WDI to IDE adaptor (4)

[Mike Arnold]

Aron, I had a (final) thought relating to testing the board. I have not specified any test points because my prototype is DIP therefore I just hang a probe off a chip leg. I cannot do that in SMD-land. Although I have identified many potential test points, there are only 7 where I think we need a distinct pad. All the others can be accessed by DIP chip legs. If it is possible to put in 7 pads (as below) that should be really helpful.

TP1 U22-5 2s timer output

TP2 U23-1 50ns W60h pulse

TP3 U45-6 50ns R60h pulse

TP4 U28-6 50ns delayed W60h pulse

TP5 U40-8 50ns burst clock trigger

TP6 U27-2 50ns primer clock trigger

TP7 U26-2 22.579MHz clock.

If I get time over my holidays I will continue work on the board test schedule along with the software (EEPROM based).

Thanks for all your support in this.

Mike

[Mike Arnold]

To give an idea of the test ROM software that I will be writing in due course, I have drafted some user instructions. I am open to suggestions and improvements. Note that I have not done any coding for this yet. The idea is to be able to identify faults on the board so that it can be repaired.
Mike

[Mike Arnold]

An update... 

After a bit of a break, Aron and I are back on the case.

I am part way through building a second (hand wired) prototype because the first prototype is now my day-to-day hard disk. Also, I am wiring the second one from the Kicad drawings as another verification of the circuit. When that is done, I will write the diagnostics and Aron will be in a position to finish the layout.

[Mike Arnold]

Well, I have got the second hand-wired prototype booting Cromix so its initial test is promising. This was wired from the Kicad circuits. More testing to come and then to write the test software.

Because I have a few different components in my shed this circuit is slightly different to the Kicad version (eg incorporates +5V to -5V inverter, uses original oscillator design, DIP RAM chip, original circuit around MAX232) but no major deviations really.

Aron, the only error I found was that U103-11 needs to be tied to +5V. Hats off to you for transcribing a big circuit so faithfully.

For those that are interested, it took me 50 man-hours to wire and debug the second prototype. 

Mike

[Aron Hoekstra]

Wow, looks awesome! I'd love to see the underside of that board! I have never had the desire to learn wirewrapping, but seeing how many chips you can squeeze on the top side with this approach, I can see the appeal.

Glad the schematic was mostly accurate! Wow yeah U103-11 wasn't connected to anything at all, not sure how that was missed - that is now updated.

--
You received this message because you are subscribed to the Google Groups "Cromemco" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cromemco+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cromemco/ff504b42-0215-40e3-8f7b-192c327e78d8n%40googlegroups.com.

[Mike Arnold]

I didn't use wire wrap because it becomes difficult in high density layouts and the IC sockets and pins are very expensive. I soldered it all using 0.19mm enamelled wire and plenty of patience! I estimate around 900 solder joints and lots of cups of tea to keep me going. This has taught me that a PCB is definitely the way to go!

I attach a picture of the knitting on the underside as requested. I also include a picture of what happened to prototype 1 in case you wondered.

[Aron Hoekstra]

Oh my god, as expected: this underside gives me anxiety just looking at it! Amazing work though, and I really like how that enclosure turned out.

To view this discussion visit https://groups.google.com/d/msgid/cromemco/ea6d3c2f-b789-40ac-8974-7af02f77283cn%40googlegroups.com.

[Mike Arnold]

A quick update. 

I am still working on the diagnostics ROM for the W2I to make it easier to debug any problems with the PCBs. 

I was delayed by my CDOS development system which started crashing and locking up frequently. In a fit of paranoia I assumed that the W2I attached to the system might be exhibiting a deep bug. However, the answer was simple, the power supply in the chassis was browning out occasionally. So, I had to do a full nut and bolt rebuild of the chassis and power supply (plus adding some PSU monitoring) before I could resume coding. 

I am up to test 12 of 16 but with the distractions of Christmas (and some rather nice French red wines) I may not complete it until the new year.

Wishing you all a very happy Christmas and all the best for Marcus dealing with his health issues.

Mike

[Aron Hoekstra]

Hey Mike, thanks for the update. Happy to hear your system is back up and running at least! No rush on anything, I've got plenty to keep me busy. :) I am working on reproducing the 88-CCC boards which is the controller for the Cyclops digital camera: https://forum.vcfed.org/index.php?threads/cromemco-cyclops-88-ccc-board-set-reproduction.1255558 - Hope to get a camera built and the controller recreated here in the coming months.

-Aron

To view this discussion visit https://groups.google.com/d/msgid/cromemco/f85e820c-e09f-4f68-be40-35b06c53b0ccn%40googlegroups.com.

[Mike Arnold]

Aaron, you are definitely on an adventure with the Cyclops.. Sounds fun All you have to do after that is to recreate the SDI board set!

Incidentally, when do you want some money for the W2I boards?

Mike

[Aron Hoekstra]

Redoing the SDI board set would be awesome - but then I also need all the supporting boards for that high end configuration which I don't own. One day!

I may be mistaken but was I waiting on anything from you either on the schematic or board layout? I think there was some talk of test points? but maybe you already provided me with those..

To view this discussion visit https://groups.google.com/d/msgid/cromemco/c2e9a96f-c690-419f-b9ef-5645efde876en%40googlegroups.com.

[Mike Arnold]

Oops! I forgot that I had some things to do for the PCBs. I will get it sorted.

[Aron Hoekstra]

Thank you!

To view this discussion visit https://groups.google.com/d/msgid/cromemco/0e84274d-bab0-4caa-ae2c-278749cb0accn%40googlegroups.com.

[Mike Stein]

Hi guys,

I'm sorry to say that the years are taking their toll and I've been out of the loop so long that I've forgotten exactly what this board is;

Is it essentially a 7710 emulator when used with a PATA IDE drive?

Does it still require a WDI?
Is either a WDI or a WDI-II OK?
Can it effectively be larger than 11MB?

Sounds like you guys are doing an amazing job!

m

To view this discussion visit https://groups.google.com/d/msgid/cromemco/CAD9iBDLW6AFo_8jiB-KYTtqrvkVzpoCvM%3DGGs0eVcDH8kqyqhg%40mail.gmail.com.

[Mike Stein]

Did you know::

In 2007 the Computer History Museum’s Storage SIG voted the IMI 7710 as a significant product in HDD history for its first usage of a brushless dc motor in a production.

https://www.storagenewsletter.com/2020/04/06/history-1979-imi-7700-first-brushless-dc-spindle-motor-for-hdds/

On Mon, Dec 15, 2025 at 4:44 PM Aron Hoekstra <null...@gmail.com> wrote:

To view this discussion visit https://groups.google.com/d/msgid/cromemco/CAD9iBDLW6AFo_8jiB-KYTtqrvkVzpoCvM%3DGGs0eVcDH8kqyqhg%40mail.gmail.com.

[Mike Arnold]

Hi Mike, as a reminder of what the W2I is, it emulates four 11Mb IMI7710A drives. It is compatible with CDOS, Z80 Cromix, 68000 Cromix and Cromix Plus. It works with the WDI and WDI-II. Having said that, Cromemco have restrictions such as Cromix Plus does not work with the original WDI and only Cromix Plus supports four drives. It has a PATA (ATA-3) interface and can use PATA drives or, with suitable adaptors, SATA or SD cards. My design goal was to make the W2I as versatile as possible.

I have two hand wired versions of the board in my shed and Aron is doing the S100 PCB version (when I get the information to him that I promised and then forgot! Doh!). The S100 version should be quite fun as it can take a 2.5in PATA drive mounted on the card which emulates four IMIs. "A storage farm on an S100 card" (maybe that's a bit of an overstatement). Imagine that idea back in 1981! All it will require (once the PATA drive is set up) is a 34w jumper cable to the WDI (-II) in the card cage.

Mike

[Mike Arnold]

Aron

I have looked at the layout and collected up my final comments. 

1. I have applied thinking to the chip layout based on how I would have gone about a manual layout (impossible with this complexity but perhaps my logic may help). I have attached a picture representing my thoughts. I have positioned the octals in runs so that the data bus can run between them more easily. I have grouped the I/O chips closer to the relevant connectors to help with routing. I have added a reminder about the screw holes for securing the drive (4mm so that there is some wriggle room for the M3 screws). I have also suggested 3mm holes in the two corners for board extraction. Cromemco boards have them and I find them quite useful. As I said, my logic is based on old-school thinking and this may not apply to KiCad and you know how to use it well so take what you want and leave any suggestions that do not fit. You will have to measure up for locations of the connectors and mounting holes as my diagram is only rough.

2. For each test point I suggest a pad with a 1mm PTH for installation of solder posts

TP1 U22-5 2s timer output

TP2 U23-1 50ns W60h pulse

TP3 U45-6 50ns R60h pulse

TP4 U28-6 50ns delayed W60h pulse

TP5 U40-8 50ns burst clock trigger

TP6 U27-2 50ns primer clock trigger

TP7 U26-2 22.579MHz clock.

Plus two or three ground points.

3. Could I suggest that in addition to the +8V, +16V and GND pads that you add +5V. This is so that if you want to use a PATA-SATA adaptor or PATA-SD-card adaptor then you can run some wires out of the board to power it. 

4. For the schematic, the only thing that I found was that U103-11 needs to be tied to +5V. However, I think that you may have already updated that.

If there is anything more you need for me just let me know.

Mike

[Mike Arnold]

One final thought occurred to me. My card cage has no retaining bar for the cards but I recall that Z2 and system 3 do have them. You might need to shuffle the connectors a bit to avoid the bar fouling a connector.