reifsnyderb Posted February 15 Share Posted February 15 24 minutes ago, Dropcheck said: So what can go wrong now? (I didn't actually say that did I? Dumb....🤕) Lots of things, unfortunately. I always like to wait another day or two after designing a new board. That way, when I realize I missed something a day later the board isn't in production yet. I learned to do this from too many bad experiences of realizing I missed something the next day. 😞 4 Quote Link to comment Share on other sites More sharing options...
AtariGeezer Posted February 15 Share Posted February 15 1 hour ago, Dropcheck said: The BOM is done, I have managed to complete a test route of the board as 4 layer. Did you swap the part numbers for Q3 and Q8 and change R141 to 1k instead of 100 ohm? Quote Link to comment Share on other sites More sharing options...
AtariGeezer Posted February 15 Share Posted February 15 (edited) Delete dbl post Edited February 15 by AtariGeezer Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted February 15 Author Share Posted February 15 7 hours ago, AtariGeezer said: Did you swap the part numbers for Q3 and Q8 and change R141 to 1k instead of 100 ohm? Yes. Thanks for the schematics on the parallel drive board. That's part of the next stage in the evolution of the board. 2 Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted February 16 Author Share Posted February 16 Ran into this issue while verifying schematic to the board I have in hand and reading about the controversy over which jumper to use for which modem chip, W3 or W4. In @Vandal968's BOM spreadsheet He indicates that W4 should never be used as it connects +12V directly to the modem and will fry it. He advises to use W3 for both modems. According to the schematics (Atari & Sobola) that is true. Here's the rub, a direct hands on trace of both W3 and W4 reveals this board truth. Do I need to get a new multimeter? This seems to be exactly the opposite. Please someone check me! Quote Link to comment Share on other sites More sharing options...
+kheller2 Posted February 16 Share Posted February 16 On my 1450 short board, w3 is jumpered for TI. W4 is open. Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted February 16 Author Share Posted February 16 (edited) 27 minutes ago, kheller2 said: On my 1450 short board, w3 is jumpered for TI. W4 is open. That is what the schematic and @Vandal968 are indicating. But ohm out the connections. Does one side of W3 go directly to +12V? +12V can be found on the middle pin of the floppy drive power connector. I'm not saying it doesn't work. Edited February 16 by Dropcheck Quote Link to comment Share on other sites More sharing options...
+kheller2 Posted February 16 Share Posted February 16 13 minutes ago, Dropcheck said: That is what the schematic and @Vandal968 are indicating. But ohm out the connections. Does one side of W3 go directly to +12V? +12V can be found on the middle pin of the floppy drive power connector. I'll check in a bit, but have you looked at the modem card itself in the pics to make sure it's going to where you think it is? I also have pics of a an original wire soldered perf board from a 1400 board. W3 is also set, but you can see the traces better on the wires. Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted February 16 Author Share Posted February 16 (edited) 11 minutes ago, kheller2 said: I'll check in a bit, but have you looked at the modem card itself in the pics to make sure it's going to where you think it is? I also have pics of a an original wire soldered perf board from a 1400 board. W3 is also set, but you can see the traces better on the wires. Right now I'm just going with the on board connections. @Vandal968 apparently did not use the modem daughtercard. He went with a direct MM74HC942 National chip in the U35 socket. I think @AtariGeezer also went directly with the Nat chip in that socket. The daughter card is going to be a bit problematic as there are no schematics that I am aware of. If the modem on chips are available, there's no real reason to use it either. Not that I wouldn't try to offer it as an option, just that it would be the last resort if you can't find either modem chips. 🙂 Edited February 16 by Dropcheck Quote Link to comment Share on other sites More sharing options...
+bent_pin Posted February 16 Share Posted February 16 On 12/1/2023 at 10:29 PM, reifsnyderb said: Dimensions are needed Scale it off the chips of known size. Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted February 16 Author Share Posted February 16 Placed an order for the passives (Caps and Resistors) as if I didn't have a donor board to pull from. The cost of the axial caps alone was in the neighborhood of $70 from Digikey. Some of the expense is due to having to substitute higher voltage caps than called for because neither Digikey or Mouser stock the correct capacitance/voltage. Another reason is that the quantity is so small. Resistors weren't a problem though, remaining relatively cheap and available. I think for my to do list for the next version is converting the axial caps to radial. The radial seems to still have a greater selection of capacitance/voltage range available. Quote Link to comment Share on other sites More sharing options...
+kheller2 Posted February 16 Share Posted February 16 4 hours ago, Dropcheck said: That is what the schematic and @Vandal968 are indicating. But ohm out the connections. Does one side of W3 go directly to +12V? +12V can be found on the middle pin of the floppy drive power connector. I'm not saying it doesn't work. Yes it does. And that is tied to pin 15 of the modem board header. That runs through a green resistor looking cap to various other things on the daughter card. Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted February 17 Author Share Posted February 17 4 hours ago, kheller2 said: Yes it does. And that is tied to pin 15 of the modem board header. That runs through a green resistor looking cap to various other things on the daughter card. So +12V is going thru W3, and not as depicted thru W4 in the schematics. If W4 were not populated, then +12V would never get to W3 according to the schematics. So that means the schematics have to be wrong or the pcb silkscreen legend is wrong(That's happened more than once already.) Okay mystery solved. The correct pcb tracing will be reflected in both schematic and silkscreen now. Thank you @kheller2! Quote Link to comment Share on other sites More sharing options...
+kheller2 Posted February 17 Share Posted February 17 3 hours ago, Dropcheck said: So +12V is going thru W3, and not as depicted thru W4 in the schematics. If W4 were not populated, then +12V would never get to W3 according to the schematics. So that means the schematics have to be wrong or the pcb silkscreen legend is wrong(That's happened more than once already.) Okay mystery solved. The correct pcb tracing will be reflected in both schematic and silkscreen now. Thank you @kheller2! Interesting that every drawing has that as W4 going to +12V, including the TONG! It's just semantics -- fix the silkscreen, leave the schematics be. 😁 I find the silkscreen to be wrong most of the time, especially on prototypes. I needed a spreadsheet to track all the duplicated component numbers used on the 128K 800XL board. 1 Quote Link to comment Share on other sites More sharing options...
AtariGeezer Posted February 17 Share Posted February 17 (edited) Yep, the silkscreen has W3 and W4 backwards. A note to builders: W3 0 ohm, For National Modem 74HC942 W4 0 ohm, For Texas Inst Modem TMS99532 U35 layout is a direct placement for the 74HC942, the TMS99532 Has To Be Used On A Daughter board... U35 pin 12 VBB should go to -5V, not +12 as in Jerzy's schematic. Edited February 17 by AtariGeezer 2 Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted February 17 Author Share Posted February 17 2 hours ago, AtariGeezer said: Yep, the silkscreen has W3 and W4 backwards. A note to builders: W3 0 ohm, For National Modem 74HC942 W4 0 ohm, For Texas Inst Modem TMS99532 U35 layout is a direct placement for the 74HC942, the TMS99532 Has To Be Used On A Daughter board... U35 pin 12 VBB should go to -5V, not +12 as in Jerzy's schematic. Based on @Vandal968's last BOM dated 8/30/2011 W3 should be used for both modems. W4 never. Now I don't know if the TMS99532 has to be in the daughterboard. That's entirely possible. Quote Link to comment Share on other sites More sharing options...
AtariGeezer Posted February 17 Share Posted February 17 1 hour ago, Dropcheck said: Based on @Vandal968's last BOM dated 8/30/2011 W3 should be used for both modems. W4 never. Now I don't know if the TMS99532 has to be in the daughterboard. That's entirely possible. Vandal968 didn't check the datasheets I guess... The TI modem does use +12v at VDD, plus +5v and -5v. So W3 on silkscreen, W4 on schematic (+12v) is used here... W4 on silkscreen, W3 on schematic is used to tie pins 15 and 17 on the National 74HC942... I made a schematic for the TI daughter board, but can't find it at the moment. 74hc942.pdfTMS99532A.PDF 1 Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted February 17 Author Share Posted February 17 3 hours ago, AtariGeezer said: Vandal968 didn't check the datasheets I guess... The TI modem does use +12v at VDD, plus +5v and -5v. So W3 on silkscreen, W4 on schematic (+12v) is used here... W4 on silkscreen, W3 on schematic is used to tie pins 15 and 17 on the National 74HC942... I made a schematic for the TI daughter board, but can't find it at the moment. 74hc942.pdf 146.36 kB · 0 downloads TMS99532A.PDF 114.81 kB · 0 downloads Probably didn't check this one, thinking that his board the 1450 might be a bit different. Or he did not know of this schematic. Because this schematic or portion there of apparently has renumbered the various chips. Something that I think he would have noted on his BOM. Also there's R145 with no declarable value at the the junction of W3 and W4, with a negative -5V applied at the other end that is board truth. Not on any of the previously mentioned schematics. @Vandal968 lists that resistor as not installed. I wish he would respond to my IMs and email. But he appears to be MIA. Hopefully well, just not in the Atari world anymore. Mainly what I'm trying to do is to go with the board truth, regardless of what the schematics say. I'm including R145 for historical reasons on this board and because it is board truth for this version of the prototype 1450XL. It could simply have been a connection that was decided against after the board was sent to manufacturing and therefor unable to be deleted. I will correct the silkscreen and my schematics to reflect the board truth. I believe @kheller2 is using the daughterboard with a TI modem and is connecting to the physical W3 connection. I don't believe I've ever heard anyone using W4. But there's been so few actually built up this board, that's unfortunately not an absolute validation. Hopefully on the next version some of these inconsistencies will be resolved. Quote Link to comment Share on other sites More sharing options...
+kheller2 Posted February 17 Share Posted February 17 I have never seen or heard of an original Atari built board that didn’t use the daughter card. Even the original 1400 prototype boards with little to no silkscreens used the daughter card. Having said that, I don’t know what TONG wound up using onboard (as no one has ever posted hi resolution pictures nor offered to sell one to me). Dynasty boards (all 150 plus prototype 1400) are everywhere compared to TONG. Scarcity goes like this: 1450, 1400, TONG, 130XE in 800XL form. AtariGeezer’s diagram I think is from TONG. There is a whole set done by hand. Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted February 17 Author Share Posted February 17 2 hours ago, kheller2 said: I have never seen or heard of an original Atari built board that didn’t use the daughter card. Even the original 1400 prototype boards with little to no silkscreens used the daughter card. Having said that, I don’t know what TONG wound up using onboard (as no one has ever posted hi resolution pictures nor offered to sell one to me). Dynasty boards (all 150 plus prototype 1400) are everywhere compared to TONG. Scarcity goes like this: 1450, 1400, TONG, 130XE in 800XL form. AtariGeezer’s diagram I think is from TONG. There is a whole set done by hand. You are probably correct on the original Atari built boards, but I believe @Vandal968 indicated he was using the Nat modem chip directly in the U35 socket and if I remember the thread correctly was trying to convince @AtariGeezer to do the same on the bare pcb purchased from Best Electronics. I believe you are correct on the TONG modem schematics too. It's labeled as 1450XLD, which I think is the TONG by another name. Quote Link to comment Share on other sites More sharing options...
+kheller2 Posted February 17 Share Posted February 17 Here is a pick of Vandal968's board and yes he did use the NAT chip: And continuing the discussion on W3 vs W4, the 1400 schematics are the same as the 1450 (when it comes to the modem section), but the silk screens on the 1400 use W3/W2 vs W4/W3 on the 1450: I'm amazed anyone could build anything back then following any schematics Quote Link to comment Share on other sites More sharing options...
AtariGeezer Posted February 18 Share Posted February 18 (edited) 17 hours ago, AtariGeezer said: The above pic is from the 1450XLD Hand Drawn Schematics for reference on the modem circuit. And yes, these are TONG... 1450XLD Hand Drawn Schematics.zip Edited February 18 by AtariGeezer 1 Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted February 20 Author Share Posted February 20 On 2/14/2024 at 6:24 PM, reifsnyderb said: Lots of things, unfortunately. I always like to wait another day or two after designing a new board. That way, when I realize I missed something a day later the board isn't in production yet. I learned to do this from too many bad experiences of realizing I missed something the next day. 😞 You were right. While going through and cleaning up the schematics for readability I discovered I had not connected all the ram control lines to RP2. Duh....! Needless to say the board order did not go out. I'm waiting for my capacitor order to come in from Digikey Thursday to verify sizing for the electrolytic caps. Maybe this Sunday......? 1 Quote Link to comment Share on other sites More sharing options...
reifsnyderb Posted February 20 Share Posted February 20 31 minutes ago, Dropcheck said: You were right. While going through and cleaning up the schematics for readability I discovered I had not connected all the ram control lines to RP2. Duh....! Needless to say the board order did not go out. I'm waiting for my capacitor order to come in from Digikey Thursday to verify sizing for the electrolytic caps. Maybe this Sunday......? It's so easy to miss something. Good to hear you caught the problem. 2 Quote Link to comment Share on other sites More sharing options...
+mytek Posted February 20 Share Posted February 20 2 hours ago, reifsnyderb said: It's so easy to miss something. This is unfortunately so true, because I have yet to have a new motherboard that doesn't require at least 3 PCB runs to get to the final version. This has been the case for: 1088XEL, 1088XLD, 576NUC+, and the CV-NUC+ Quote Link to comment Share on other sites More sharing options...
Like us on Facebook
Follow us on Twitter
AtariAge on Instagram
Watch us on YouTube
Chat on Discord
Recommended Posts
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.