sources/BOM up on GitHub, and post some boards out to those who have requested them

Bprepared · 22 March 2018, 14:07

Well I finished soldering up the Rev 2.1 last night, and all is well with the Keyboard interface...finally!
However, for some reason the Amiga RGB port is on the fritz as although colours are present and correct, the screen was scrolling vertically

I belled out the sync lines and everything was in order, so I plugged the Rev2 board back into the riser to check that it wasn't the CD32 (I really should get around to recapping it now that the A1200 is done), but it was a nice stable picture, so not the consoles fault

The Amiga RGB is the simiplest hookup on the board as the sync lines are straight connections, aside for tapping off to Stedy's 3.3V conversion circuit, so that I think is my next point of investigation.

solidcore · 26 March 2018, 23:13

Scrolling vertically? (that reminds me of something where the horizontal was stable but v sync rolled around and around) Hmm.

Bprepared · 27 March 2018, 12:08

A problem with the V-sync line is what I think too, hence the belling out, but no joy there, all seemed good, but I didn't check for a short to ground due to being very tired and frustrated before packing up to head back to my normal place of residence, will be having another look tonight.

09 March 2018, 12:58	#62
Bprepared Registered User Join Date: Feb 2017 Location: Cirencester, UK Posts: 113	How about one of these? https://www.rapidonline.com/Adafruit...RoCrUcQAvD_BwE or https://www.sparkfun.com/products/12009 Last edited by Bprepared; 09 March 2018 at 13:09.

09 March 2018, 13:21	#64
kolla Banned Join Date: Nov 2007 Location: Trondheim, Norway Posts: 1,893	@BPrepared Hah, funny, I just dropped by a colleague who has more electronics foo, and he suggested exactly that He also had some other, simpler suggestions, like letting TX from pi go directly to RX on CD32, and just level down the TX from CD32 with a resistor, sort of like this... Code: CD32 RX -------------------------- TX Pi TX -----[2k ohm]------------- RX \| [4k ohm] \| GND -------------------------- GND Last edited by kolla; 10 March 2018 at 16:38.

15 March 2018, 11:24	#73
solidcore Registered User Join Date: Sep 2017 Location: birmingham, uk Posts: 548	@bprepared, I assume it's one of the Chinese knock-off K150 PIC programmers? : what o/s do you use it on, I managed to find working windows 10 drivers for mine which went from killing chips to programming them perfectly each time, although regarding LVP I'm unsure BUT the k150 docs say it does support the 16F628. So perhaps this is a driver issue? I use a combo of: Windows 10: + Prolific Comm Driver v3.3.11.152 ( http://www.magicdriver.com/driver158...3.3.2.102.html ) + DIyPAcK25EP2.zip for the programming software. ( http://www.ozitronics.com/download/DIYpack25EP2.zip ) Any other combination ended up with dead chips. Last edited by solidcore; 15 March 2018 at 11:29.

22 March 2018, 14:07	#78
Bprepared Registered User Join Date: Feb 2017 Location: Cirencester, UK Posts: 113	It works....sort of! Well I finished soldering up the Rev 2.1 last night, and all is well with the Keyboard interface...finally! However, for some reason the Amiga RGB port is on the fritz as although colours are present and correct, the screen was scrolling vertically I belled out the sync lines and everything was in order, so I plugged the Rev2 board back into the riser to check that it wasn't the CD32 (I really should get around to recapping it now that the A1200 is done), but it was a nice stable picture, so not the consoles fault The Amiga RGB is the simiplest hookup on the board as the sync lines are straight connections, aside for tapping off to Stedy's 3.3V conversion circuit, so that I think is my next point of investigation.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
CD32 Expansion Riser	solidcore	Hardware mods	84	05 October 2017 15:33
Helping out an Abused A4000. Zorro riser board stuck?	ElectroBlaster	support.Hardware	4	11 September 2017 00:37
A1200 expansion port "riser" question	diablothe2nd	support.Hardware	9	06 June 2012 07:28
A500 CPU riser	Eamoe	support.Hardware	5	31 January 2011 23:31

10 March 2018, 20:42	#69
Bprepared Registered User Join Date: Feb 2017 Location: Cirencester, UK Posts: 113	Looks spot in to me

14 March 2018, 15:16	#70
Bprepared Registered User Join Date: Feb 2017 Location: Cirencester, UK Posts: 113	Well, after several hours of head scratcing last night I've come to the conclusion that a K150 PIC programmer doesn't cut it when it comes to the ICSP of the SSOP version of 16F628A's (it works fine on the DIPs I've tried!) I have therefore ordered a PICKit 3 to try that instead, if that doesn't work I'm completely at a loss as to why I can't program the PIC using ICSP and will switch to DIPs instead

15 March 2018, 00:21	#72
Bprepared Registered User Join Date: Feb 2017 Location: Cirencester, UK Posts: 113	Thanks Ian, guess I didn't read that part of the documentation I think you are on to something here, unfortunately I can't find my jumper wires to test this at the moment, but the K150 doesn't do LVP. The PICkit should arrive pretty soon anyway and that does support LPV.

15 March 2018, 13:34	#75
solidcore Registered User Join Date: Sep 2017 Location: birmingham, uk Posts: 548	Fair point mate. Good luck with the other programmer or if you get round to trying Ian's suggestion let us know how it fairs .

15 March 2018, 14:29	#76
Bprepared Registered User Join Date: Feb 2017 Location: Cirencester, UK Posts: 113	Oh I do intend to test Ian's solution as compatability with as many different programmers as possible is always desirable. If it works there will be one final spin of the board with a resistor footprint for shorting pin 10 to gnd when nessecary. I just can't find my 'solidcore' breadboard jumper wires, only stranded ones which I don't trust on a surface mount board and are a pain to solder to find pitch SMD IC's without potentially shorting other pins by accident.

16 March 2018, 15:22	#77
Bprepared Registered User Join Date: Feb 2017 Location: Cirencester, UK Posts: 113	SUCCESS!!!! The PIC is now programmed using the PICKit 3 so once I get back to my cave on Tuesday I will solder down the remaining components (I only soldered down the PIC to check the programming circuit initially) and all being well and tested I'll get the sources/BOM up on GitHub, and post some boards out to those who have requested them

26 March 2018, 23:13	#79
solidcore Registered User Join Date: Sep 2017 Location: birmingham, uk Posts: 548	Scrolling vertically? (that reminds me of something where the horizontal was stable but v sync rolled around and around) Hmm.

27 March 2018, 12:08	#80
Bprepared Registered User Join Date: Feb 2017 Location: Cirencester, UK Posts: 113	A problem with the V-sync line is what I think too, hence the belling out, but no joy there, all seemed good, but I didn't check for a short to ground due to being very tired and frustrated before packing up to head back to my normal place of residence, will be having another look tonight.

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