Jump to content
Shift838

838 - IO Card (Flex Interface card replacement!)

Recommended Posts

I can't wait to get my new peb ribbon.

Long gone are the bulky firehose days of yesterday. But I'll keep it just in case this new one dies on me. Just saying.

Newtons fault

  • Like 1

Share this post


Link to post
Share on other sites
19 minutes ago, Vorticon said:

Agree. I actually keep 2 RS232's in my system alongside the TIPI, one for parallel port interfacing and comms, and the other for printing to a dot-matrix printer and HDX connectivity. I did however get rid of my ramdisk as it seemed redundant. I assume you keep yours because it's faster in access than the TIPI?

The ramdisk is faster in some respects but I keep it for many of the same reasons I have other hardware - I like to use and 'play' and experiment with the variety :)  

  • Like 4

Share this post


Link to post
Share on other sites
1 hour ago, InfiniteTape said:

Any characterization on the length of the ribbon?

 

1 hour ago, Shift838 said:

so far I have tested it with a 4 foot cable with no issues.  planning on testing a 6 foot next few days, but that is as long as I will test.

 

I have tested with this one:

 

Amazon.com: Antrader 10Ft/3M IDC 40Pin 1.27mm Rainbow Color Flat Ribbon Wire Cable for 2.54mm Connectors: Computers & Accessories

 

 

 

Propagation delay in copper is about 5ns per meter (wire gauge affects this, so you have to look it up / calculate based on the wire being used).  However, you will probably run into noise problems before you hit a length problem.

 

The PEB cable is basically an extension of the CPU bus and runs at ~3MHz, so having a ground conductor between every signal conductor would be highly recommended (IDE cables do this, for example).  Also, the Fire-Hose cable is probably shielded, which is also a good thing for reliability.

 

There is no error checking in the 99/4A, so you will only find out about the reliability problems via random problems and/or corrupt files, etc.

Share this post


Link to post
Share on other sites

There are ways to reduce interference besides shielding.  A bank of capacitors across the data lines, ferrite beads, extra ground lines (like an IDE cable uses,) are a few examples.

 

  • Like 1

Share this post


Link to post
Share on other sites
6 hours ago, atrax27407 said:

The firehose cable IS shielded - hence the layer of foil.

Yes, but this new design does not have shielding.  I was just offering information for consideration.

 

2 hours ago, Pheonix said:

There are ways to reduce interference besides shielding.  A bank of capacitors across the data lines, ferrite beads, extra ground lines (like an IDE cable uses,) are a few examples.

And each one has electrical and physical trade-offs and considerations to account for, and there are no silver bullets.  Adding capacitance to your high-speed data lines can put you into a world of hurt really fast, and IDE cables are designed for short distances and internal use.  The fire-hose is external and typically much longer, so a lot more opportunity to introduce problems.

 

RF, grounding, and shielding is non-trivial.  Hopefully this design will "just work" and most of these concerns can be ignored.

  • Like 1

Share this post


Link to post
Share on other sites

I would assume that the shielding of the firehose cable is not primarily there for functional considerations, but for the FCC's EMI requirements.  After all, we're not talking excessive lengths or high speeds.  Long, unshielded parallel connections were quite common back in the day in commercial (i.e. non-residential) environments were EMI regulations were less strict.

  • Like 1

Share this post


Link to post
Share on other sites

All methods of noise reduction are a trade off of some sort.  Ferrite beads add resistance to the line.  Capacitor filtering adds, well, capacitance to the line.  Foil shielding adds physical restraints to the line.  Extra ground lines also add physical restraints.  They can all add extra latency.  As for length, a properly shielded & filtered IDE (or SCSI for that matter,) cable can reach up to 10' before a repeater point of some sort is needed.  In the case of SCSI, just having a drive somewhere along the length serves this purpose.  Hopefully, as Hans23 pointed out, the age of the system will allow for more flexibility in this.  Many of the designs in the late 70's & 80's showed the US's paranoia where EMI is concerned.  Take the metal "tail" on the C64's dattasette for example.  Only present on US models, and completely unnecessary.  But, required by regulation back then.  If someone is worried about signal loss, with modern components, an error correcting encoder/decoder circuit could be added to each end without taking up much space at all.  But that's well beyond my capabilities.

  • Like 2

Share this post


Link to post
Share on other sites
16 hours ago, matthew180 said:

Also, the idea of "speed" is the wrong mental model for what is happening.  Chips all operate at the same "speed", i.e. electron flow which is roughly equivalent to the speed of light. 

Er ... no. You may want to talk to those guys with the accelerator rings like LHC if you want to have particles like electrons travel near light speed.

 

In fact, electrons travel at a surprisingly slow pace, some millimeters per second. If you take alternating current, most of the electrons in the wires in your light switch never get the chance to leave the switch.

 

But the electric field propagates at the speed of light, which makes the electrons move along the full extent of the wire. This is what makes circuits switch, light bulbs glow, motors turn.

 

 

  • Like 3

Share this post


Link to post
Share on other sites
58 minutes ago, mizapf said:

Er ... no. You may want to talk to those guys with the accelerator rings like LHC if you want to have particles like electrons travel near light speed.

 

In fact, electrons travel at a surprisingly slow pace, some millimeters per second. If you take alternating current, most of the electrons in the wires in your light switch never get the chance to leave the switch.

 

But the electric field propagates at the speed of light, which makes the electrons move along the full extent of the wire. This is what makes circuits switch, light bulbs glow, motors turn.

 

 

That is actually very interesting. Did not know that.

  • Like 1

Share this post


Link to post
Share on other sites

You can most easily explain this with your garden hose. When it is completely empty, it takes some time until the water squirts out. However, when it is full, and you turn on the water, it instantly comes out of the nozzle. In that sense, electric conductors are filled with movable electrons.

 

See https://en.wikipedia.org/wiki/Electric_current#Drift_speed

  • Like 4

Share this post


Link to post
Share on other sites

"When it is completely empty, it takes some time until the water squirts out"..

 

And always seems to happen when I'm looking into the hose wondering where's the water. 

Damn wet electrons.

  • Haha 5

Share this post


Link to post
Share on other sites
17 hours ago, InsaneMultitasker said:

The ramdisk is faster in some respects but I keep it for many of the same reasons I have other hardware - I like to use and 'play' and experiment with the variety :)  

Yes, the HRD 4000 is a fantastic device and by no means obsolete because of TIPI. Just different use cases.

 

  • Like 3

Share this post


Link to post
Share on other sites
4 hours ago, mizapf said:

Er ... no. You may want to talk to those guys with the accelerator rings like LHC if you want to have particles like electrons travel near light speed.

 

In fact, electrons travel at a surprisingly slow pace, some millimeters per second. If you take alternating current, most of the electrons in the wires in your light switch never get the chance to leave the switch.

 

But the electric field propagates at the speed of light, which makes the electrons move along the full extent of the wire. This is what makes circuits switch, light bulbs glow, motors turn.

 

 

Sure, I understand that, but I was trying to stay on topic (answer a question) and provide a simplification to help clarify an idea about propagation delay, rather than getting into hole vs. electron flow through a PN junction, or current vs. wave E and M field theory, etc.

 

For your sake I will go make a correction, since, as you pointed out, it is not the actual electrons flowing from one end to the other that carries the information.  Is that satisfactory?

  • Like 2

Share this post


Link to post
Share on other sites

Please, no big issue. :) I just got alerted by that one statement (and I know that many people indeed believe it).

 

 

  • Like 1

Share this post


Link to post
Share on other sites
On 8/13/2021 at 12:26 AM, Shift838 said:

The issue I have found is that if the PEB is not powered up and when the 99/4A is powered up the LED on the PEB card lights up.  It's odd as there is no VCC lines passing to the side port card from the 99/4A console.  But of course the LEDs work as expected when both the PEB and console are powered up.  As I said it does not affect the functionality of the card.  Once I figure that out then it will be perfect and after some more testing be ready for the wild.  If anyone has any ideas on the issue please comment and I can check it out.  

 

I have been able to run down my LED issue described above.  I should have tested this first!

 

The schematic has a LS244 controlling the LED via the DBIN signal along with the 2n3904 transistor.  Well looks like I had a bad LS244!  Replaced it and BAM, no more LED on the PEB card when the 99/4a is powered on by itself.

 

Now for some more testing over the next week or so.

 

 

  • Like 8

Share this post


Link to post
Share on other sites

I came to a conclusion over this weekend after a some comments around HCT vs LS chips.  The naming of the board as Turbo Interface was to take advantage only if the user had installed the 32k ram internally on the 4A's 16 bit bus or if the user has performed a crystal upgrade.  Which both of these would allow (according to everything I have read) the data to transfer faster between the PEB bus and the 99/4A bus.  But after some thought,  how many of us actually have done either of those upgrades or could benefit from it.

 

I have decided to rename the card from 838-Turbo Interface to something more adequate like 838 IO Interface.

 

So if this project makes it to production please be aware the name will change.

  • Like 1
  • Thanks 2

Share this post


Link to post
Share on other sites

Moving the 32K into the console does not make anything faster as far as the CPU bus cycle time is concerned.  It might affect the overall performance of the *computer* if 32K console mod avoids the 16-to-8 bit multiplexer, but I have not looked at the mod in detail so I'm not sure.

 

However, changing the crystal in the system, which affects the CPU's clock, that *does* impact the bus cycle time and devices on the bus will need to respond in less time, or introduce less propagation delay.

 

How many people have done those mods is anyone's guess.  It is probably best to make your hardware projects so they work with a stock console, IMO.

  • Like 4

Share this post


Link to post
Share on other sites
8 hours ago, matthew180 said:

Moving the 32K into the console does not make anything faster as far as the CPU bus cycle time is concerned.  It might affect the overall performance of the *computer* if 32K console mod avoids the 16-to-8 bit multiplexer, but I have not looked at the mod in detail so I'm not sure.

Yeah, there are multiple mods, and at least one of them puts the RAM on the 16-bit bus and disables the multiplexer. I had it in my console back in the day, but for my newer mod I opted for simpler over faster. ;)

 

  • Like 3

Share this post


Link to post
Share on other sites

I got around to finishing up a case for the 838-IO Sideport card that connects to the 838-IO PEB card.  After a few revisions, it all fits good.

 

the silver top slides in nicely to the black case via some channel cuts.

 

the total time to print both pieces is about 6 hours.

 

the PCB in the case is from the original test run and has the old logo and name on it, the new PCBs will only be simply called 838-IO.

 

I opted to bolt the PCB directly through the case for stability.

 

 

838-io-spc1.jpg

IMG-2783.jpg

IMG-2777.jpg

IMG-2789.jpg

IMG-2787.jpg

IMG-2786.jpg

IMG-2784.jpg

IMG-2788.jpg

  • Like 5
  • Thanks 1

Share this post


Link to post
Share on other sites
9 hours ago, Vorticon said:

No passthrough port for the speech synth?

no, the 838-IO is designed after the original flex cable interface card.  It is designed to be the last device on the 99/4A bus to the PEB.  So the 838-IO will either plug into the side port of the 99/4a itself or a different side port expansion car, like the speech synthesizer, but as said it must be the last one in the chain. 

 

The speech requires power to be passed from the 4A, the flex cable side port card (original)as well as the 838-IO side port card does not pass power to anything, both get their power from the PEB via the cable hooked up between the two.

 

 

Edited by Shift838
  • Like 3

Share this post


Link to post
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...