WDI to IDE adaptor

[Mike Arnold]

My WDI to IDE adaptor is complete, tested and working!
This has been a long journey (not helped by me only working on it part time) with periods of frustration and elation along the way.
I attach the user manual which gives you a good deal of information about how it works and the hardware schematics. Tomorrow I will package up the software into a zip file with a link to my drop box.
I am under no illusions that few will build one because most of you are sane enough not to share my love of wiring up chips for hours on end - and there are quite a few chips. However, if you wish to build one, I will be happy to provide support to ensure it works.
Alternatively, the information that I provide could allow someone to build their own design of adaptor using a standard SBC (eg BeagleBoard) and some interface chips.
Mike

[Mike Arnold]

Wrong attachment! Try this.

[Aron Hoekstra]

Wow, congratulations on finishing this! Very cool - I definitely want to try building one.. would you mind if I used the included schematic to design up a PCB in Kicad?

--
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/fb3dc1db-d936-4b1e-a802-fe0436d0b7e4n%40googlegroups.com.

[Richard Muse]

Absolutely fantastic job Mike! Congrats. And stupendous Docs. 

Richard M

--

[jima]

Mike,  thanks so much for this epic piece of work. I've been looking for something to attach to my WDI-II since my HD-20 died. Time to break out the Blitz Bus :-D

[Mike Stein]

Wow! That's determination and a labour of love; congratulations!

Years ago, when my IMI drives were starting to fail and STDCs hard to find I contemplated the same thing but for the WDI-II; it looked fairly simple, but when I dug into the details I realized I'd never finish it, so kudos to you!

m

--

[Mike Stein]

Boy, if you did that I might even try building one; none of these newfangled thumb-nail sized SBCs and FPGA's for me!

And yes, Mike: fantastic documentation as well!!!

To view this discussion visit https://groups.google.com/d/msgid/cromemco/CAD9iBDLOkPXYmhbR%2B_va0uJe94LuMFacnLT%3DmvbibB8hByfU_Q%40mail.gmail.com.

[Peter Higgins]

I've got to try building one as well. I fear the single IMI drive I that I own will, inevitably some day, stop working.

Looking forward to seeing the build details. I cannot tell from the single photo in the manual whether the prototype was soldered point-to-point wiring, or wire-wrap construction.

[Mike Arnold]

Peter

The underside of the board has all the "knitting". It is point to point using self fluxing wire and a wiring pen.

And if you are currently using a real IMI it has the advantage of being dead silent.

Mike

To unsubscribe from this group and stop receiving emails from it, send an email to unsub...@googlegroups.com">cromemco+unsub...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cromemco/a347cd45-bc0d-4015-aa5a-660d5a1bbf14n%40googlegroups.com.

[Mike Stein]

A man after my own heart! I think wiring pens (Road Runners etc.) and the solderable wire are badly underrated; no stripping, less cutting, no special sockets, etc. required!

m

To view this discussion visit https://groups.google.com/d/msgid/cromemco/2aa9f8dc.47d9a.197ad94d597.Webtop.30%40talk21.com.

[Mike Stein]

I gave my only 7710 to Aron but I do still have a few 5 and 20MB 5.25 IMI drives left; they are indeed starting to fail, alas.

m

To view this discussion visit https://groups.google.com/d/msgid/cromemco/a347cd45-bc0d-4015-aa5a-660d5a1bbf14n%40googlegroups.com.

[Mike Arnold]

Aron, I would be delighted if you could look at producing a pcb. I have been left behind in the stone age compared to the modern pcb cad packages. 
Mike

PS did you notice the picture of the IMI7710 was yours?

[Mike Arnold]

The software is in a zip on my shared Dropbox. Link here -> https://www.dropbox.com/scl/fi/p7ko9almsm2imjl36uef9/W2I.zip?rlkey=ge3jow57elk3bcze3ubw50m23&dl=0

Have fun... laughing at my Z80 code!

:-)

Mike

[Aron Hoekstra]

Mike, I just noticed are these schematics hand-drawn? Wow this is some kind of lost art - very nice job. My first step in working with Kicad is always to start with a schematic so I will be doing that part first. I'm definitely going to have questions though.. :) But it's so nice when you do it this way because then designing the PCB is simple because Kicad already knows the nets and valid connections. It even has a pretty decent autorouter which can do like 90% of the trace work for you. Then you run a checker which makes sure everything is connected, and nothing is connected that shouldn't be. At some point I'll also need the Z80 code/HEX but that's a long way off. Would you want to set up a github repo for this?

To view this discussion visit https://groups.google.com/d/msgid/cromemco/14883fba-9f7e-4e44-a43f-77c96887c7b9n%40googlegroups.com.

[Mike Arnold]

Aron

Yes, they're hand drawn so I am afraid you cannot just import a file. If there's any way I can make it easier then let me know.

I did have this idea of doing a pcb that was an S100 card that picked up its power from the bus, had a 2.5in drive on the board and had a short 34w cable between the board and the WDI. I did not take that any further because I only have a 6 slot S100 frame and I've run out of slots.

I have published the software but something odd is going on with the forum. I put the message on the forum with a link to the zip on Dropbox on Thursday but I only got confirmation today. Usually it's immediate.

Anyway let me know if there's any help you need from me but I'm off for a 10 day holiday so may be slow in replying.

Mike

From: Aron Hoekstra
Sent: Jun 28, 2025 at 6:25 PM
To: crom...@googlegroups.com
Subject: Re: WDI to IDE adaptor

To unsubscribe from this group and stop receiving emails from it, send an email to unsub...@googlegroups.com" target="_blank">cromemco+unsub...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/cromemco/14883fba-9f7e-4e44-a43f-77c96887c7b9n%40googlegroups.com.

--
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 unsub...@googlegroups.com">cromemco+unsub...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cromemco/CAD9iBD%2BAv%3D4ZcZXJjq-i7XmvHKVH-OOWWL3hj_Hk%3DqqZ%2B%3DGKjw%40mail.gmail.com.

[Peter Higgins]

Mike - have you tested your adapter using a CF card via a CF-to-IDE adapter?

[Mike Arnold]

Hi Peter

I have not tried a compact flash drive. I am guessing that there are PATA to cf adaptors and, if they obey the older ATA spec and don't take more than 1s to come ready, they should be fine.

Mike

To view this discussion visit https://groups.google.com/d/msgid/cromemco/df8844ab-3f8f-4062-b1bb-6f775cb76269n%40googlegroups.com.

[Mike Stein]

That access issue may be my fault; as co-moderator I received a message about a post being held as suspected spam but couldn't clear it because due to some health issues I've been away from home for a couple of weeks and didn't have my password with me. Sorry if that was the issue, but it sounds like Marcus cleared it for you.

To view this discussion visit https://groups.google.com/d/msgid/cromemco/3f33aed5.4bf7d.197b7b19adf.Webtop.75%40talk21.com.

[Aron Hoekstra]

I like the idea of putting it on a S-100 board - eliminates the need for another power supply and additional wires. If it were all to go on a single board, would that eliminate some of the components for the front panel display or would they all still basically be necessary? I still like the idea of keeping the LED's, maybe just using some right-angle ones on the top edge of the board. The downside is this would restrict you to only 2.5" drives, I guess unless an additional connector was added to power/connect a 3.5" drive.

[Mike Stein]

Just a slightly OT thought: Cromemco supplied a conversion kit to convert HD20 drives from a WDI-II interface to the STDC interface; AFAIK IBM also used the same drive (as an ST506/412), so there might be a few more of those still out there than the IMI version that could replace a failed HD20 by swapping boards in the opposite direction...

Probably not enough to be worth while, but I'll have to see if I have a working ST506 HD20 (5006H/5012H/5018H) and whether putting an IMI board on it would actually work.

If a CF card works then you might as well put an (optional?) IDE<>CF card adapter on the board as well.

To view this discussion visit https://groups.google.com/d/msgid/cromemco/CAD9iBDK%3DgTQizLxHS2cOXm8q9Cu9g0TZN7OV7Z%3Dx8u2rVi0kDg%40mail.gmail.com.

[Aron Hoekstra]

@Mike Arnold

 FYI I've made pretty good progress on transcribing the schematic - I'm on sheet 3 of 5.. Just have a few questions compiling.

Here's one - I'm going to assume that CMDBUS #'s not lining up with the IMIBUS Data line #'s is ok? as shown here CMDBUS6 = D7, CMBBUS7 = D6, and so forth? just wanted to make sure that's correct.

But my biggest concern at the moment is that for some of the 74-series parts, the Kicad symbol pin labels aren't matching up to your schematic. It's probably not a huge deal because I'm assuming the pin numbers are what's most important but I wanted to provide a couple of examples for what I'm talking about and get your thoughts..

Take this part for example, U11 74HCT374. Your labels D7-D0 (pins 13, 8, 14, 7, 17, 4, 18, 3) all line up with the Z80 Data bus D7-D0.

But in Kicad, the pin labels D0-D7 don't line up to the same pin numbers that you have. They have D0=3 which aligns with yours. But then they have D1=4, whereas you have D1=18. What gives? Like I said I'm ignoring the pin labels for now and using your pin numbers.

Thanks,

Aron

To view this discussion visit https://groups.google.com/d/msgid/cromemco/CAHHfo1tpa2DNMiO51xZZQHaLhxzjwykGz8skMb_%3Dy-38a%3DjJdw%40mail.gmail.com.

[Patrick Linstruth]

Not related to your issue, but if you want /W00h to be W00h with a line over it, you can use ~{W00h} for the label. I am not sure if you knew that.

On Jul 4, 2025, at 5:03 PM, Aron Hoekstra <null...@gmail.com> wrote:

@Mike Arnold

 FYI I've made pretty good progress on transcribing the schematic - I'm on sheet 3 of 5.. Just have a few questions compiling.

Here's one - I'm going to assume that CMDBUS #'s not lining up with the IMIBUS Data line #'s is ok? as shown here CMDBUS6 = D7, CMBBUS7 = D6, and so forth? just wanted to make sure that's correct.

<image.png>

But my biggest concern at the moment is that for some of the 74-series parts, the Kicad symbol pin labels aren't matching up to your schematic. It's probably not a huge deal because I'm assuming the pin numbers are what's most important but I wanted to provide a couple of examples for what I'm talking about and get your thoughts..

Take this part for example, U11 74HCT374. Your labels D7-D0 (pins 13, 8, 14, 7, 17, 4, 18, 3) all line up with the Z80 Data bus D7-D0.

<image.png>

But in Kicad, the pin labels D0-D7 don't line up to the same pin numbers that you have. They have D0=3 which aligns with yours. But then they have D1=4, whereas you have D1=18. What gives? Like I said I'm ignoring the pin labels for now and using your pin numbers.

<image.png>

Thanks,

Aron

To view this discussion visit https://groups.google.com/d/msgid/cromemco/CAD9iBDKh0RKOkHR1_8WfD%3D6My-H7yYkHWP1bjLmVLpNDkRB1jQ%40mail.gmail.com.

[Aron Hoekstra]

I did not, thank you! That's so much prettier! :)

To view this discussion visit https://groups.google.com/d/msgid/cromemco/E471355E-BBA0-46C8-834E-246C381A4172%40deltecent.com.

[Mike Arnold]

Aron

Thanks for the quality check - I missed a few points.

You are right about the wiring to the command bus. The pins on the 34 way connector do not flow in the way one expects and I forgot that. I will correct my master schematic.

 

Well spotted for the pin numbers on the octals. I designed some SCCs a few years back and used a different gate/latch order for D7-0 in order to make the pcb laybout easier. It was a tape and pad pcb mask, not CAD based! I have simply copied that order ever since. As you realise, it does not matter in what order the gates/latches are used provided that the inputs correspond to the outputs. If you want to change them to match Kicad then do that. It will probably make things less confusing.

 

Finally, I am impressed at the speed you are getting this into Kicad. If you want to send me an image of the schematics then I am happy to go over them to see if I have made any other slip-ups.

Mike

------ Original Message ------
From: null...@gmail.com
To: crom...@googlegroups.com
Sent: Friday, July 4th 2025, 22:04
Subject: Re: WDI to IDE adaptor
 

@Mike Arnold

 FYI I've made pretty good progress on transcribing the schematic - I'm on sheet 3 of 5.. Just have a few questions compiling.

 

Here's one - I'm going to assume that CMDBUS #'s not lining up with the IMIBUS Data line #'s is ok? as shown here CMDBUS6 = D7, CMBBUS7 = D6, and so forth? just wanted to make sure that's correct.

 

 

 

But my biggest concern at the moment is that for some of the 74-series parts, the Kicad symbol pin labels aren't matching up to your schematic. It's probably not a huge deal because I'm assuming the pin numbers are what's most important but I wanted to provide a couple of examples for what I'm talking about and get your thoughts..

 

Take this part for example, U11 74HCT374. Your labels D7-D0 (pins 13, 8, 14, 7, 17, 4, 18, 3) all line up with the Z80 Data bus D7-D0.

 

But in Kicad, the pin labels D0-D7 don't line up to the same pin numbers that you have. They have D0=3 which aligns with yours. But then they have D1=4, whereas you have D1=18. What gives? Like I said I'm ignoring the pin labels for now and using your pin numbers.

 

 

Thanks,

 

Aron

 

On Sat, Jun 28, 2025 at 3:49 PM Mike Stein <mhs....@gmail.com> wrote:
 

To view this discussion visit https://groups.google.com/d/msgid/cromemco/CAD9iBDK%3DgTQizLxHS2cOXm8q9Cu9g0TZN7OV7Z%3Dx8u2rVi0kDg%40mail.gmail.com.
 

 

-- 
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 unsub...@googlegroups.com" target="_blank">cromemco+unsub...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/cromemco/CAHHfo1tpa2DNMiO51xZZQHaLhxzjwykGz8skMb_%3Dy-38a%3DjJdw%40mail.gmail.com.
 

 

-- 
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 unsub...@googlegroups.com">cromemco+unsub...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/cromemco/CAD9iBDKh0RKOkHR1_8WfD%3D6My-H7yYkHWP1bjLmVLpNDkRB1jQ%40mail.gmail.com.
 

[Christopher Mallery]

If you need another set of eyes I am happy to help also.

To view this discussion visit https://groups.google.com/d/msgid/cromemco/57c81623.612f5.197da43edea.Webtop.155%40talk21.com.

[Aron Hoekstra]

Thanks Chris, I'll post the schematic PDF and Kicad files here when I'm done for review, I'm sure I'll have missed a few things.

To view this discussion visit https://groups.google.com/d/msgid/cromemco/CAPKaxwKDecL08013gbOFcuu9eRLXf0FKgzQo4s%3D7dgzHYdVgyQ%40mail.gmail.com.

[Aron Hoekstra]

Well spotted for the pin numbers on the octals. I designed some SCCs a few years back and used a different gate/latch order for D7-0 in order to make the pcb laybout easier. It was a tape and pad pcb mask, not CAD based! I have simply copied that order ever since. As you realise, it does not matter in what order the gates/latches are used provided that the inputs correspond to the outputs. If you want to change them to match Kicad then do that. It will probably make things less confusing.

Well I'm not very concerned with it being 'correct' as far as Kicad is concerned.. if it works, it works - so I'm going to leave those alone. :) 

[Aron Hoekstra]

Ok well here goes.. I can post the Kicad files but google is blocking the .zip file - let me know if it'd make reviewing these easier. Some of this got a little ugly and I'm sure I messed up some things..  so I could use a hand reviewing from whoever feels like helping - much thanks! I will also continue to review.

Experimented briefly with just getting all the components on a PCB and wow, with the 2.5" drive mounted it's going to be very close whether this can all even fit on an S-100 board..  In the manual you write that all IC's should have a decoupling cap but I'm wondering if I can save some space by eliminating some?  If all the chips can physically fit, this PCB will likely have to be 4 or 6 layers. Will see what I can do but I'd like to get the schematic right before I spend any time on that.

Aron

[Aron Hoekstra]

Oh I forgot to mention, I still have to add a few things like pulling power from the S100 bus and some power regulators..

[Mike Arnold]

General note: I realise that this thread is generating quite a lot of "chatter" which may not be appreciated by forum members. Therefore, after this post, I will communicate directly with Aron by email to help with the schematic and PCB.

Aron, I am still on holiday so cannot do a full check on the schematics but I have noticed that I have forgotten the -WE pins (27) on the 61256 and 28C256. That was down to my haste to get them drawn up. There could be more. If I remember correctly, pin 27 on the 28C256 is tied to +5V whilst pin 27 of the 61256 is connected to -W on the Z80. I will revise the drawings when home.
Great job in transcribing and just a few points that I have noticed.

- U45B is missing the 1K and 1k5 resistors on its input.
- The legend against U13-9 should be DataBus7.
- As you had pointed out earlier, the legends against the IMI bus 34-way connector should be in the order: IMIbus6, IMIbus7, IMIbus4, IMIbus5, IMIbus2, IMIbus3, IMIbus0, IMIbus1.
- On RP1 pins 1 & 15 are NC. The resistor pack contains 14 potential dividers so the internals are a bit different from the way that KiCad shows. Only 12 terminators are needed.
- U38-4 connects to U31-1, U32-1 & U32-19.
- On the 44-pin IDE connector: 41 & 42 are +5V, 43 is ground.

I take your point that the PCB layout will be tricky on an S100 board. Note the the supply decoupling capacitors are 0.1uF rather than the larger 1uF (keep the MAX232 pump capacitors at 1uF). You could consider using SMD chip capacitors for decoupling as they take up very little space and could be mounted on the underside. However, routing could be an issue. I used very small axial devices which fitted above the board but under the IC sockets - you could consider the same. They also have the advantage of fitting between holes that can be towards the top and bottom of chips which eases routing. Likewise, the resistors could be SMD or vertically mounted 1/8W axials.
The regulators need a bit of thought. My W2I with a 2.5in SSD was drawing about 0.9A so a single 7805 might just be enough. If a rotating drive is used it will draw more. I don't have enough with me to do the thermal calculations but putting two 7805s on one of those heatsinks that Cromemco used might crack it. The 7905 for the -5V supply will not need a heatsink.
Finally, I hope that KiCad knows that the 2681 is a 0.4in spaced DIP package and not the more common 0.6in.
Regards, Mike. Again, great job!

[Aron Hoekstra]

Thanks Mike, I was apprehensive about posting this since I knew you are on holiday - so no rush on any of this. I will correct these things you mentioned and post a new version when I have a chance. I'm like 99% sure that I chose the 0.4" footprint for the 2681.. I did catch that during a review of that section. :)

That's a good point about potentially mixing in some SMD components.. Just things like the caps & resistors.. I can have those populated by the manufacturer.. I've done a few runs now with JLCPCB having them stuff components and they've turned out perfect every time. To keep costs down, all SMD components would need to be on just one side of the PCB, so I think the back side would make sense.

I'll hold off working on the PCB until you get back and have some more time, thanks for giving it a quick look though.

On the power side of things, I would want the board to be able to handle a spinning hard drive as well (primarily since I personally have a handful of laptop drives laying around here). Might have to go with a Pololu switching regulator like this one? Goes up to 3.2A https://www.pololu.com/product/3782

--

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/f9ea150b-3b20-4461-9bcc-10074569d629n%40googlegroups.com.

[Aron Hoekstra]

• decoupling capacitors are 0.1uF rather than the larger 1uF Updated
• pin 27 on the 28C256 is tied to +5V Updated
• pin 27 of the 61256 is connected to -W on the Z80 Not sure what pin this is? I'll wait for your updated drawing on that
• U45B is missing the 1K and 1k5 resistors on its input. Updated
• The legend against U13-9 should be DataBus7. Wow, good catch - updated!
• The legends against the IMI bus 34-way connector should be in the order Updated
• On RP1 pins 1 & 15 are NC Updated
• U38-4 connects to U31-1, U32-1 & U32-19.  Not sure how I missed all that! haha - updated
• On the 44-pin IDE connector: 41 & 42 are +5V, 43 is ground. Ah, meant to come back to those pins, then forgot! - updated
• I also updated various "Front Panel" labels so they match up now & create nets

Unless someone complains, I'm not too concerned about the chatter in here.. sometimes I actually enjoy reading through the minutiae of others' projects.

I think Christopher also mentioned he'd like to help. Since Mike already gave it a once over, it wouldn't hurt to have another set of eyes review. Maybe if someone else can give it another look, we'll be in a good spot for when you return. (updated schematic attached)

Aron

--

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/f9ea150b-3b20-4461-9bcc-10074569d629n%40googlegroups.com.

[Peter Higgins]

Amazing work Aron! I am definitely up for building a couple of these boards if you do a production run.

A few comments:

1. Could you add labelling for U3, the 61256 RAM chip

Pin 27 = /WE (should connect to /W on the Z80)

Pin 22 = /OE

Pin 20 = /CE

2. Could you add labelling for the currently unlabelled pins on U4, the 2681 UART.

3. I think a Pololu D36V28F5 (3.2A max) or maybe even a D24V22F5 (2.5A max) would be an excellent choice, rather than using standard 7805 regulators. I use these exclusively on all new S100 board designs, and they work great - no need for a heat sink, which will save some board real estate. These regulators should provide adequate 5V power to both the ICs and a 2.5" IDE SSD.

[Aron Hoekstra]

Peter, thanks for the feedback - will get these labels added.  Yep I think the D24V22F5 would be the way to go, I think 2.5A should be plenty.. Maybe I just need to add a larger cap for a startup surge, or do you think the caps on the pololu would eliminate that need?

Aron

To view this discussion visit https://groups.google.com/d/msgid/cromemco/38f04bd8-9457-4de8-9d55-7f7b00f40e3an%40googlegroups.com.

[Peter Higgins]

I think it is still good practice to add a capacitor directly at the input and output of these switching regulators, something like:

- input side: 33uF, 25V aluminum organic polymer (eg Kemet A758EK336M1EEAE040 )

- output side: 10uF, 25V aluminum organic polymer (eg Kemet A758BG106M1EAAE070 )

[Peter Higgins]

Another thought - the circuitry for the 3.686MHz and 27MHz oscillators could each be replaced by 8-pin DIP package oscillators. I could not easily find an 8-pin DIP package 22.5792MHz oscillator, though this frequency is readily available as an SMD component.

[Aron Hoekstra]

Oh I see what you mean about needing labelling for those parts, U3 & U4. I couldn't find a premade Kicad symbol for the 2681 or the 61256 chips, so I just picked a generic 24 & 28 pin connector.

I've also added the S100 bus header, Pololu regulator (and related caps as suggested) and the 7905.

In the process, I also discovered I completely missed the negative supply to U36 & U37! Yikes.

There are still some unmapped pins on U36, 3 & 10?

Should I add some caps to the 7905?

Update schematic attached.

Aron

On Tue, Jul 8, 2025 at 7:22 PM Peter Higgins <higgin...@gmail.com> wrote:

To view this discussion visit https://groups.google.com/d/msgid/cromemco/38f04bd8-9457-4de8-9d55-7f7b00f40e3an%40googlegroups.com.

[Peter Higgins]

U36 (75107 line receiver) has an internal "no connect" on pins 3 and 10. On your schematic, you could mark these connections with an "x" just like pins 1, 2, and 4.

You can add the same input/output caps for the 7905 as for the Pololu 5V regulator.

[Mike Stein]

"a spinning hard drive as well" ?

As well as what? SSD drives and CF cards?

To view this discussion visit https://groups.google.com/d/msgid/cromemco/CAD9iBDLO1SVXFAhVyJCFcYM2N33qGKehEnd%2BxPFV3oLbEL4d8A%40mail.gmail.com.

[Aron Hoekstra]

Yep, exactly (or well in our case you'd be using a 2.5" IDE to SD/CF adapter such as https://www.amazon.com/gp/product/B08SK4G7RP)

The difference here being as I'm sure you can imagine, the spinning disks use far more power than the solid state devices. So I'd like to make sure this board can put out enough power to support them as well as solid state solutions.

To view this discussion visit https://groups.google.com/d/msgid/cromemco/CAHHfo1uDMvFyG8jTg5to9K5yH2S%2BJ5DG%2BnqrNQJP%2BN3Du1zTbA%40mail.gmail.com.

[Peter Higgins]

Until someone actually tests connecting a mechanical 2.5" IDE drive to the board's power supply, I'd call this an "unsupported feature". Though small, these drives can draw almost an amp of power during spinup and half an amp when seeking. My concern is the effect this heavy power draw on the board's logic circuitry - particularly if it is sudden and intermittent as during seek operations - if the same 5V regulator provides power to both.

[Aron Hoekstra]

I was just starting to look at some part selection and am realizing that JLCPCB carries SMD versions of the 74LS/74HCT chips. I know I don't have all these chips just lying around, and sourcing them all from eBay, etc gets really expensive. Just curious since we're already thinking about using some SMD components, would anyone be opposed to using SMD 74x parts and having them stuffed during the mfg process? Prices are really reasonable, from .25ea to .75ea depending on the part. I realize that takes away some of the vintage authenticity/appearance of the finished board.. Parts like the Z80, 61256, etc would still all be through-hole. Thoughts? 

To view this discussion visit https://groups.google.com/d/msgid/cromemco/c140787f-f877-4c67-8b00-0c0d61fa397en%40googlegroups.com.

[Peter Higgins]

I am not opposed to having the board partially populated by SMD parts. "Vintage authenticity" is not my primary goal for a WDI replacement.

[Peter Higgins]

Incidentally, if you are going the route of replacing some of the DIPs with SMDs you could also consider replacing the obsolete 61256 by the current production 62256, which is in stock as an SMD at JLCPCB.

On the other hand if you want to stay old school with DIPS, all 74LS and 74HCT parts are still current production, and are in stock at Mouser. The older 74-series and 74F parts will need to be sourced at secondary sources like Unicorn or Anchor Electronics (or eBay).

On Wednesday, July 9, 2025 at 7:57:57 PM UTC-7 Aron Hoekstra wrote:

[jima]

Actually, Mike, I'm very much enjoying the chatter so unless it is onerous to keep it going here please feel free to continue.

James

[arnol...@talk21.com]

Thanks, but if others have a strong adverse feeling then please let us know.

To view this discussion visit https://groups.google.com/d/msgid/cromemco/2964761c-c031-440d-8835-95cd05eec6bcn%40googlegroups.com.

[Mike Stein]

Sorry to interrupt this interesting thread but I've been away for a while and have missed a few things, especially whether this project has switched from a standalone converter between the WDI  and a PATA drive to an S100 card emulating both the WDI and the IDE HD? And does it now emulate the later IMI drives that used the WDI-II ?

Great work; wonderful to see folks working together like this.

To view this discussion visit https://groups.google.com/d/msgid/cromemco/CAD9iBDLSssaDFCqnEaji99GAcUQ%2BhYRBPB8BsawOP-jz-KK63g%40mail.gmail.com.

[Mike Arnold]

Hi Mike,

The WDI to IDE adaptor has not changed in any big way. It still will have a 34 pin connector to take a cable to the WDI or WDI-II. It will also have an IDE connector for the drive (various drives can be used). Aron is doing his magic and trying to fit in both the old 40-pin connector and the more recent 44-pin connector. He is laying out a PCB that, coincidentally, is an S100 card but only so that it can pick up its power from the bus. It will not interact with the S100 bus itself.

The adaptor will only emulate the IMI7710A. It does not emulate the 5007 or 5018.
Mike

[Mike Stein]

Ah, thanks for clearing that up for me; nice to hear that it works with both a WDI and a WDI-II.

I'm oldskool like you in this instance, but these days I get discouraged when I think that some wizard could put almost all of that into a single chip the size of a postage stamp (remember those? ;-).

To view this discussion visit https://groups.google.com/d/msgid/cromemco/5126dee2-4fe3-44b3-bf01-567a997cc8a5n%40googlegroups.com.