View unanswered posts | View active topics It is currently Thu Mar 28, 2024 9:54 pm



Reply to topic  [ 266 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6, 7, 8 ... 18  Next
 EPiC - A new 68k multi-processor motherboard project 
Author Message

Joined: Tue Jun 03, 2014 2:40 pm
Posts: 127
Bad news... I shorted something out and fried my CPU. I noticed the address bus was dead and the chip was very hot, so I put it on its own separate breadboard and hooked nothing but power to it and once again the chip started burning up. More are on order, and things are on hold until they come in. :(


Fri Jul 04, 2014 6:00 pm
Profile

Joined: Wed Jan 09, 2013 6:54 pm
Posts: 1780
Ouch!


Fri Jul 04, 2014 7:44 pm
Profile
User avatar

Joined: Tue Jan 15, 2013 5:43 am
Posts: 189
It'd nice if we knew how the CPU got cooked so it doesn't happen again. :|

In a previous post you were talking about power supplies, and perhaps you've hard a curve ball pitched in your direction. Surprisingly, perhaps, some PSUs have a specification for minimum load current. IOW you need to draw at least a certain amount of current or else it might misbehave. And the misbehavior could include the voltage rising above nominal -- enough to fry your CPU, perhaps. If you have a PSU with no specs available it's best to experiment first. Try it with no load and see if the voltage is correct. If it's not, connect a dummy load and see if things settle down. Or use a different PSU.

Another possible cause of cookedness is electrostatic discharge, discussed on 6502.org here:
http://forum.6502.org/viewtopic.php?f=1&t=2859

-- Jeff

_________________
http://LaughtonElectronics.com


Sat Jul 05, 2014 8:05 am
Profile WWW

Joined: Tue Jun 03, 2014 2:40 pm
Posts: 127
First off, a small update. As you can see from the attached picture, the dev board is starting to take shape. I've got the address decoder mounted and connected finally, so as soon as I get more wires - and a new CPU - I can once again proceed. The breadboard-mounted PS/2 ports came in and I got waveforms out of a keyboard using my logic analyzer, so I think I have a pretty good idea now of how I'm going to tie them into an additional 68681 to handle keyboard and mouse input. All the components are pretty much in place, until I add the parts necessary for onboard video.



Dr Jefyll wrote:
It'd nice if we knew how the CPU got cooked so it doesn't happen again. :|


I know exactly what I did which caused the short, I just don't fully understand why it killed the 68K. I shorted positive to negative on one of the breadboard's power rails, right next to the BGACK and DTACK lines. I would think the short would only mess up things from itself back towards the power supply, but I can only assume that the short's close proximity to the CPU made it catch some damage as well.



Dr Jefyll wrote:
In a previous post you were talking about power supplies, and perhaps you've hard a curve ball pitched in your direction. Surprisingly, perhaps, some PSUs have a specification for minimum load current. IOW you need to draw at least a certain amount of current or else it might misbehave. And the misbehavior could include the voltage rising above nominal -- enough to fry your CPU, perhaps. If you have a PSU with no specs available it's best to experiment first. Try it with no load and see if the voltage is correct. If it's not, connect a dummy load and see if things settle down. Or use a different PSU.


I actually thought of that before I switched to a PC PSU for my power source. I tested all the lines and even under zero load, they all put out an appropriate voltage - in fact, it was surprisingly precise! I tested the board for almost a full two days of straight free-run on PSU power with zero adverse effects. Until I clumsily crossed wires, that is. :roll: All the other components which were connected at the time - the clock pulse generator and address decoder - still function flawlessly. Only the CPU was affected.


Attachments:
EPiC breadboard layout.jpg
EPiC breadboard layout.jpg [ 420.83 KiB | Viewed 10885 times ]
Sun Jul 06, 2014 3:03 am
Profile

Joined: Tue Jun 03, 2014 2:40 pm
Posts: 127
Do I need bypass caps on just on the CPU-like chips (such as the 68K itself and the 68681s) or on each and every one (the RAM, EPROM, RTC, MAX232, etc.)?


Sun Jul 06, 2014 12:29 pm
Profile

Joined: Wed Jul 02, 2014 9:36 pm
Posts: 32
Unless you actually need the power and/or the voltages you get from the PC PSU, I would ditch that for a supply with current limiting capabilities. That way you could set it to just slightly higher than your needs, and the supply would go offline instead of killing your components.

I've shorted the supply a couple of times myself, and it's saved me just fine then.


Sun Jul 06, 2014 12:33 pm
Profile

Joined: Tue Dec 11, 2012 8:03 am
Posts: 285
Location: California
mercury0x000d wrote:
Do I need bypass caps on just on the CPU-like chips (such as the 68K itself and the 68681s) or on each and every one (the RAM, EPROM, RTC, MAX232, etc.)?

The lack of them probably won't do physical damage to the parts, but they definitely help operation. Solderless breadboards are already absolutely the worst of all worlds for performance in this kind of circuit (because of the long, curved connections), so yes, you need the bypass capacitors on every digital IC.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources


Sun Jul 06, 2014 3:05 pm
Profile WWW

Joined: Tue Jun 03, 2014 2:40 pm
Posts: 127
stefanberndtsson wrote:
Unless you actually need the power and/or the voltages you get from the PC PSU, I would ditch that for a supply with current limiting capabilities. That way you could set it to just slightly higher than your needs, and the supply would go offline instead of killing your components.

I've shorted the supply a couple of times myself, and it's saved me just fine then.


I do intend the final motherboard to conform to the ATX form-factor so that it can easily be mounted in a standard PC case and use typical PC peripherals, drives and the PSU. What type of supply do you use?



Garth wrote:
mercury0x000d wrote:
Do I need bypass caps on just on the CPU-like chips (such as the 68K itself and the 68681s) or on each and every one (the RAM, EPROM, RTC, MAX232, etc.)?

The lack of them probably won't do physical damage to the parts, but they definitely help operation. Solderless breadboards are already absolutely the worst of all worlds for performance in this kind of circuit (because of the long, curved connections), so yes, you need the bypass capacitors on every digital IC.


Okay, thanks. Will do.


Mon Jul 07, 2014 2:10 am
Profile

Joined: Wed Jul 02, 2014 9:36 pm
Posts: 32
mercury0x000d wrote:
I do intend the final motherboard to conform to the ATX form-factor so that it can easily be mounted in a standard PC case and use typical PC peripherals, drives and the PSU. What type of supply do you use?


I have a UNI-T UTP3702. Probably not all that spectacular, but it was available in a store just nearby. Two outputs, 0-30V, 0-3A on each. So far it's been very useful.


Mon Jul 07, 2014 10:38 am
Profile

Joined: Tue Jun 03, 2014 2:40 pm
Posts: 127
I'm negotiating with a fab house for a quantity discount on a batch of 68060-to-breadboard adapters to facilitate the creation of low-speed prototypes using this chip. Would any of you here be interested in purchasing one? Once I've gauged the interest I can get a per-piece quote.


Mon Jul 07, 2014 7:30 pm
Profile

Joined: Tue Dec 11, 2012 8:03 am
Posts: 285
Location: California
Every board house I've dealt with has a price formula they apply. Especially for smallish quantities (under a thousand pieces), they probably won't be open to negociation. You can however get online quotes with many of them, entering the number of layers, holes, hole sizes, trace & space size, size, qty, etc., and see how varying the different things will affect the price. Even if they won't give you the exact forumula they use, you can kind of do a curve fit after you get enough data points, and figure out how much each of the various things cost on a per-piece basis and how much the hidden set-up fee may be (in addition to explicit tooling charges).

If you're doing your own design, I strongly recommend showing it on the forum to run past others before getting it made, since fast processors and other ICs ("fast" referring especially to rise times, even if you run it at a slow clock rate) can easily be disabled by stray inductance. Even after being in this stuff for decades, I'm still becoming more and more impressed by the effects. And again, refer back to the links in my first post on page 1 of this topic.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources


Mon Jul 07, 2014 8:16 pm
Profile WWW

Joined: Tue Jun 03, 2014 2:40 pm
Posts: 127
Garth wrote:
Every board house I've dealt with has a price formula they apply. Especially for smallish quantities (under a thousand pieces), they probably won't be open to negociation. You can however get online quotes with many of them, entering the number of layers, holes, hole sizes, trace & space size, size, qty, etc., and see how varying the different things will affect the price. Even if they won't give you the exact forumula they use, you can kind of do a curve fit after you get enough data points, and figure out how much each of the various things cost on a per-piece basis and how much the hidden set-up fee may be (in addition to explicit tooling charges).

If you're doing your own design, I strongly recommend showing it on the forum to run past others before getting it made, since fast processors and other ICs ("fast" referring especially to rise times, even if you run it at a slow clock rate) can easily be disabled by stray inductance. Even after being in this stuff for decades, I'm still becoming more and more impressed by the effects. And again, refer back to the links in my first post on page 1 of this topic.


Thanks for the input - I was thinking about the whole inductance issue and am hoping it won't be as big an issue if I leave the designing to them. After they design a preliminary model, I'll run it by you guys before proceeding. Hopefully it'll work out well. :)


Mon Jul 07, 2014 8:52 pm
Profile
User avatar

Joined: Tue Jan 15, 2013 5:43 am
Posts: 189
Quote:
68060-to-breadboard adapters
Good idea -- inevitable, in fact. (You'll need something!) And the same PCB could also serve as a 68060-to-wirewrap adapter.

No doubt you've realized the adapter is necessary simply to physically spread out a dense concentration of pins into a not-so-dense concentration of pins. :) But its other crucial function is to get enough bypass capacitors intimately connected to the chip! So you want to lay out the caps and the supply traces first, then the signal traces after. You might need to start over a few times, but the caps and the supply traces always come first.

For PCB fab, check out https://oshpark.com. Haven't used their service (yet) but it sure seems well suited to hobbyists. AND they have free, open source design tools (KiCAD) and an online community to offer support & advice.

BTW I'm not a prospective buyer for your PCB -- wish I had the time to play with one! But the oshpark prices make it easy to go it alone.

-- Jeff

Quote:
if I leave the designing to them
ps- oops, your post hadn't yet appeared when I was typing mine. But maybe someone in the KiCAD community would be able to manage the preliminary design.

_________________
http://LaughtonElectronics.com


Mon Jul 07, 2014 9:12 pm
Profile WWW

Joined: Tue Dec 11, 2012 8:03 am
Posts: 285
Location: California
Dr Jefyll wrote:
Quote:
68060-to-breadboard adapters
Good idea -- inevitable, in fact. (You'll need something!) And the same PCB could also serve as a 68060-to-wirewrap adapter

which should have an arrangement something like a PGA layout (not an ultra long DIP) to reduce connection length.

Quote:
But its other crucial function is to get enough bypass capacitors intimately connected to the chip! So you want to lay out the caps and the supply traces first, then the signal traces after. You might need to start over a few times, but the caps and the supply traces always come first

and, I would add, clock signals come next. Having a clean clock distribution is more critical than other buses. Keep the ground connections as short as you can. I wish we had not gotten stuck with 74-family power and ground pins being out at the corners, something that was standardized back before they anticipated speeds getting where they are today. Even on DIPs, they should have done what they do today on for example memory ICs that come in SOIC or SOJ packages, or on the PIC microcontrollers, which have power and ground right next to each other in the middle of each row, as close as possible to the die, like pins 7 & 8 and pins 21 & 22 of a 28-pin package.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources


Mon Jul 07, 2014 9:42 pm
Profile WWW
User avatar

Joined: Tue Jan 15, 2013 5:43 am
Posts: 189
Another thought: maybe you should solder the '060 directly to the adapter board (thus saving the cost of an expensive socket). Then the entire CPU-and-pcb assembly could be plugged in to whatever. Here's a rough layout:
Attachment:
060 adapter.gif
060 adapter.gif [ 4.87 KiB | Viewed 10818 times ]

Along the edges I've got double rows of connections to keep the width from getting ridiculous. For use with a <shudder> breadboard, you'd populate one row with male pins on the bottom and the other row with a female header on top. For wirewrap use, there'd be just a double row of male pins on the bottom. In that case the assembly could engage a comparatively inexpensive female connector like this Samtec SSQ-150-03-G-D (about $17 at DigiKey).
Attachment:
Samtec SSQ-150-03-G-D (cut in half).JPG
Samtec SSQ-150-03-G-D (cut in half).JPG [ 16.98 KiB | Viewed 10818 times ]

This would be challenging to build -- I'm not pretending otherwise. But you seem committed, and anyway is there an easier alternative? I'd be interested to see other ideas.

-- Jeff

_________________
http://LaughtonElectronics.com


Mon Jul 07, 2014 10:41 pm
Profile WWW
Display posts from previous:  Sort by  
Reply to topic   [ 266 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6, 7, 8 ... 18  Next

Who is online

Users browsing this forum: No registered users and 8 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group
Designed by ST Software