Jump to content
IGNORED

CV-NUC+ DIY Builders Thread


mytek

Recommended Posts

Over in the Atari forum a separate topic like this one was started for the DIY crowd to post their experiences on building the 1088XEL, XLD and I believe the 576NUC+. It turned out to be a great resource for people trying to troubleshoot any issues that came up. And it also became a gallery showing each persons completed boards.

 

here's my gallery...

 

This is my 1st assembled board (V1.3) where I discovered a placement error in the Cart Port and noise issues with 5V supply

CV-NUC_V1.3.thumb.png.94dd0956fb4a938dd300e5ec0d69c98d.png

 

V1.4 Cart Port Position Corrected and MSX Module audio mixer improvements -- new low noise switch mode 5V regulator in use

CV-NUC_V1.4.thumb.png.f517b15cad0789964caa385c4ba301ff.png

 

 

V1.5 had unneeded -5V circuit removed and CPLD relocated to the backside of the MSX Module (this is the final version ready for production)

CV-NUC_V1.5.thumb.png.80eacad53da3950ac84cc5bbaa8dd8de.png

 

Versions 1.0-1.2 were never made as real boards, and merely existed as early prototypes on paper. V1.3 already had the concept of a Quadrature module implemented, but the Module itself was still in the R&D phase, eventually adding the USB Mouse ability. V1.3-V1.4 had dual Green Leds coming from my 576NUC+ stock, but I soon realized that this part had gone obsolete, with only the Dual Yellow version still being manufactured. So starting with V1.5 Yellow LEDs were being used. V1.3 ultimately was disassembled with many of its components going into V1.4 which got shipped off to Canada to my first Beta Tester and collaborator @chart45. When I received the first V1.5 boards, bare boards were sent to both @chart45 and @MacRorie, with MacRorie becoming my 2nd Beta Tester, and soon after the sole manufacturer and distributor of what had officially become a product and not just a prototype.

 

Read all about the project development phase here...

 

  • Like 4
Link to comment
Share on other sites

boards and CPLD ordered at TBA - will need some time (and money ;-) ) to travel over the ocean to Europe...

actually sourcing missing components - very few are out of stock at DigiKey, but finding other sources is quite simple

"rare" chips came in from China already a few weeks back - hopefully they will work...

 

exciting... ;-)

  • Like 1
Link to comment
Share on other sites

On 6/19/2023 at 1:27 AM, LarryL said:

actually sourcing missing components - very few are out of stock at DigiKey, but finding other sources is quite simple

I kept updating the BOM while I was building up my protos, since things would disappear or become out of stock. Had to keep finding and trying alternatives, and then change the entries in the BOM. That is why I also list a manufacturer's part number to aid someone in finding alternatives themselves, since I don't plan on doing constant BOM updates into eternity ;)

 

An End of Life status for the 32K Static RAM occurred while I was still working on the final board version. Although those should be easily gotten through various alternative sources for sometime to come (e.g., eBay, AliExpress). Presently there are well over 1,100 in stock at Digi-Key P/N: 800-1433-5-ND but OctoPart shows them as the only source with any stock left.

 

EDIT: I recently started using UtSource, and always get working chips from them.

 

TMS9128NL

TMS4416-12NL

YM2149F

SN76489AN

FMS6400

 

Note1: Although it's sometimes cheaper to get the YM2149 instead of the YM2149F this is not advisable since the audio mixing in the CV-NUC+ has been designed for the YM2149F specifically. From what I gather the non-'F' version of the YM chip doesn't have the built-in clock divider or the extra bit of resolution in the sound envelope generator. This makes it a better match for the AY chip. And even though the CV-NUC+ doesn't use the clock divider*, the extra bit of audio level resolution is expected to be present.

 

*Note2: I wish I had known about the YM2149F's clock divider feature during development, since it would have allowed me to eliminate one of the 74HCT74 chips :ponder:

  • Like 1
Link to comment
Share on other sites

On 6/19/2023 at 10:27 AM, LarryL said:

boards and CPLD ordered at TBA - will need some time (and money ;-) ) to travel over the ocean to Europe...


package arrived.

But…

I at least thought that I have ordered the CV-NUC PCBs…

seems I was too stupid to read the TBA website right, and only ordered the programmed ATF1500, but not the PCBs 😞

 

anyway… sent the Gerbers to JLCPCB.

now, soon I will have 5 sets - so 4 would be available for others

preferably somebody from Europe (I am located in Germany) - PM me, if you are interested

Shipping to the US will not make any sense (esp having TBA as local source there)

 

BTW: to make shipping charges reasonable, I had also ordered the PCBs for the 1088XLD - so, very soon another one will come to life 🙂

 

 

  • Like 1
Link to comment
Share on other sites

Not being entirely happy with my 3D printer's abilities and quality, I had TBA print me up a case using one of their Ultimaker printers instead.

 

Here's the result...

2081225299_CV-NUC_top.thumb.JPG.6f1483116cf7b2e7fd9eef5067b681a2.JPG

173764224_CV-NUC_back.thumb.JPG.fa215bd4356ea7f0aefb451d87402579.JPG

150430534_CV-NUC_btm.thumb.JPG.17e142f03dfc273c31f28dcb2fa2a6bf.JPG

 

This came out much better than what I was able to do on my AnyCubic Kobra Plus. Of course with the price difference between these, it's no real surprise. The top and bottom have a mirror finish. And for the keypad mounting, I counter sunk the holes myself to accept flat head screws (personal preference). This is not something easily 3D printed due to break out issues, so alternatively pan head screws work out just fine, thus not requiring any case modding. The bottom holes come already countersunk (no break out issues to worry about).

 

These cases can be purchased for around $45 from TBA and I think they are well worth it :)

 

When you order one of the fully assembled units, this case is part of that deal.

  • Like 1
Link to comment
Share on other sites

2 hours ago, LarryL said:

package arrived.

But…

I at least thought that I have ordered the CV-NUC PCBs…

seems I was too stupid to read the TBA website right, and only ordered the programmed ATF1500, but not the PCBs 😞

Sorry for the mix up :_(

 

However I do look forward to seeing your build progress. Be sure to post pics :)

Link to comment
Share on other sites

6 hours ago, mytek said:

Sorry for the mix up :_(

 

However I do look forward to seeing your build progress. Be sure to post pics :)


thanks - has been completely my own fault

 

I will surely post pics!

 

  • Like 1
Link to comment
Share on other sites

Hi,

 

components are arriving, step by step…

PCBs are on the way in from China - maybe another week…

 

but the case arrived today.

got it printed (by a service provider nearby) in silver - came out quite well…


IMG_0442.thumb.jpeg.ee14ece399781a21e2bd9fa872a49dbe.jpeg

 

IMG_0443.thumb.jpeg.0ab45a35af184284e8467bcf98816be5.jpeg

 

IMG_0444.thumb.jpeg.c3c15783607d81cd3a0e0eafdc10c8f8.jpeg

 

IMG_0445.thumb.jpeg.3a6fcf750e297b35fb0ebe8d7c5d8427.jpeg

 

Since I am traveling the next 3 weeks (summer holidays) the build has to wait until I return

 

cheers

Michael

 

  • Like 1
Link to comment
Share on other sites

3 hours ago, LarryL said:

but the case arrived today.

got it printed (by a service provider nearby) in silver - came out quite well…

Looks quite good 👍

 

Something to keep in mind when doing the assembly with the PCB in place, is that first of all it's a tight fit between the front and the back, and that back panel will tend to bow outwards until the the top is secured with the top edge of the back panel sliding into the slot on the top. I also see that the sides are slightly bowed outwards (kinda normal for some prints). This keeps the sides from lining up completely with the top.

 

image.png.10509bcb6b0ce69ce86691b3d8793c06.png

 

My own fix for this... after I'm sure that all is working properly with the electronics, I put one drop of a quick dry adhesive in the top of the side slots and press the sides inward until the adhesive has dried enough to hold the assembly together. I specifically call out one drop of glue, because just in case you need to remove the Main PCB at a future date it's probably still going to be a possibility.

Link to comment
Share on other sites

Good advice, cool

until now my case is still empty, so I just quickly put the back inside and dropped the top on the bottom

nothing is screwed together, yet

just wanted to get an impression, how the case will look like

 

but, yes - it seems to be a tight fit

 

I remember, when assembling the 576NUC+ it was also pretty tight…

 

 

  • Like 1
Link to comment
Share on other sites

On 7/4/2023 at 10:22 AM, LarryL said:

PCBs arrived - black beauties…

Looking good 👍

 

Something to take note of when it comes to assembling the Main PCB, is that the coaxial power jack pads are a bit tight on the first run boards. I didn't catch this until after @MacRorie had already ordered his first batch of boards. It kinda comes down to the luck of the draw with how tight the pad holes vs. the pins on the power jacks will be, with some taking only a slight pressure to insert, whereas others may need a little wiggling and more pressure to get them to seat. But it is important to make sure that the bottom of the plastic housing is flush with the board before soldering it in place, otherwise there may be alignment problems with the hole in the case's back panel.

 

PwrJack.png.5595a5cb2dec2f45238c940c8e655df2.png

 

This was properly resolved before I uploaded the gerbers to my site, so anyone going the completely DIY path and had boards made from those files should be good :)

  • Like 1
Link to comment
Share on other sites

19 minutes ago, mytek said:

Looking good 👍

 

Something to take note of when it comes to assembling the Main PCB, is that the coaxial power jack pads are a bit tight on the first run boards. I didn't catch this until after @MacRorie had already ordered his first batch of boards. It kinda comes down to the luck of the draw with how tight the pad holes vs. the pins on the power jacks will be, with some taking only a slight pressure to insert, whereas others may need a little wiggling and more pressure to get them to seat. But it is important to make sure that the bottom of the plastic housing is flush with the board before soldering it in place, otherwise there may be alignment problems with the hole in the case's back panel.

 

PwrJack.png.5595a5cb2dec2f45238c940c8e655df2.png

 

This was properly resolved before I uploaded the gerbers to my site, so anyone going the completely DIY path and had boards made from those files should be good :)

"how tight the pad holes"
You make the project sound sexy....

  • Haha 2
Link to comment
Share on other sites

  • 4 weeks later...

So, finally started - and progressing

IMG_0917.thumb.jpeg.56455ebf4769571119f3b0e8e29e79f8.jpeg

 

still waiting for some parts, since I have realized that some of my parts did not perfectly fit.

like the DC barrel jack (way to large pins) and the DB9 (mine have sockets for screws)

 

and yes, I have read the post, that the barrel jack pins are a tight fit - but mine do not really have pins, but way larger „soldering lugs“

 

Anyway, satisfied so far…

 

…will it boot first time? Let’s see 🙂 

 

edit: now the whole bunch…

IMG_0927.thumb.jpeg.93b11ce2d224ad47cd5fbc9b1a5eeb95.jpeg

Link to comment
Share on other sites

On 8/4/2023 at 8:15 AM, LarryL said:

still waiting for some parts, since I have realized that some of my parts did not perfectly fit.

like the DC barrel jack (way to large pins)... and yes, I have read the post, that the barrel jack pins are a tight fit - but mine do not really have pins, but way larger „soldering lugs“

You must have ordered through an alternate vendor or manufacturer. Since these are the ones from Digi-Key that are listed on the BOM. Previously on one of my other boards (1088XEL) I used the jack you are referring to, and had to make the pad holes very large. Later I discovered these which are much nicer and of course a smaller pin.

 

image.thumb.png.89dd0e7f21acdb670cd8a4ff28656023.png

 

On 8/4/2023 at 8:15 AM, LarryL said:

...and the DB9 (mine have sockets for screws)

Well the ones listed in the BOM do have a place for a jack screw to go (which we don't use), but it sounds like you got the ones with a jack screw integrated with the housing (non-removable).

 

image.thumb.png.7d2e51b21efaf9e9e59e8575fee82020.png

 

-----------------

 

Your boards are looking good 👍

Link to comment
Share on other sites

Yeah, I had the parts in my stock and was too lazy to check/compare them against the BOM…

 

but I got them quite quickly and was able to complete the board today 🙂 

 

IMG_0609.thumb.jpeg.3ba4cec39ddbe0978129d770547ddf9c.jpeg

 

tomorrow is „smoke test“ 🙂

 

 

 

  • Like 1
Link to comment
Share on other sites

6 hours ago, LarryL said:

tomorrow is „smoke test“ 🙂

Remember power test first just as you have it to verify 5V, then go ahead and plug stuff in (with power off).

 

EDIT: Looks like you got different 40 pin sockets as well. the ones I spec'ed only have a center crossbar, whereas yours are machine pin variants with two cross bars. Looks like you still got it to work fine never the less :)  In fact it looks like through sheer luck the nested part placement was good for two cross bars as well.

 

This shows the sockets specified based on the BOM listing. The nested components were arranged with this style socket in mind for clearance. They are actually excellent dual leaf sockets that I've found to retain a very solid grip even after repeated chip pulls, and they go for a much cheaper price than a machine pin version (although one time I bought a bunch for cheap from China, but they turned out to be junk unfortunately).

image.thumb.png.a735e0127283e802207ccbe501e586d2.png

  • Thanks 1
Link to comment
Share on other sites

41 minutes ago, mytek said:

Remember power test first just as you have it to verify 5V, then go ahead and plug stuff in (with power off).

 

EDIT: Looks like you got different 40 pin sockets as well. the ones I spec'ed only have a center crossbar, whereas yours are machine pin variants with two cross bars. Looks like you still got it to work fine never the less :)  In fact it looks like through sheer luck the nested part placement was good for two cross bars as well.

 

This shows the sockets specified based on the BOM listing. The nested components were arranged with this style socket in mind for clearance. They are actually excellent dual leaf sockets that I've found to retain a very solid grip even after repeated chip pulls, and they go for a much cheaper price than a machine pin version (although one time I bought a bunch for cheap from China, but they turned out to be junk unfortunately).

image.thumb.png.a735e0127283e802207ccbe501e586d2.png


right, used the machine sockets I already had laying around.

had some bad experience with cheap Dual wipe sockets recently…

 

thought you designed it intentionally 🙂

was a perfect fit with the two bars!
IMG_0610.thumb.jpeg.54ac0d9f3f20ebb7e45f59d5306a43fd.jpeg

 

IMG_0611.thumb.jpeg.73be1c7ac6e5cee551435f6d9212b71e.jpeg


but during checking, I figured out a mistake I did:

since I did not read the build manual carefully enough, I added the wrong RN1,2 - 9pin, not 8pin

I then used SMD 10k resistors on the back from the 9th pin to pin 10 of the LS541 (not from the empty pin to pin 10)

stupid… although the silk screen shows exactly a 8pin resistor array 

Now I have to try to get the RN1,2 out again without damaging stuff…

 

 

Link to comment
Share on other sites

1 hour ago, LarryL said:

I added the wrong RN1,2 - 9pin, not 8pin

I have to take a little bit of blame, since the BOM originally had a description of a 9-pin SIP even though the part number and the schematic were calling out an 8-pin device. This got pointed out to me a couple weeks ago, and has been corrected on the download from my website. I should have probably also mentioned it here as well - sorry about that :(

 

EDIT: It used to be a 9-pin SIP providing a pull-up on all 8 inputs of the 74LS541 chips until it was discovered that AtariSoft games didn't like it (thanks for discovering that @Falonn ). So the silk screen got altered to only show an 8-pin device and hence the instructions to add the 10K resistors from that 9th pin to GND. Unfortunately when correcting the BOM to reflect the change I missed updating the description. The absolute best recourse would have been to re-layout that part of the board to provide a proper through-hole spot for those pull-down resistors, but it would have been a big challenge since there was virtually no room to do so.

 

image.thumb.png.b4259b9c8cfe05b712a9e19434ae7b5f.png

Link to comment
Share on other sites

Yeah, it is really tight in this area…

seems I got confused by the BOM I have downloaded some weeks ago and took an existing 9p RN from my stock

No problem, I will manage…

 

BTW: since the system works even with the 9p RN, I gave it a try - and it booted right the first time 🙂

of course I did the voltage checks upfront (without chips)

got confused a second, measuring 12V at the CD4066 - but checking the schematics proves that this is right 🙂

 

now I have to find my AV2HDMI adapter, because my PAL LCD shows quite funny colors trying to display NTSC ;-)

I know I bought one recently and also successfully used it with another NTSC project (the N8 from retrobrewcomputers.org) - but where is it? 

at least the CV-NUC shows a picture and I was able to play a short round of Donkey Kong 🙂 

Joystick and Keypad works, Sound works…

so far so good

 

Had some problems with the  QUAD board initially - no USB device detected.

but after checking and reflowing the SMD chip, all went well and I got it programmed

 

but first I have to fix the resistor arrays, then I will do some more intensive checks with the test ROMs you provided (copied them already to the Atarimax Cart)

 

it is really a great project, all fitting so nicely together - a really excellent design @mytek

 

  • Thanks 1
Link to comment
Share on other sites

Ok, got the resistor networks out, using the process from above video

worked surprisingly well.

they did not fall out immediately, but with a little bit of wiggling they have lost the battle ;-) 

 

IMG_0614.thumb.jpeg.012849483c173be642b45f3192c41de9.jpeg

 

 

now I need to wait for the new resistor arrays to arrive

 

did some case work in the meantime.

needed to remove a bit of PLA with a Dremel at the front inner side of the case (as described in one of the above posts)

now the back panel goes in smoothly

 

also melted the M2.5 threaded inserts into the top of the case - but this will require longer screws compared to e.g. the XEP-80 II

seems that M2.5x30mm will fit

 

IMG_0612.thumb.jpeg.d41b884037d890b8b62ee9b0daf2291e.jpeg

 

Keypad also fits very well - Used small M2.5 screws to secure it

 

IMG_0613.thumb.jpeg.44b81d5ac22a68f5b98ea02b71e24a1f.jpeg

 

 

 

 

  • Like 1
Link to comment
Share on other sites

4 hours ago, LarryL said:

Ok, got the resistor networks out, using the process from above video

worked surprisingly well.

they did not fall out immediately, but with a little bit of wiggling they have lost the battle ;-)

Glad that worked for you. It is a much safer way to remove SIPs vs. a solder sucker.

 

8 hours ago, LarryL said:

BTW: since the system works even with the 9p RN, I gave it a try - and it booted right the first time 🙂

Perfect 👍

 

8 hours ago, LarryL said:

now I have to find my AV2HDMI adapter, because my PAL LCD shows quite funny colors trying to display NTSC ;-)

I know I bought one recently and also successfully used it with another NTSC project (the N8 from retrobrewcomputers.org) - but where is it? 

at least the CV-NUC shows a picture and I was able to play a short round of Donkey Kong 🙂 

Joystick and Keypad works, Sound works…

so far so good

So far so good. Please be sure to post a screenshot when you eventually find the adapter.

 

4 hours ago, LarryL said:

did some case work in the meantime.

needed to remove a bit of PLA with a Dremel at the front inner side of the case (as described in one of the above posts)

now the back panel goes in smoothly

Yep although it might have been better to leave a bit more slop in the design, having to do a tiny amount of adjustment with a Dremel or file really isn't all bad.

 

4 hours ago, LarryL said:

also melted the M2.5 threaded inserts into the top of the case - but this will require longer screws compared to e.g. the XEP-80 II

seems that M2.5x30mm will fit

The case is 10 mm taller than either the XEP80-II or the 576NUC+ so that length should work fine. As a precaution I like to run a tap through the insert to clear out any PLA that might have squirmed its way inside.

 

Looks like your printer did a great job :)

 

8 hours ago, LarryL said:

it is really a great project, all fitting so nicely together - a really excellent design @mytek

Thank you. I put a lot of love and patience into this one 🥰

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...