RetroBrew Computers Forum
Discussion forum for the RetroBrew Computers community.

Home » RBC Forums » General Discussion » Jackalope Progress
Jackalope Progress [message #2502] Fri, 21 April 2017 13:09 Go to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
Jackalope boards arrived today!

First ERRATA: 5 Volts is connected to wrong pin on J1. On Gryphon and Xagdin it was pin 1. On Jackalope it is pin 4. Beware - power connections through J1 have to be changed
Re: Jackalope Progress [message #2503 is a reply to message #2502] Fri, 21 April 2017 13:56 Go to previous messageGo to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
ERRATA #2: None of the TTL or larger chips have silk screen locations. Looks like a bug with KiCad. I assume the locations are the same as Xagdin. This needs to be fixed in next version.
Re: Jackalope Progress [message #2504 is a reply to message #2502] Fri, 21 April 2017 14:18 Go to previous messageGo to next message
mikemac is currently offline  mikemac
Messages: 249
Registered: March 2017
Senior Member
yoda wrote on Fri, 21 April 2017 13:09
Jackalope boards arrived today!



Yes it did! Along with 3 breakout boards I assume for the FPGA, video, and Ethernet chips.

So who's a reputable supplier of 68030s, 68881s, 68681s, ... Just in case I go crazy and try yo populate my board. Smile



Mike

[Updated on: Fri, 21 April 2017 14:19]

Report message to a moderator

Re: Jackalope Progress [message #2505 is a reply to message #2502] Fri, 21 April 2017 14:49 Go to previous messageGo to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
Well I got my parts off eBay a long while ago - I buy here and there and accumulate.

I hope you are not going to just sit on the board. The whole idea of the limited set of boards is to help debug and validate the design, so get off the side lines Razz
Re: Jackalope Progress [message #2506 is a reply to message #2502] Fri, 21 April 2017 14:57 Go to previous messageGo to next message
mikemac is currently offline  mikemac
Messages: 249
Registered: March 2017
Senior Member
yoda wrote on Fri, 21 April 2017 13:09

First ERRATA: 5 Volts is connected to wrong pin on J1. On Gryphon and Xagdin it was pin 1. On Jackalope it is pin 4. Beware - power connections through J1 have to be changed


According to http://pinouts.ru/Power/BigPower_pinout.shtml, the Jackalope board is correct. Luckily the socket is symmetrical so you can insert it which ever way works for you.



Mike
Re: Jackalope Progress [message #2509 is a reply to message #2506] Fri, 21 April 2017 16:33 Go to previous messageGo to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
mikemac wrote on Fri, 21 April 2017 16:57
yoda wrote on Fri, 21 April 2017 13:09

First ERRATA: 5 Volts is connected to wrong pin on J1. On Gryphon and Xagdin it was pin 1. On Jackalope it is pin 4. Beware - power connections through J1 have to be changed


According to http://pinouts.ru/Power/BigPower_pinout.shtml, the Jackalope board is correct. Luckily the socket is symmetrical so you can insert it which ever way works for you.


Yes I know - but it is different than the previous board so anyone that worked on Gryphon or Xagdin would have problems. Also it is strange that Xagdin schematic is same as Jackalope but board connector is not so something happened in translation.
Re: Jackalope Progress [message #2511 is a reply to message #2506] Fri, 21 April 2017 21:16 Go to previous messageGo to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
Actually I went back an looked at the ATX power supply I have. If you look at the silk screen on the board, the double white bar should represent the flat side of the connector so 5V should be towards the 68230. The issue is the square pin should be at the bottom away from the 68230. Either the square pin is wrong or the silk screen with respect to pin one is wrong.
Re: Jackalope Progress [message #2513 is a reply to message #2502] Sat, 22 April 2017 08:07 Go to previous messageGo to next message
lynchaj is currently offline  lynchaj
Messages: 1017
Registered: June 2016
Senior Member
Hi, Yes, J1 is reversed compared to Xagdin. It was a bug on Xagdin so it is flipped on Jackalope to match the footprint.

Thanks, Andrew Lynch
Re: Jackalope Progress [message #2514 is a reply to message #2503] Sat, 22 April 2017 08:12 Go to previous messageGo to next message
lynchaj is currently offline  lynchaj
Messages: 1017
Registered: June 2016
Senior Member
Yes, looks like a bug with the footprints. Replaced the old DIP footprints with new ones with bigger pads and they conveniently put the labels on the back silkscreen Sad Learn something new about KiCAD everyday.
Re: Jackalope Progress [message #2515 is a reply to message #2514] Sat, 22 April 2017 08:14 Go to previous messageGo to next message
lynchaj is currently offline  lynchaj
Messages: 1017
Registered: June 2016
Senior Member
Let's make a list of errata and they'll get fixed the next version. Hopefully its minor stuff. Always happens though...
Re: Jackalope Progress [message #2518 is a reply to message #2513] Sat, 22 April 2017 09:29 Go to previous messageGo to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
lynchaj wrote on Sat, 22 April 2017 10:07
Hi, Yes, J1 is reversed compared to Xagdin. It was a bug on Xagdin so it is flipped on Jackalope to match the footprint.

Thanks, Andrew Lynch


I think the pin designations in Xagdin were wrong but now I think the silk screen is wrong now. Look at photo. To get the voltages right the molex connector must have the flat side towards the inside where as the silk screen with double line is toward the outside (which I think means flat side out) is wrong? You have to flip connector from silk screen or you are going to apply 12 volts to the 5V rail.

index.php?t=getfile&id=375&private=0
  • Attachment: IMG_0183.JPG
    (Size: 418.67KB, Downloaded 1809 times)
Re: Jackalope Progress [message #2519 is a reply to message #2515] Sat, 22 April 2017 09:31 Go to previous messageGo to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
lynchaj wrote on Sat, 22 April 2017 10:14
Let's make a list of errata and they'll get fixed the next version. Hopefully its minor stuff. Always happens though...


I will make an errata list on the wiki once I start uploading to there.
Re: Jackalope Progress [message #2522 is a reply to message #2519] Sat, 22 April 2017 09:41 Go to previous messageGo to next message
Andrew B is currently offline  Andrew B
Messages: 467
Registered: October 2015
Location: Near Redmond, WA
Senior Member
Administrator
Do you guys need a new page in the 'dev' wiki section?
Re: Jackalope Progress [message #2523 is a reply to message #2519] Sat, 22 April 2017 09:42 Go to previous messageGo to next message
lynchaj is currently offline  lynchaj
Messages: 1017
Registered: June 2016
Senior Member
Yes, those darn molex power connectors have caused many problems on boards! I think John Coffman made a new footprint with the little wedges cut out on the sides so the connector gets installed right the first time. I'll fix it in the update. Keep digging and posting errata. We'll sort it out. Thanks, Andrew Lynch

The good news is your 12V (yellow wire) is in the right place so that's good news. It is a good practice to always verify power and ground *before* installing components or applying power to board! Things happen.

[Updated on: Sat, 22 April 2017 09:44]

Report message to a moderator

Re: Jackalope Progress [message #2526 is a reply to message #2522] Sat, 22 April 2017 10:12 Go to previous messageGo to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
Andrew B wrote on Sat, 22 April 2017 11:41
Do you guys need a new page in the 'dev' wiki section?

Hi Andrew

I will be creating a page under my id in builders for bringup. There is a source code repository in GitLab the Ewout set up.
Re: Jackalope Progress [message #2527 is a reply to message #2514] Sat, 22 April 2017 10:15 Go to previous messageGo to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
lynchaj wrote on Sat, 22 April 2017 10:12
Yes, looks like a bug with the footprints. Replaced the old DIP footprints with new ones with bigger pads and they conveniently put the labels on the back silkscreen Sad Learn something new about KiCAD everyday.


I actually prefer the old footprint with smaller pads. I find them easier to solder and they use less solder in end, jut my preference in the future.
Re: Jackalope Progress [message #2530 is a reply to message #2523] Sat, 22 April 2017 12:53 Go to previous messageGo to next message
jcoffman is currently offline  jcoffman
Messages: 332
Registered: October 2015
Senior Member
Ahh ... those Molex power connectors.

I recall building an early board, having some question about where +5v power came in (red wire), and discovered that the pin used was at the wrong end. Pin 4 correctly carries +5v. The square pin in the footprint is pin 1, which carries +12v. At least there is no confusion over pins 2 and 3, which both are GND. Smile

--John
Re: Jackalope Progress [message #2536 is a reply to message #2530] Sat, 22 April 2017 18:30 Go to previous messageGo to next message
lynchaj is currently offline  lynchaj
Messages: 1017
Registered: June 2016
Senior Member
Thanks John,
If I use the right pins to the right signals, the footprint has to be right since there is no way to "flip" it relative to the numbered pins.

The Xagdin power input connector had the right pins but the signals were flipped. The darn footprint is just ambiguous. I don't think the double lines actually mean anything.

Do the double lines denote the flat side of the bornier connector?

[Updated on: Sat, 22 April 2017 18:36]

Report message to a moderator

Re: Jackalope Progress [message #2537 is a reply to message #2536] Sat, 22 April 2017 19:20 Go to previous messageGo to next message
mikemac is currently offline  mikemac
Messages: 249
Registered: March 2017
Senior Member
lynchaj wrote on Sat, 22 April 2017 18:30
Thanks John,
If I use the right pins to the right signals, the footprint has to be right since there is no way to "flip" it relative to the numbered pins.

The Xagdin power input connector had the right pins but the signals were flipped. The darn footprint is just ambiguous. I don't think the double lines actually mean anything.

Do the double lines denote the flat side of the bornier connector?


I'd suggest using the symbol from backplane-8 or backplane-3. They are correct and use the beveled corners that match the socket. I think they're much clearer.

The one you're using looks like it came from backplane-12. It has +5V on pin 1.

6x0x-ATX-6U uses a combination of bevels and extra straight line.



Mike
Re: Jackalope Progress [message #2553 is a reply to message #2537] Mon, 24 April 2017 12:44 Go to previous messageGo to next message
lynchaj is currently offline  lynchaj
Messages: 1017
Registered: June 2016
Senior Member
Good progress so far, am looking forward to seeing photos in the builder's logs on how Jackalope build & test proceeding. Good luck! Thanks, Andrew Lynch
Re: Jackalope Progress [message #2554 is a reply to message #2553] Mon, 24 April 2017 13:06 Go to previous messageGo to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
I should have something up in the next couple of days - have to work tonight so won't make much progress. Board is assembled and initial smoke test was good - no fires, smoke etc and 5Volts at appropriate pins. Next up is plug in minimal components and start with ROMLoop with logic analyzer and verify basic signals, then get serial port working. Hope to get a lot of that done tomorrow night if I don't run into too many issues. I have upgraded my tool chain to the latest and greatest and need to do verification of it as well. We should know by mid week if it is alive or not.

Dave

[Updated on: Mon, 24 April 2017 13:07]

Report message to a moderator

Re: Jackalope Progress [message #2560 is a reply to message #2554] Wed, 26 April 2017 19:08 Go to previous messageGo to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
Just a quick update - FlashLoop is working. Next up is serial port testing. Will try to get to that tomorrow or over the weekend.


index.php?t=getfile&id=380&private=0
Re: Jackalope Progress [message #2579 is a reply to message #2560] Sun, 30 April 2017 14:37 Go to previous messageGo to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
New Update:

Successfully have serial port and sramtest now check out. I had to add a pull up to DTACK_SER* connected to unused 10K pull up on RN3. I will probably have to do the same for DTACK_PAR* as well. Not seeing a good reset yet so I may have to play around with the resistors on the 555 timer to get a good reset pulse. Hello world is going off into never never land so I have to do some more debug on that. Once I have Hello World going then bringing up the basic monitor should be easy as it means I have full C development environment working. I will be traveling this week so won't get much done this week.
Re: Jackalope Progress [message #2580 is a reply to message #2579] Sun, 30 April 2017 14:44 Go to previous messageGo to next message
mikemac is currently offline  mikemac
Messages: 249
Registered: March 2017
Senior Member
Sounds like you're making good progress! Thanks!!


Mike
Re: Jackalope Progress [message #2599 is a reply to message #2580] Tue, 02 May 2017 00:20 Go to previous messageGo to next message
ewout is currently offline  ewout
Messages: 30
Registered: October 2015
Member
Wow nice work. Just received the board last weekend (damn customs) and still waiting on some parts (sockets/headers especially) too.
Re: Jackalope Progress [message #2729 is a reply to message #2599] Thu, 11 May 2017 04:21 Go to previous messageGo to next message
lynchaj is currently offline  lynchaj
Messages: 1017
Registered: June 2016
Senior Member
Hi, does anyone have any updates on Jackalope? Thanks, Andrew Lynch
Re: Jackalope Progress [message #2733 is a reply to message #2729] Thu, 11 May 2017 11:26 Go to previous messageGo to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
Have been traveling and next week too - so might not be anymore progress for a couple of weeks. I am working on Hello World loop. I think I have messed up the linker script that I need to look at. Once Hello World in C works the basic visual monitor should work immediately then on to debugging individual I/O chips, auto vector interrupts, etc.
Re: Jackalope Progress [message #2735 is a reply to message #2733] Thu, 11 May 2017 11:54 Go to previous messageGo to next message
lynchaj is currently offline  lynchaj
Messages: 1017
Registered: June 2016
Senior Member
OK, Thanks
Re: Jackalope Progress [message #2863 is a reply to message #2735] Mon, 22 May 2017 06:37 Go to previous messageGo to next message
lynchaj is currently offline  lynchaj
Messages: 1017
Registered: June 2016
Senior Member
Hi
Any news from anyone? Anything?
Re: Jackalope Progress [message #2994 is a reply to message #2863] Tue, 06 June 2017 15:50 Go to previous messageGo to next message
lynchaj is currently offline  lynchaj
Messages: 1017
Registered: June 2016
Senior Member
Hi
Any news on the Jackalope builds? Thanks, Andrew Lynch
Re: Jackalope Progress [message #3078 is a reply to message #2994] Wed, 14 June 2017 07:38 Go to previous messageGo to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
Well I have run into a snag. It appears that I have a flaky address line. I don't know if it is CPU or socket yet. It works for a while then as it warms up it crashes to the point I can't even boot the ROM image. I need to put logic analyzer on and see which address line is not working. It looks like it is one of the high address lines. I know John Coffman said he had problems with an 030 socket and had to drill out one of the pins (I hope it doesn't come to that).
Re: Jackalope Progress [message #3079 is a reply to message #3078] Wed, 14 June 2017 07:54 Go to previous messageGo to next message
jcoffman is currently offline  jcoffman
Messages: 332
Registered: October 2015
Senior Member
RE: 68030 socket

My bad experience was with a cheap, old, socket from eBay. The socket was so old that several pins were corroded and I did not notice same when I soldered the socket to the PC board. When the completed board had problems, an inspection of solder connections through a jeweler's 8X magnifier found the bad connections. It took a lot of scraping and a lot of solder flux on the offending pins. Luckily, they were all in a small region where, perhaps, a drop of condensation had promoted the corrosion. Finally, a solder bridge was established to the PC board, and inspection of the pins' sockets in that region on the top showed no corrosion that the 68030 chip could be confidently re-inserted. This re-vamped board is running Will's 68000 Linux port successfully.
Re: Jackalope Progress [message #3080 is a reply to message #3079] Wed, 14 June 2017 08:14 Go to previous messageGo to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
My sockets were from eBay as well - I thought I had checked them because I had remembered reading your post. I will use the logic analyzer and see if I can identify the region to inspect closely. Thanks for the reply.
Re: Jackalope Progress [message #3089 is a reply to message #3080] Wed, 14 June 2017 10:25 Go to previous messageGo to next message
lynchaj is currently offline  lynchaj
Messages: 1017
Registered: June 2016
Senior Member
Well that's good news in a fashion... work continues. Kind of sucks about the bad address line though. Maybe it's a cold solder joint? Won't know for sure until you know which line is bad though.
Re: Jackalope Progress [message #3102 is a reply to message #3089] Tue, 20 June 2017 20:47 Go to previous messageGo to next message
Andrew B is currently offline  Andrew B
Messages: 467
Registered: October 2015
Location: Near Redmond, WA
Senior Member
Administrator
There is a new wiki page for Jackalope at - https://www.retrobrewcomputers.org/doku.php?id=dev:boards:ja ckalope_68030:start

I went through the previous forum thread and uploaded the .pdfs and KiCAD files that should correspond to the boards that were mailed out to everyone.

Happy testing!
Re: Jackalope Progress [message #3158 is a reply to message #3102] Tue, 04 July 2017 19:36 Go to previous messageGo to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
Here is a quick update and teaser. Finally figured out the wild jump. It turns out my RAM chip had a shorted address line A9 - never seen that before. Anyways, replaced and Hello World plus base Monitor is working - need to start working on other devices now.

index.php?t=getfile&id=532&private=0

Re: Jackalope Progress [message #3196 is a reply to message #3158] Fri, 07 July 2017 16:43 Go to previous messageGo to next message
Andrew B is currently offline  Andrew B
Messages: 467
Registered: October 2015
Location: Near Redmond, WA
Senior Member
Administrator
Exciting!
Re: Jackalope Progress [message #3206 is a reply to message #3196] Sat, 08 July 2017 19:30 Go to previous messageGo to next message
yoda is currently offline  yoda
Messages: 125
Registered: October 2015
Location: Cedar Rapids, IA
Senior Member
Update: Parallel port, timers (68230), RTC, Keyboard controller, and sound chip are working. A few bugs to clean up and I will start putting in the GitHub repository. I will be working on interrupts and IDE interface next. Then the dreaded DRAM work. After that I am moving this to a lower priority and see if anyone is really working on this. I fear not. All that will be left then would be ethernet and video to check out. I will do that after I catch up on other projects on my workbench.
Re: Jackalope Progress [message #5945 is a reply to message #2502] Mon, 11 February 2019 12:37 Go to previous messageGo to next message
stevemoody is currently offline  stevemoody
Messages: 13
Registered: February 2019
Junior Member
I've been looking at designing a board based on the 68030 and was looking at the Gryphon/Jackalope schematics as a starting point rather than having to start from scratch.

I've got a question about the DRAM controller being used here. Considering the last post has yoda's comment "Then the dreaded DRAM work" I was wondering if there's been any luck getting it working?

Looking at the schematic for the DP8422V it seems that the design seems to have been taken out of the Interfacing with the 68030 microprocessor app note. After spending some time reading through the datasheet and app note it looks like there's possibly an error in the schematic or I'm missing something

The app note design is for 32M of RAM using 2 banks for DRAM not 1. Currently on the Jackalope the address line A4 is connected to B1 (pin 40) on the DP8422.

From the app note:
Quote:
The memory banks are interleaved on every four word (32-bit word) boundry. This means that the address bit (A4) is tied to the bank select input of the DP8422A (B1).


From my understanding of the app note this would RAS 0/1 and CAS0/1 to access each bank depending on the state of A4. It seems if it's not using a dual bank design then having A4 connected to B1 means that the A4 line is not going to be used when accessing ram. Also only using 1 bank of ram instead of 2 would restrict you to only 16M of ram instead of 32M.

I could be that I'm reading this completely wrong as the datasheet is not the easiest to read through and I've still not worked out the programming bits correctly yet but looking at the schematic as it is I'm struggling to see how it's going to work.

[Updated on: Mon, 11 February 2019 12:38]

Report message to a moderator

Re: Jackalope Progress [message #5946 is a reply to message #5945] Mon, 11 February 2019 17:02 Go to previous messageGo to previous message
mikemac is currently offline  mikemac
Messages: 249
Registered: March 2017
Senior Member
If you haven't already, you might check out the design for the KISS-68030. It works!

Jackalope never really got any momentum towards working out the bugs.



Mike
Previous Topic: Home Computer??
Next Topic: Z1RCC, a RC2014-compatible, RomWBW-capable 2"x4" Z180 SBC


Current Time: Thu Mar 28 01:27:45 PDT 2024

Total time taken to generate the page: 0.00858 seconds