Full size H17 controller with VDIP1 and HSFE support!

36 views
Skip to first unread message

Norberto Collado

unread,
Sep 8, 2021, 2:27:16 AM9/8/21
to se...@googlegroups.com

I decided to take the current H17 design (http://koyado.com/Heathkit/H8-H17-H37-H67-USB_files/H8_H17_Controller.png) that includes the VDIP1 + HSFE and dropped into a full size board. This allows to have the H37/H67 by itself. Also allows to move the VDIP1 from the CPU into the H17 to allow more space between the H8 Control Panel and the CPU board without wasting a backplane slot.

 

Also supports in just having the H17 with HSFE and VDIP1 in the H8 if not interested on the H37/H67 combo. For me the best configuration is the H17 + H67 (174Q, 170Q) as we have a lot of files in H17 H8D format.

Thanks,

Norberto

 

 

Glenn Roberts

unread,
Sep 8, 2021, 5:40:00 AM9/8/21
to se...@googlegroups.com
Good decision. The piggyback arrangement was innovative but still essentially requires two slots due to the thickness, so no real savings for the trouble.

Once I get past VCF I want to get back to documenting/sharing the “jukebox” and baselining the associated software (future REMarks issue! REMarks is just on summer hiatus).  Once you store your H8D images in the jukebox you have no need for the H17 board or drives or floppy disks. It is a pure software implementation - all you need is the z67IDE+ with a z80 cpu board. I hope also to be able to develop a CP/M jukebox (currently only HDOS). I currently have over 500 H8D “disks” (basically Les’ archive plus some of my own) any of which I can instantly call up with just a MOUNT command … and the jukebox has room for a total of 4,096 of them

Sent from my iPad

On Sep 8, 2021, at 2:27 AM, Norberto Collado <norberto...@koyado.com> wrote:


--
You received this message because you are subscribed to the Google Groups "SEBHC" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/1F68EF73-B767-463C-B993-F428C28D1E01%40koyado.com.

Les Bird

unread,
Sep 8, 2021, 7:48:43 AM9/8/21
to SEBHC
Jukebox sounds like the way to go. I still have not built out the Z67IDE board. I've been looking at sourcing parts for it, would be nice to have that instead of using the GIDE set up which works only in CP/M.

Les

Les Bird

unread,
Sep 8, 2021, 8:08:08 AM9/8/21
to SEBHC
Norby, I've been thinking about a new CPU board design with 64K, ORG0, H17 and H8-4 (dual port). Has this been explored already? I think that encompasses the baseline H8 system.

Les

Les Bird

unread,
Sep 8, 2021, 8:45:20 AM9/8/21
to SEBHC
Here's my thoughts... so let me talk through it. The section that is marked with the red X, if I remember correctly when I did the original V2 design, I believe that logic is just to turn the front panel ION light on/off by decoding the Z80 instruction set and looking for EI/DI. I took this from the original Heath Z80 design but in my opinion we don't need that. We can just leave the ION light on all the time and get rid of those chips. Free up a lot of space. We can then move the ORG0 section down there, optimize the right half of the board (address decoding logic and buffering - I think you already did this with the V4 design, and scoot the RAM and Z80 CPU over to the right side some. Should free up some space on the left side for the H17 logic.

And of course if space is an issue we can put all the logic on an FPGA and just have a giant CPU board with s single chip on it (haha).

Thoughts?

H8-Z80-64-30-F_SilkS.jpg

Douglas Miller

unread,
Sep 8, 2021, 8:55:14 AM9/8/21
to se...@googlegroups.com

I'm not sure about eliminating ION (I don't own any real hardware, so maybe don't get to have an opinion). That has been an important debug tool for problems in the past - the state of ION tells you a lot about situations where the system hangs. Still, we could probably eliminate that circuitry and replace it with a GAL/CPLD chip, if that saves any space.

That being said, I've always been suspicious about the circuitry with Z80/Z180 instruction streams, as I believe there are some possible Z80/Z180 instructions that will result in /M1 and 0xF3/0xFB patterns that are not EI/DI instructions. Still, probably never causing a problem for most users.

--
You received this message because you are subscribed to the Google Groups "SEBHC" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.

Glenn Roberts

unread,
Sep 8, 2021, 8:57:09 AM9/8/21
to se...@googlegroups.com

I had an uncompleted board from a while back that I’d never gotten around to building out. I recently completed that build to help with my VCF demo.  I had no trouble sourcing any of the parts, however Norberto had sent me the two PLCC chips and the oscillator back when I got the board from him.  final board worked first time!  I bought 8G CF cards on Amazon.  I’ve used the Transcend ones simply because they’ve worked (“rock solid”) since I first began playing with the IDE device, which is probably approaching 10 years ago now?  This board, more than anything, brought me back to the hobby as I had previously spent many hours fooling with floppy disks and their associated reliability issues, which has become much worse over time.  It just wasn’t fun

 

I know somewhere in Norberto’s pipeline is the SD equivalent of this board but the CF one is a tried and true warrior!  Terry has done some really cool things with his hardware, as witnessed in his recent posts here…

Glenn Roberts

unread,
Sep 8, 2021, 9:14:37 AM9/8/21
to se...@googlegroups.com

While this would be a fun intellectual exercise, I would argue that there’s minimal need for such a solution.  The Rev. 3 & 4 boards already incorporate RAM, ORG0, and dual serial ports so the only “new” enhancement is the H17, which (IMHO) is of limited value moving forward.  However as a study in contrasts – what would a “modern” version of a 1980-era basic H8 configuration look like, it would be interesting to show – just don’t know if that’s enough to justify the effort.  I agree with Douglas it would be an error to just “hard wire” the ION light – it serves a purpose.

 

If you do implement a modern H17 solution it would be very useful to eliminate the S2350 chip as they’re becoming scarce and expensive.  perhaps that’s something that an FPGA could help with?  Hmm. While you’re at it could an FPGA chip be programmed to replicate either/both of the H17 and H37?  That would be a challenge (and useful as the Western Digital chips for the ’37 are also becoming hard to get)!

 

From: se...@googlegroups.com <se...@googlegroups.com> On Behalf Of Les Bird
Sent: Wednesday, September 08, 2021 8:45 AM
To: SEBHC <se...@googlegroups.com>

Subject: [sebhc] Re: Full size H17 controller with VDIP1 and HSFE support!

 

Here's my thoughts... so let me talk through it. The section that is marked with the red X, if I remember correctly when I did the original V2 design, I believe that logic is just to turn the front panel ION light on/off by decoding the Z80 instruction set and looking for EI/DI. I took this from the original Heath Z80 design but in my opinion we don't need that. We can just leave the ION light on all the time and get rid of those chips. Free up a lot of space. We can then move the ORG0 section down there, optimize the right half of the board (address decoding logic and buffering - I think you already did this with the V4 design, and scoot the RAM and Z80 CPU over to the right side some. Should free up some space on the left side for the H17 logic.

 

And of course if space is an issue we can put all the logic on an FPGA and just have a giant CPU board with s single chip on it (haha).

 

Thoughts?

 

On Wednesday, September 8, 2021 at 8:08:08 AM UTC-4 Les Bird wrote:

Norby, I've been thinking about a new CPU board design with 64K, ORG0, H17 and H8-4 (dual port). Has this been explored already? I think that encompasses the baseline H8 system.

 

Les

 

On Wednesday, September 8, 2021 at 2:27:16 AM UTC-4 Norby wrote:

I decided to take the current H17 design (http://koyado.com/Heathkit/H8-H17-H37-H67-USB_files/H8_H17_Controller.png) that includes the VDIP1 + HSFE and dropped into a full size board. This allows to have the H37/H67 by itself. Also allows to move the VDIP1 from the CPU into the H17 to allow more space between the H8 Control Panel and the CPU board without wasting a backplane slot.

 

Also supports in just having the H17 with HSFE and VDIP1 in the H8 if not interested on the H37/H67 combo. For me the best configuration is the H17 + H67 (174Q, 170Q) as we have a lot of files in H17 H8D format.

Thanks,

Norberto

 

 

--

You received this message because you are subscribed to the Google Groups "SEBHC" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.

image001.jpg

Norberto Collado

unread,
Sep 8, 2021, 11:06:43 AM9/8/21
to se...@googlegroups.com

Hello Les,

 

I agree with Glenn if you could just focus on an H17/H37/H67 FPGA solution for now. I do have the USB BLASTER II and the Quartus software, so I will just need your FPGA H17/H37/H67 solution to start testing. For learning purposes, I will just start with the H67 solution and once it is working, move to the H17, and last on the H37. That will be great if you can do this as I do not know nothing about FPGA’s; just to program them.

 

We need the ION light and that logic can be replaced with a GAL to gain more board space. As well the Single step interrupt circuit.

 

If you want to do a CPU board, I will recommend to take my Z80 V4.0 CPU board and updated it to a Z180 with 1MB of RAM with battery backup as it is still pending on my To-Do-List, but H17/H37/H67 FPGA should be the priority.

 

Thank you for sharing,

 

Norberto

Norberto Collado

unread,
Sep 8, 2021, 11:15:48 AM9/8/21
to se...@googlegroups.com

And if you could add the Z67IDE+ into the FPGA that will be really cool. Driving 2x CF or SD cards. 😊

Glenn Roberts

unread,
Sep 8, 2021, 11:33:00 AM9/8/21
to se...@googlegroups.com

And having a modern (e.g. FPGA) H37/67 interface would be very useful as, unlike the H17, the H37 interface still has good usefulness for either 3.5” floppy drives (which are still serviceable and much more reliable than the old 100K siemens drives) and the Gotek, which provides a software emulation for soft sectored drives.

image001.jpg

Les Bird

unread,
Sep 8, 2021, 12:26:51 PM9/8/21
to SEBHC
Norby, Glenn,

Just to let you know, I do not know anything about FPGAs. I'm very early in the learning process, looking at sample code and trying to connect the dots. But at this stage, anything like an H17, H37, H67 is far out.

However, there are FPGA cores already written for the WD1771 which, despite the regression in chip numbering, is the successor to the WD1791 and is operationally compatible with the 1791, as far as I can tell. So that at least speeds up the possibility of an H37 on FPGA.

The way I see this working is, ultimately, putting the cores for H17, H37 and H67 on FPGA and then interfacing to real hardware via the GPIO pins, so there'll be a need for additional chips to handle the interfacing I think. I'm not sure if the GPIO can directly talk to the floppy drives.

Les

Les Bird

unread,
Sep 8, 2021, 12:36:44 PM9/8/21
to SEBHC
Glenn, yes I kind of figured that "yet another CPU board" is probably not in the interest of the group. Thought I'd throw the idea out to gauge. My interest is in more of the "classic" H8 configuration and not really moving it into the future as Norby's designs are doing.

Perhaps this is a project I pursue on my own then... what I'm discovering is that when building out a classic H8 (64K, ORG0, H8-4, H17, HA-8-3 and H37) it is getting very cramped in the chassis with only a few slots to spare so by combining the H8-4 and H17 on the CPU board it clears up some space. Basically the rear half of the chassis (in this photo) would be freed up which means I can comfortably add a H8-2 Music board and H8-5 (for cassettes) in there.

IMG_6523.JPG

Norby

unread,
Sep 8, 2021, 6:15:28 PM9/8/21
to se...@googlegroups.com
I’m my todo list I’m looking into expanding the H8 bus into a second chassis to have more slots available. 
Or
If you can develop a board with 2-serial ports + H17 + H67 on a full size H8 board that will be great! For me this is the best configuration and to free two more H8 slots.

On the Z67-IDE+ board I did send you include the SCSI controller + IDE controller + 33 MHz oscillator. These are the parts hard to get. I will suggest to get it up and running with latest FW from TerryS. Adding the jukebox capabilities is a great experience.


Sent from my iPhone

On Sep 8, 2021, at 9:36 AM, Les Bird <les...@bellsouth.net> wrote:

 Glenn, yes I kind of figured that "yet another CPU board" is probably not in the interest of the group. Thought I'd throw the idea out to gauge. My interest is in more of the "classic" H8 configuration and not really moving it into the future as Norby's designs are doing.

Perhaps this is a project I pursue on my own then... what I'm discovering is that when building out a classic H8 (64K, ORG0, H8-4, H17, HA-8-3 and H37) it is getting very cramped in the chassis with only a few slots to spare so by combining the H8-4 and H17 on the CPU board it clears up some space. Basically the rear half of the chassis (in this photo) would be freed up which means I can comfortably add a H8-2 Music board and H8-5 (for cassettes) in there.

Reply all
Reply to author
Forward
0 new messages