EDIT: Project finished.
To order, fill the spreadsheet here: https://docs.google.com/spreadsheets/d/1GleRhKIo71r2kXYFSYJaaIWSvXhHe1CcnDU0wpnyp_w/edit?usp=sharing (https://docs.google.com/spreadsheets/d/1GleRhKIo71r2kXYFSYJaaIWSvXhHe1CcnDU0wpnyp_w/edit?usp=sharing)
I will contact you via PM. Shipping may vary.
Original post:
Hi,
I'm planning to make myself a DDI-1 interface by following the schematics in the service manual. After taking a look at the schematic I have some questions:
1. Is the standard 3.5" Floppy compatible with DDI-1 without any modifications? *Other than the different pinout which I will adjust.
2. Does the Interface support 2 drives? I see that there is a Drive select 0 & 1, but only one Motor On signal. Is the Motor On signal common for both?
3. Why is there a capacitor, C115, in series with the +5V Rail? As far as I can recall, DC is blocked by capacitors...
4. Can I use a 32K EPROM instead of the 16K one and use A14 to switch between AMSDOS and PARADOS? Are they interchangeable?
5. What are LK1, LK2, LK3 jumpers for?
6. The 74LS240 seem to be tri-state buffers rather than inverters, I presume that the 1G and 2G inputs are grounded to be used as a simple inverter.
7. Can I use 2N2222 transistors instead of KTC1815? Seems that the spec is even better than needed.
Thank you very much!
P.S.: Yes, I know there are some clones and DDI3, but I prefer to do it myself because I already have the required components and I need only the PCB.
Hi Silvergreen93,
some answers:
1 - No, a PC 3.5in drive doesn't output a "Ready" signal, so the drive needs to be modded to do this. Also, PC drives are hardwired as DS1, meaning that they are drive B:, this also needs to be modified.
2 - Yes, it supports 2 drives and yes only one Motor-on signal, common for both.
3 - This is the tricky part and you will have to decide how you get around this one. You are correct, capacitors block DC. The DDI-1 actually gets its power from the FD-1 Floppy drive, not from the CPC. You could replace this capacitor with a bridge so that the CPC supplies the power, but Amstrad obviously had problems with the reset sequence and this was their solution.
4 - Yes, there's instructions on the Wiki explaining how to do this. They are interchangeable as long as you don't leave A14 floating.
5 - LK-1 sets the /EXP signal, possibly so that the 464 could detect that a DDI-1 is connected. However, I don't think any software or the firmware actually use this signal. LK-2 and LK-3 are to adjust the pre-compensation level of the data seperator. However I think that they are always set with both links present.
6 - Correct.
7 - Any general purpose NPN transistor could be used here. However, take note that the KTC1815 has a different pin order (BCE) than most modern transistors such as the 2N2222 (CBE), so you'll have to route the PCB slightly differently.
Bryce.
Silvergreen93,
Where did you get the NEC765 disc controller IC or are you using a FPGA (or similar) to emulate the 765?
rpalmer
Quote from: rpalmer on 12:34, 19 September 17
Silvergreen93,
Where did you get the NEC765 disc controller IC or are you using a FPGA (or similar) to emulate the 765?
rpalmer
Well, I found about 5pcs left at my local electroincs shop at $1.50 a piece :D And they were genuine NEC parts.
But I did search on ebay and found http://www.ebay.com/itm/NEC-UPD765AC-2-D765AC-2-Floppy-Disk-Controller-IC-DIP40-x-1pc-/162094807385?epid=607929550&hash=item25bd9a7959:g:tqIAAOSwuxFYwQ9a (http://www.ebay.com/itm/NEC-UPD765AC-2-D765AC-2-Floppy-Disk-Controller-IC-DIP40-x-1pc-/162094807385?epid=607929550&hash=item25bd9a7959:g:tqIAAOSwuxFYwQ9a)
I wanted to use a FPGA to emulate the whole DDI-1 interface, this seems doable, but I do not have much info of the exact operation of the 765 and the 9229
Quote from: Bryce on 11:49, 19 September 17
1 - No, a PC 3.5in drive doesn't output a "Ready" signal, so the drive needs to be modded to do this.
3 - This is the tricky part and you will have to decide how you get around this one. The DDI-1 actually gets its power from the FD-1 Floppy drive, not from the CPC.
Thank you for complete answers!
1 - How can the FDD be modeed to support the Ready signal? I remember I had a 5.25" drive with a jumper to select Disk Change or Ready signal, but I don't recall having a 3.5" with this option.
Maybe I can include the mod in the PCB to make it compatible with any FDD.
3 - Is there any problem if I do power the DDI and the FDD from the expansion port of the Amstrad? As the Amstrad doesn't have any local regulators, if the external 5V PSU is good enough, it technically should work fine.
1 - The mod needs to be done on the drive itself. The only external mod would leave the Ready signal permanently enabled which messes up some software that needs to know you changed disks. There are many threads and articles here and around the web to show which drive makes/models are convertable. some are simple jumper moves, some require PCB changes.
2 - Oh, there's no 2, on to number 3.
3 - I have my DDI-1 powered from the CPC. It works most of the time, but some cold starts result in ROM 7 (The AMSDOS in the DDI-1) not being detected.
Bryce.
Quote from: SilverGreen93 on 11:05, 19 September 17
5. What are LK1, LK2, LK3 jumpers for?
If LK1 is not set then the ROM appears as ROM 0, which causes the ROM software to auto-boot CPM. (There is a pull-up resistor within the CPC which pulls the inputs to the XOR gate high.
Quote from: Bryce on 11:49, 19 September 171 - No, a PC 3.5in drive doesn't output a "Ready" signal, so the drive needs to be modded to do this. Also, PC drives are hardwired as DS1, meaning that they are drive B:, this also needs to be modified.
While bryce is correct that PC drives are defaulted to DS1, swapping the respective wires in the cable fixes that (as can be seen in PC floppy cables). There are older drives which could select DS0/DS1 and these are rare today (I have a TEAC FD-235F which select DS0/D1 as the default setting).
Bryce is also correct about the ready signal for many of the recent floppy drives, however there are older drives which can select to have "Ready" signal or a "Disk Changed" signal (such as the TEAC FD-235F I mentioned).
Some people have found success with a "Ready" Signal by connecting it to ground, but some software does not like this and will complain bitterly. I think that maybe connecting "Ready" to the respective drive select signal or "Motor On" signal can be a good choice as both are active low to match the active low of "Ready".
@IanS (http://www.cpcwiki.eu/forum/index.php?action=profile;u=629): Hm... So LK1 is a quick way to boot CPM from disk? I will try that.
@rpalmer (http://www.cpcwiki.eu/forum/index.php?action=profile;u=379): I have a lot of 10-12 FDDs laying around from old PCs and I know for certain that I have 3 TEAC FD-235HF with an "H", hope it's the same thing. If nothing else works I'll try a jumper to GND.
After some unfortunate delays, I finally managed to re-write the DDI-1 schematic using Proteus Design Suite.
I only modified the floppy interface to 34 pin standard and the double-size EPROM for dual ROM.
I attached a picture of the schematic if anyone likes to take a look and maybe find mistakes, if any.
I will now start to design the PCB and then order some in China. If anyone will be interested, I will sell the rest and/or maybe make some DIY kits, as I only need 1 piece.
Updates will be posted.
P.S.: If anyone does want to request the original Proteus file, I will post it here after the PCB design is done.
Looks good, assuming you've connected all the TTL power pins and you've got your footprints correct.
Bryce.
The first version of the PCB is almost finished. I need to check that all components fit without interfering with the CPC or connectors and also to tweak some various traces or small changes.
If anyone has any other design suggestion in this phase before sending to production it is really welcome.
The board is aprox 9.7x9.9cm and is designed to be mounted vertically, the floppy connectors being on the back.
(The red connector in the 3D is the card-edge one)
It's very tall and will block the screen for many users. You should spread the design horizontally like the original DDI so that it's not blocking your view.
Bryce.
Quote from: Bryce on 11:13, 14 November 17
It's very tall and will block the screen for many users. You should spread the design horizontally like the original DDI so that it's not blocking your view.
Thank you for feedback Bryce!
I haven't thought of that, in my case I have an LCD monitor that is 40 cm away and about 15cm up from the table, so it doesn't block my view. I cannot see why this would be a problem for many users, the monitor/TV shouldn't be so close to the CPC I think.
Anyway, I'll try to make another design with the ROM underneath the edge connector to lower the overall position, because extending the design over 10cm will cost a lot more to produce.
Many users still use an original CTM Monitor which has 8cm power cables to/from the CPC, so putting the monitor higher/further away isn't an option.
Bryce.
Quote from: Bryce on 11:35, 14 November 17
Many users still use an original CTM Monitor which has 8cm power cables to/from the CPC, so putting the monitor higher/further away isn't an option.
I see. I didn't know that, I haven't had the original monitor, but it sounds like the CRT is dangerously close to the eyes. :D
Quote from: SilverGreen93 on 11:38, 14 November 17
I see. I didn't know that, I haven't had the original monitor, but it sounds like the CRT is dangerously close to the eyes. :D
Depends how long your arms are :D
Bryce.
Quote from: SilverGreen93 on 11:20, 14 November 17
Anyway, I'll try to make another design with the ROM underneath the edge connector to lower the overall position, because extending the design over 10cm will cost a lot more to produce.
Another design you could try is move the board so that it fits next to the drive instead of next to the computer. This would require an additional ribbon cable, but then if you can fit the board e.g. under the drive, it could allow for a pretty enclosure design.
A couple of suggestions:
- It could be interesting to add a +5V line out on the board to power the drive. The original power supply of the CTM isn't enough, but if you are using an external power supply, it should be enough. No need to add yet another PSU just for the drive.
- Or the other way around, and you power the board with the PSU you planned to use with the drive. Like Amstrad did. You need an additional PSU if your drive needs +12V anyway (modern drives don't). Which would go well with my first suggestion of fitting the board next to the drive.
- You may want to add a jumper / switch to short pins 31-32 to access the other side of the drive.
Can the DDI-I be used to connect two drives directly?
The board is designed to power the floppy without any additional power supply. See the 4 pin connector beneath the FDC connector.
I cannot move the board away from the Amstrad as I don't have any ribbon cable edge connectors to make a cable for it.
Quote from: khaz on 13:52, 14 November 17- You may want to add a jumper / switch to short pins 31-32 to access the other side of the drive.
The disks are not double sided on the original DDI-1? Do I need to add a jumper to switch sides?
I see that an output signal /SIDE1 exists and is driven directly by the uPD765...
I asked in the first page if the interface will be able to use both drives and Bryce said that it is.
Moving the PCB to the other end of the cable (at the drive end) won't work. The expansion port signals aren't buffered and are too weak to drive a signal down a meter of cable. They are also much more likely to damage the CPC through static discharge - Long plastic cable connected directly to CPU pins = bad idea. The DDI-1 has buffered outputs to allow the cable to the drive to be that long.
The CTM PSU does have enough power to drive the DDI-1 (as long as you don't have 5 or 6 other expansions connected), the seperate 5V supply in the FD-1 was done for other reasons. But your idea about powering it from the drive supply is still a good idea as it takes the load off the CPC. Just make sure that they are isolated from each other, otherwise you'll have issues that could involve flames. :)
Bryce.
Quote from: SilverGreen93 on 13:59, 14 November 17
The disks are not double sided on the original DDI-1? Do I need to add a jumper to switch sides?
I see that an output signal /SIDE1 exists and is driven directly by the uPD765...[/font]
Parados, and other custom DOS can access both sides directly, but only for floppies formatted for it (like ROMDOS format or games with a custom DOS like Orion Prime). If you try to put two standard DATA dsk images on your 3.5, you will only be able to access one.
Quote from: Bryce on 14:04, 14 November 17
Moving the PCB to the other end of the cable (at the drive end) won't work. The expansion port signals aren't buffered and are too weak to drive a signal down a meter of cable. They are also much more likely to damage the CPC through static discharge - Long plastic cable connected directly to CPU pins = bad idea. The DDI-1 has buffered outputs to allow the cable to the drive to be that long.
A meter haha, I wasn't planning to put the drive on the other side of the room. Even something like 30cm is too long for that?
I had no idea about that sort of challenge.
Quote from: khaz on 14:11, 14 November 17
Parados, and other custom DOS can access both sides directly, but only for floppies formatted for it (like ROMDOS format or games with a custom DOS like Orion Prime). If you try to put two standard DATA dsk images on your 3.5, you will only be able to access one.
Ok, I will add a jumper to ground if you say that it is required for other types of floppies. Does the standard BASIC floppy type require this to use both sides?
@Bryce (http://www.cpcwiki.eu/forum/index.php?action=profile;u=225), is this modification ok for all floppy types?
A 3 pin jumper like this:
/SIDE1 >--------*
*--------------> to pin 32
GND >-----------*
Even 30cm could be an issue without buffers / line-drivers.
The side select is fine for 3.5in drives, but it would need to be switched if a 3in drive is connected.
Edit: It's not a simple "jumper to ground". Here's how it has to be connected: http://www.cpcwiki.eu/index.php/Side_Select_Switch
Bryce.
Quote from: Bryce on 14:21, 14 November 17
Edit: It's not a simple "jumper to ground". Here's how it has to be connected: http://www.cpcwiki.eu/index.php/Side_Select_Switch (http://www.cpcwiki.eu/index.php/Side_Select_Switch)
Well, isn't that what I suggested above? a switch between GND and /SIDE1 connected to pin 32? I don't quite get the picture from the link, what has it to do with pin 1 and pin 34?
Yes, I didn't read you ASCII art properly. I though you were suggesting grounding pin 32 without cutting it.
Bryce.
Quote from: Bryce on 09:01, 15 November 17
Yes, I didn't read you ASCII art properly. I though you were suggesting grounding pin 32 without cutting it.
Well then, I'll try to respin the board in a couple of days, including the side switch/jumper and the lower profile, and come back.
Quote from: Bryce on 14:21, 14 November 17
Edit: It's not a simple "jumper to ground". Here's how it has to be connected: http://www.cpcwiki.eu/index.php/Side_Select_Switch
What's wrong with simply grounding the connection? This is how most cables are made, and it works. See for example http://www.cpcwiki.eu/index.php/Guide_on_how_to_connect_a_3.5%22_drive_to_a_CPC6128/664
Because you are essentially connecting the output of a 74LS38 in the floppy controller to ground. This output can supply a maximum current of 250µA. So after a while you will have destroyed at least this output channel of the 74LS38 if not the entire IC. Which is why I recommend people to avoid these types of cable (although they provide me with a steady supply of repair work :) ).
Bryce.
New design ready! (v2)
Added SIDE selection jumper as discussed above.
Reduced the overall dimensions to (LxH): 97x93mm
Positioned the ROM beneath the expansion connector.
Are you sure that there is enough space under the edge connector for that ROM without the PCB hitting the table?
Bryce.
Quote from: Bryce on 09:19, 17 November 17
Are you sure that there is enough space under the edge connector for that ROM without the PCB hitting the table?
Yes, I double checked on my Schneider model. Please measure yourself if there is any Amstrad that needs an DDI-1 and is different.
I left exactly 1 inch from the center of the edge connector to the lowest point on the DDI-1 PCB. The table is at about 1.20 inch from the connector center. So it should not touch the table.
Also, I checked if the ROM+Socket would not hit the Amstrad itself, and as far as I can tell, it it just perfectly flush. I will triple check it again before sending to production.
Only the Classic 464's really need a DDI, so if it fits on your Schneider it will fit everything it needs to.
Bryce.
That's great then!
One question about the Edge connector: I am planning to make them from ISA motherboard connectors. Is there any easy way I can get the "positioning slot" between two contacts to make it align perfectly with the card edge? Or should I just use some end-caps near the last pins to stop it inserting in the wrong position?
The PCB mounted edge connectors are readily available. Use real ones instead of faffing about with a hacksaw on ISA connectors.
Bryce.
Hmm... I did a search some time ago on ebay, farnell and tme, but haven't found any exact 50 pin ISA-type connectors. If you know some, please share :)
Quote from: SilverGreen93 on 10:25, 17 November 17
Hmm... I did a search some time ago on ebay, farnell and tme, but haven't found any exact 50 pin ISA-type connectors. If you know some, please share :)
https://www.ebay.com/itm/5pc-Industrial-Card-Edge-Slot-Socket-Connector-25x2P-50P-2-54mm-0-1-3A-240-50/
The part No. is AMP 5530841-5
https://www.mouser.de/ProductDetail/TE-Connectivity/5530841-5/?qs=%2fha2pyFaduh67bfhXJdtNcb6skNETH7L%252bUNDoW3P56RQc3ATtodEKg%3d%3d
https://www.digikey.de/product-detail/en/te-connectivity-amp-connectors/5530841-5/5530841-5-ND/2259225 (https://www.digikey.de/product-detail/en/te-connectivity-amp-connectors/5530841-5/5530841-5-ND/2259225)
Bryce.
Quote from: Duke on 10:28, 17 November 17
https://www.ebay.com/itm/5pc-Industrial-Card-Edge-Slot-Socket-Connector-25x2P-50P-2-54mm-0-1-3A-240-50/ (https://www.ebay.com/itm/5pc-Industrial-Card-Edge-Slot-Socket-Connector-25x2P-50P-2-54mm-0-1-3A-240-50/)
Oh, great! I will order this one.
Quote from: Bryce on 10:30, 17 November 17The part No. is AMP 5530841-5
https://www.mouser.de/ProductDetail/TE-Connectivity/5530841-5/?qs=%2fha2pyFaduh67bfhXJdtNcb6skNETH7L%252bUNDoW3P56RQc3ATtodEKg%3d%3d (https://www.mouser.de/ProductDetail/TE-Connectivity/5530841-5/?qs=%2fha2pyFaduh67bfhXJdtNcb6skNETH7L%252bUNDoW3P56RQc3ATtodEKg%3d%3d)
https://www.digikey.de/product-detail/en/te-connectivity-amp-connectors/5530841-5/5530841-5-ND/2259225 (https://www.digikey.de/product-detail/en/te-connectivity-amp-connectors/5530841-5/5530841-5-ND/2259225)
Unfortunately these are too expensive and the shipping price is huge.
Mouser shipping is cheap within Germany, not sure what they charge outside though. I rarely order single parts though, so the shipping then becomes irrelevant.
Bryce.
Looks like the shipping for both mouser and digikey is 20EUR.
That's why I order all my parts from tme or farnell, which for the majority of parts have free shipping in Romania.
Quote from: SilverGreen93 on 11:04, 17 November 17
Looks like the shipping for both mouser and digikey is 20EUR.
That's why I order all my parts from tme or farnell, which for the majority of parts have free shipping in Romania.
If you buy for 50 euro at mouser you get free shipping. Arrow is also pretty good, iirc. free shipping above 50 us once you have bought some items there before (perks customer or something).
Nice work, getting hold of DDI-1s can be expensive these days.Next - design a custom Gotek board to plug directly into the floppy port on your expansion :-)
Follow the design with the OLED screen :-)
My orders are usually around the €200+. I only order once a month at the most.
Bryce.
Quote from: Duke on 11:13, 17 November 17
If you buy for 50 euro at mouser you get free shipping. Arrow is also pretty good, iirc. free shipping above 50 us once you have bought some items there before (perks customer or something).
Yeah, I know, but I won't buy that many parts. I will do a BOM at the end and see what is the cost for 1 pc. and then decide if and where from will I order the components.
I already have most of the components available in my parts jelly bin to make 1 fully populated board (including cutting the ISA connector), so the only part that I ordered was the FDC data separator.
Quote from: Sykobee (Briggsy) on 11:19, 17 November 17[/size]Nice work, getting hold of DDI-1s can be expensive these days.Next - design a custom Gotek board to plug directly into the floppy port on your expansion :-)Follow the design with the OLED screen :-)
That's why I'm doing it. I think you can hook up a Gotek to this interface, as it is 100% compatible with the original one. I don't know what custom board are you expecting, maybe some integration with a USB directly connected to the interface, but I see that others have done that already.
The next logical step to follow is to make a FPGA design to replace the D765 and the FDC data separator. If I manage to do that, the next batch/version of DDI-1 will be a lot easier to produce, more reliable and way smaller.
I think I have a load of data seperators somewhere if you need more.
Bryce.
Also forgot Arrow has edge connector too, I bought some of those once, they are fine for CPC too:
https://www.arrow.com/en/products/5530843-5/te-connectivity
It's fairly cheap and you get 20% off.. Perks shipping is free on US 20.
Ordered the PCBs on Sunday, 19/11/2017, received them today via DHL. The manufacturer that I chose is ALLPCB.
I am very pleased with the quality of the boards and especially with the fast processing.
Here are some pictures.
I will start the first assembly hopefully this weekend, and then do some testing.
You gonna be selling these? Having recently got a cpc 464 again and wanting to avoid another zaxon device.....
Sent from my ONEPLUS 3t using Tapatalk
Quote from: CraigsBar on 01:00, 24 November 17
You gonna be selling these? Having recently got a cpc 464 again and wanting to avoid another zaxon device.....
Just for curiosity, what's wrong with zaxon's devices?
I will be making a full list of Bill-of-Materials after the successful prototype testing and, after such demands, I plan to offer them in 3 variants: board only, board + parts for DIY, or already assembled & tested.
Please let me know which of them are you interested in and I will be calculating the final cost afterwards.
Quote from: SilverGreen93 on 08:22, 24 November 17
Just for curiosity, what's wrong with zaxon's devices?
I will be making a full list of Bill-of-Materials after the successful prototype testing and, after such demands, I plan to offer them in 3 variants: board only, board + parts for DIY, or already assembled & tested.
Please let me know which of them are you interested in and I will be calculating the final cost afterwards.
Board and parts sounds like fun.
Sent from my ONEPLUS 3t using Tapatalk
Great job.
I will be interested on one diy kit if there haven't got SMD components. If has, a mounted one will be my option.
Waiting the results.
There are no SMD components in the design. The main chips will be socketed.
Thanks SilverGreen93.
One board + parts for DIY for me, please.
I finally assembled the first board.
The ROM works perfectly in both PARADOS and AMSDOS, but I can't seem to get the floppy to work properly.
I always get the "Drive A: is missing" error when accessing any floppy with the CAT / DIR / |A command
Do I need to format the disk in CP/M before use in standard BASIC? Can I format it without using CP/M?
If not, it means that I have to find a PC with a floppy controller to write a CP/M disk first and then test if CP/M boots first.
P.S.: Yes, I have a floppy with ready signal.
You can format with Parados.
|drive
http://www.cpcwiki.eu/imgs/0/0d/Parados.pdf
Quote from: khaz on 02:30, 26 November 17
You can format with Parados.
|drive
http://www.cpcwiki.eu/imgs/0/0d/Parados.pdf (http://www.cpcwiki.eu/imgs/0/0d/Parados.pdf)
Thanks a lot! That seems to have done the trick.
So, initial testing looks very promising :D
Edit: In the second picture, why is ParaDOS showing 4 times on the start screen?
I tested CP/M also, but I have one problem. It seems that after loading any CP/M program, when returning to CP/M, the error "read fail" occurs and I can never recover.
If I shut off the Amstrad and turn back on, then I can again enter CP/M using the |CPM command. If I return to AMSDOS it works, but if I want to go back to CP/M, it doesn't load, again with the error "read fail".
Something weird is going on. The disk reads fine the first time, but the second time it tried to access it, it gives the read fail error.
Any clues? I don't know if it's something wrong with the components I used or the floppy drive itself is partially bad.
Also, If I CAT the CP/M system disk it works, if I change the disk to standart BASIC one, CAT still works, but if I return to CAT the CP/M disk it again says read fail. This is very weird.
Later edit: If I disconnect the power cable from the floppy and connect it back when pressing Retry it works fine, it reads it again.
Quote from: SilverGreen93 on 10:43, 26 November 17
Thanks a lot! That seems to have done the trick.
So, initial testing looks very promising :D
Edit: In the second picture, why is ParaDOS showing 4 times on the start screen?
How are you decoding the ROM index which selects ParaDOS? Are you comparing against all 8 bits?
Quote from: SilverGreen93 on 17:17, 26 November 17
I tested CP/M also, but I have one problem. It seems that after loading any CP/M program, when returning to CP/M, the error "read fail" occurs and I can never recover.
If I shut off the Amstrad and turn back on, then I can again enter CP/M using the |CPM command. If I return to AMSDOS it works, but if I want to go back to CP/M, it doesn't load, again with the error "read fail".
Something weird is going on. The disk reads fine the first time, but the second time it tried to access it, it gives the read fail error.
Any clues? I don't know if it's something wrong with the components I used or the floppy drive itself is partially bad.
Also, If I CAT the CP/M system disk it works, if I change the disk to standart BASIC one, CAT still works, but if I return to CAT the CP/M disk it again says read fail. This is very weird.
Later edit: If I disconnect the power cable from the floppy and connect it back when pressing Retry it works fine, it reads it again.
I think it'll be doing seeks and recalibrates and read ids to determine the disc format.
Then when it determines the disc format, it then does a read data.
Does that help?
Quote from: arnoldemu on 17:54, 26 November 17How are you decoding the ROM index which selects ParaDOS? Are you comparing against all 8 bits?
You can see the schematic in the first pages of this topic. The ParaDOS is mounted in the same place as the AMSDOS.
Quote from: arnoldemu on 17:55, 26 November 17
I think it'll be doing seeks and recalibrates and read ids to determine the disc format.
Then when it determines the disc format, it then does a read data.
Does that help?
No.
But I managed to find another floppy disk drive and with this one it does work. So my conclusion is that the floppy drive was having some problems, but I don't know why.
The first floppy was a TEAC FD-235HF 3823-U5 and now I have replaced it with a NEC FD1231H.
So far all seems good for now with the new drive.
Since Parados is initialized 4 times, some bit is not decoded correctly.
That would be either databit 1 or 2, when doing a rom select, since the 464 only initializes roms 0-7 and basic isn't overwritten (bit 0). I did not study the schematic.
Quote from: Duke on 21:43, 26 November 17
Since Parados is initialized 4 times, some bit is not decoded correctly.
That would be either databit 1 or 2, when doing a rom select, since the 464 only initializes roms 0-7 and basic isn't overwritten (bit 0). I did not study the schematic.
This is strange because the AMSDOS is initialized only once.
Anyway, the problem seems to have been fixed after replacing the 74LS136 with a 7486 chip.
Quote from: SilverGreen93 on 21:55, 26 November 17
This is strange because the AMSDOS is initialized only once.
Anyway, the problem seems to have been fixed after replacing the 74LS136 with a 7486 chip.
AMSDOS is silent though, so it would not be obvious unless you check the rom space.
EDIT:
You could put maxam on and check the upper roms 0-7 to look for any mirrors.
Or just check from basic ? himem
Quote from: SilverGreen93 on 21:55, 26 November 17Anyway, the problem seems to have been fixed after replacing the 74LS136 with a 7486 chip.
Not a good idea, the 136 is used specifically because it has open-collector outputs, and the 3 outputs used by the rom number decode are tied together. With an 7486 the outputs will be fighting each other.
Quote from: IanS on 00:54, 27 November 17
Not a good idea, the 136 is used specifically because it has open-collector outputs, and the 3 outputs used by the rom number decode are tied together. With an 7486 the outputs will be fighting each other.
Yeah, I know, but now it works. Maybe I had a faulty one. Until I get new parts, I will continue to use the 7486. Although, if I look at the schematic I think the only "damage" that can be made is to the 7486 itself, due to the tied outputs of the 3 decoder gates.
LE: On a second thought, this may impact also the functionality. Anyway, the exact parts should arrive from TME next week.
All the testing is completed. It works great. :)
Unfortunately I don't have all the necessary parts yet to make more. I have only 2 complete boards for now.
I plan to ship the first board in January to the first person that sent me a message.
The price will be around 45EUR + shipping. Standard shipping should be around 12EUR to any country in the EU. I will need to quote the final price afterwards.
Due to the fact that not all the parts I found are from the same manufacturer or era, I want to test them before shipping to make sure you don't get a faulty one, so I might ditch the DIY part and sell only fully assembled and tested ones.
If this is ok with you all guys, I will keep you posted about the progress.
P.S.: I can also sell them without the D765 and FDC9229BT if anyone wants them like that or already has those parts, because those are the parts I am mostly short of right now.
Quote from: SilverGreen93 on 11:26, 22 December 17
All the testing is completed. It works great. :)
Unfortunately I don't have all the necessary parts yet to make more. I have only 2 complete boards for now.
I plan to ship the first board in January to the first person that sent me a message.
The price will be around 45EUR + shipping. Standard shipping should be around 12EUR to any country in the EU. I will need to quote the final price afterwards.
Due to the fact that not all the parts I found are from the same manufacturer or era, I want to test them before shipping to make sure you don't get a faulty one, so I might ditch the DIY part and sell only fully assembled and tested ones.
If this is ok with you all guys, I will keep you posted about the progress.
P.S.: I can also sell them without the D765 and FDC9229BT if anyone wants them like that or already has those parts, because those are the parts I am mostly short of right now.
I'd rather have the kit for the fun of soldering it myself. But if that is no longer available the I am still interested in an assembled board.
Sent from my ONEPLUS 3t using Tapatalk
Congratulations!.
No problem if you assemble it.
One question for all. Colud be used an expansion cable for this?. How long it could be?
Quote from: Querubin on 09:21, 23 December 17
Congratulations!.
No problem if you assemble it.
One question for all. Colud be used an expansion cable for this?. How long it could be?
Well, in theory you could try, but keep it very short.
In practice, it is NOT recommend to use any cable. This will probably result in unreliable behavior and/or data corruption.
Ok, Thanks.
Happy Holidays guys!
Here are some photos of my final CPC464 setup.
I've made a dual-floppy-drive unit, using some cardboard and solid core wire to make hooks to hold it in place on an angle on the CPC case.
Both floppies are modded with wires to provide the READY signal and they are both powered from the DDI-1 interface.
I've also made a gamepad from a SNES controller and a Parallel Centronics adapter for printer.
And of course, an SCART/CGA to VGA adaptor from ebay :) to hook it up an LCD monitor.
Merry Christmas.
Great! Hope have mine soon.
Quote from: Querubin on 10:40, 26 December 17
Merry Christmas.
Great! Hope have mine soon.
me too, Now I am reliving my first computer with my newly restored CPC 464 I need a floppy drive for it LOL.
Just wanted to update you on the status of this project.
I'm still waiting for a couple of components to arrive (D765s and EPROMs). Meanwhile, I need to finish my exams and then I will write a short documentation with pictures for how to assemble the KIT yourself. After publishing it here, I will start shipping for the first units :) . Hopefully this month.
Cheers.
PS.: Meanwhile, anyone that wants it can add himself to the list here (click the link): https://docs.google.com/spreadsheets/d/1GleRhKIo71r2kXYFSYJaaIWSvXhHe1CcnDU0wpnyp_w/edit?usp=sharing (https://docs.google.com/spreadsheets/d/1GleRhKIo71r2kXYFSYJaaIWSvXhHe1CcnDU0wpnyp_w/edit?usp=sharing)
Fingers crossed, no spammers. I will contact each one of you through private message. And I will update the list with shipping dates.
One on the list.
Waiting for my DDI-1.
Finally I received last week the DDI-1 from SilverGreen93 and only can say one thing AMAZING!. Fully recommended. And with the interface comes a printed guide made by SilverGreen93 very usefull.
SilverGreen93, if you make a dk'tronics memory expansion board clone where you can plug on its back the DDI-1 you will be God to me :P .