H19/Z19 Questions & Project Status

[Alexander 'Z' Pierson]

Good morning folks,

I've been tinkering with my H19 standalone project, with varying degrees of success.  Considering it's cobbled together with spare parts from ebay I'm actually really surprised it got this far.  It sat gathering dust for a month or two, and when I re-assembled it, I started encountering difficulties.  Then again, that's to be expected considering this is just a terminal logic board and a keyboard.  I've got a +/- 15V and a +9V power supply to simulate the +/-16 and +8.5V transformer taps respectively.  You're seeing the video via a composite combiner circuit.  It worked last time I fired it up (pictured below), but today I experienced some strange failures...

U401, VCC-1's 7805 regulator gets very hot within the span of about 20 seconds. I've also noted that the V-sync signal doesn't seem to be playing nice, but I have a hunch that fixing the power malfunctions may fix the sync too.    Here are my questions:

1: Does anyone know how many variations of schematic/board layout were made for the H19/Z19 were made?  I seem to have two very different layouts (the right pictured below one came directly out of my H89), each with a few minor differences.

2: Does the U405 7905 -5V regulator do anything on an H19 board?  I can't seem to find anywhere that it actually connects on my schematics, nor can I fathom anything that would need -5 volts on the terminal logic board itself...

3: Has anyone else done a full power supply replacement on an H89 or H19?  I want to if someone has tread this path before and might have advice to share.

Any advice you could provide would be greatly appreciated.

Thank you,

-Alexander 'Z' Pierson

[Glenn Roberts]

I have not replaced an H19 power supply but I have replaced my H8 power supply with a PC Switching supply.  It runs great.  Very cool running.  In fact I’m sure my H8 would run just fine without the ventilation slots in the cover (or with a bunch of stuff piled on top).  It’s never warm.

 

Since I had to rob a CRT from one of my H19’s I’ve got this notion that some day I’ll install an LED screen in the opening and hook up a smart board, say maybe a Raspberry Pi, to the actual H19 keyboard, replacing all the original Heathkit “guts”.  Then run something like George’s H19 emulator software (or even Mark’s H89 emulator!).  The result would be something that looks and feels just like an H19/H89 but with modern electronics and screen.  You could perhaps even just use it as a Unix workstation or do all kinds of stuff a real ’19 never could!

 

Somewhere on my Heathkit “bucket list”…

 

-          Glenn

--
You received this message because you are subscribed to the Google Groups "SEBHC" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.
To post to this group, send email to se...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/bd671946-7ae4-4f81-a6a8-3d4870e4aac6%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Norberto Collado]

Perhaps the attached file could help!

Norberto

-------- Original Message --------
Subject: [sebhc] H19/Z19 Questions & Project Status
From: "'Alexander 'Z' Pierson' via SEBHC" <se...@googlegroups.com>
Date: Wed, January 27, 2016 3:34 pm
To: SEBHC <se...@googlegroups.com>

Good morning folks,

I've been tinkering with my H19 standalone project, with varying degrees of success.  Considering it's cobbled together with spare parts from ebay I'm actually really surprised it got this far.  It sat gathering dust for a month or two, and when I re-assembled it, I started encountering difficulties.  Then again, that's to be expected considering this is just a terminal logic board and a keyboard.  I've got a +/- 15V and a +9V power supply to simulate the +/-16 and +8.5V transformer taps respectively.  You're seeing the video via a composite combiner circuit.  It worked last time I fired it up (pictured below), but today I experienced some strange failures...

U401, VCC-1's 7805 regulator gets very hot within the span of about 20 seconds. I've also noted that the V-sync signal doesn't seem to be playing nice, but I have a hunch that fixing the power malfunctions may fix the sync too.    Here are my questions:

1: Does anyone know how many variations of schematic/board layout were made for the H19/Z19 were made?  I seem to have two very different layouts (the right pictured below one came directly out of my H89), each with a few minor differences.

2: Does the U405 7905 -5V regulator do anything on an H19 board?  I can't seem to find anywhere that it actually connects on my schematics, nor can I fathom anything that would need -5 volts on the terminal logic board itself...

3: Has anyone else done a full power supply replacement on an H89 or H19?  I want to if someone has tread this path before and might have advice to share.

Any advice you could provide would be greatly appreciated.

Thank you,

-Alexander 'Z' Pierson

[Norberto Collado]

Does the U405 7905 -5V regulator do anything on an H19 board?

It is there for any ROM that supports such voltage. See attached circuit.

[Lee Hart]

From: 'Alexander 'Z' Pierson' via SEBHC <se...@googlegroups.com>

>I've been tinkering with my H19 standalone project...
>U401, VCC-1's 7805 regulator gets very hot...

See if that VCC line is shorted to ground. Check the little tantalum capacitor just below U401; they occasionally fail shorted. If not that, an IC somewhere may have died and failed shorted. Luckily, Heath socketed all the ICs. :-)

>1: Does anyone know how many variations of schematic/board layout

> were made for the H19/Z19?

There are 4 different boards that I know of. Look for the 85-xxxx part number silkscreened on the board:

85-2650-1 late model (H19A); the left board in your photo
85-2376-1 or -2 early (H19)
85-2237-2 early (H19)
85-2166-2 earliest (H19)

The circuit is basically the same throughout. But the late model H19A board was re-laid out to reduce EMI. To confuse things further, they also changed the power connector from 10 to 11 pins, and re-assigned the component designators between the early/late models.

>2: Does the U405 7905 -5V regulator do anything on an H19 board?

Early boards used EPROMs that needed the -5v, +12v, and +5v supplies. Later boards replaced them with ROMs that only needed a +5 supply. Heath kept the -5v supply and jumper options in place so customers could still use these early EPROMs if desired.

>3: Has anyone else done a full power supply replacement on an H89 or H19?

I've upgraded H19's into H89's (Heath had a kit for it). This included replacing the power supply. I've also put an H19 board in a separate PacTec case, and powered it with a +5v/+12v/-12v linear "brick" supply left over from an early computer (Coleco, I think). I bypassed the on-board regulators and used the ones already in the power supply. The H19 board needs less than an amp at 5v, and very little at +12/-12v.

--
Excellence does not require perfection. -- Henry James
--
Lee A. Hart http://www.sunrise-ev.com/controllers.htm now includes the GE EV-1

[Alexander 'Z' Pierson]

Thank you for all the responses. 

Glenn: You're thinking on a similar wavelength as I am.  Imagine a time comes when an H19 or H89 is functional, save for the CRT.  No CRT replacements around?  The ability to remove the video board and CRT, throw in an LCD display, and have the terminal logic board output composite video to the new LCD.  A day may come when such a repair becomes more commonplace, for Heathkits and many other machine. 

Norberto: That document actually does give me some ideas, if my current power supply configuration fails.  I also followed a trace from the -5V line, and as you suggested, it ended up leading to the character decode ROM U445.

Lee: That was my first thought.  I replaced most of the tantalums and electrolytics soon after I acquired this board.  I'm going to replace the pair next to U401 again, just to be certain.  I've also pulled the more precious large IC's to avoid damaging them during my next power-on test following the capacitor replacement.

I checked, and my H89's internal board is a 85-2376-1, but we're focusing on the left board in the photo, the 85-2650-1.

If my current external power supplies don't work out, I'll explore the route of removing the offending 78XX/79XX series regulators, and installing either an ATX or other sort of brick power supply to feed +/-12V and +5 directly to the terminal logic board. In my searches, I found an image of someone else's similar job: http://www.physics.purdue.edu/~jones105/pdp-11/images/IMG_2917.JPG

In the mean time, I'm going to hunt for any short circuits.  Thank you for the help, I'll provide more information soon.

-Alexander 'Z' Pierson

[Alexander 'Z' Pierson]

Update time! Alot has happened...

I asked the folks over at the VCForums about common deployments of the MC6845 CRT controller since it's in a whole slew of different computers, not just our beloved H19/Z19's.  I figured they might know of good NTSC configurations involving that chip.  Someone was nice enough to walk through a bit of the math to get timing register values that should work based on the H19's CRT controller clock speed.  They would make a more kosher composite signal, when compared to the default from H19 mask ROMs.  That means making custom ROMs... I didn't have the ability to dump or burn EPROMs at the time, so I made efforts to fix that.

So I learned how to burn ROMs with a few tips from Lee (thank you again!).  I made backups of all the ROMs in my H19, and compared them to the one's on Les's site.  Seems reading 8332's is a pain, when you can't find the correct datasheet for their unique pin configurations.  Once I dug up the right datasheets, I was able to dump the mask ROM and work from there.  I rooted around through the ROM data, and I created a version with the new CRTC timing values.  Since I don't know where to find an appropriate EA8332 pin compatable EPROM, I burned a pair of appropriate 2716 EPROMs -- one for the upper 2K and one for the lower 2K.  My board is a late revision, so it's configured to use just one ROM socket, with a EA8332, and the jumper rows are hard-wired in the traces.  I reconfigured my board to use both ROM sockets again, and allow for changes in the jumpers just like an early revision.

While I was at it, I eliminated 4 of the 5 linear regulators, and just give this project a modern power supply.  The only remaining regulator is the 7905 for the -5V rail.  So I installed jumpers to bypass the now missing linear regulators, then carted my gear to a local hackerspace to test stuff with their complex bench power supplies that could handle the +5V, +12, and -12V's that I needed to run this monster until I have something permanently built for the task.  When I fired it up, I was getting a MASSIVE draw on the 5V rail -- over 2 amps, which is excessive.  The +12V and -12V rails current draws were about what I expected, so no worries there.  The board was clearly not working, so I decided to come back to the project with my head fresh.

After a two week break, I tried again.  With the help of some tools on loan, namely a small bench power supply and a FLIR attachment for a smartphone, I hunted down the source of the 5V rail shorts.  I took out ALL of the IC's, then looked for shorts in the board traces while the board was powered.  After that, I put chips back in groups of 3 or 4, starting with the largest ones.  I kept track of the amperage for every batch that I added, and took FLIR photographs to see if something was getting hotter than the other components.  Turn off the power, plug in a few chips, turn it back on, record values, repeat. 

Sure enough, towards the end when I snapped in a 2114 RAM chip, the amperage went up dramatically, and it put off significantly more heat than the other components.  After that, two 7400 series chips also had the same symptoms.  Seems I had three bad IC's total: a 7474, a 7486, and that 2114.  Fortunately, I have plenty of 7400 series stuff on hand, so those were easy to replace. The 2114? Not so much. Already ordered a spare on ebay.  When it arrives, tests will resume.

-Alexander 'Z' Pierson

[Alexander 'Z' Pierson]

Greetings, folks!

Good news,my power bypass modification appears to be working!

Would anyone happen to have a photo of the ROM jumper section of an early revision H19 board, preferably with two 8316 mask ROMs?  I'm getting some strange results, and I get the feeling I'm not setting jumpers correctly.  I wanted to have something to compare against, since all the reference photos I have are of middle or late era boards.

Thanks,
-Alexander 'Z' Pierson

[Paul Cushing]

Alexander, would you be so kind as to share the datasheet(s) for the mask programmed 8332.  All I have is the pin-out found in  the H-19A Operations Manual

Thank You

Paul Cushing

[Lee Hart]

Great work, Alexander. That's some serious hacking and debugging you've done!

Do you need the -5v regulator for anything? If you're using standard 5v-only 2716 EPROMs, is there anything on the board that needs -5v?

Using FLIR is pretty cool. I've always had to use my finger.

If you replace the 2114 RAMs with 2148 (same pinout, but faster), you can also install my "flicker free" mod. It eliminates the "black flecks" that appear when the Z80 writes to the screen. No extra parts needed; just a couple patch wires, and also speeds up the Z80 from 2 to 3 MHz.

Virus-free.www.avast.com

--

You received this message because you are subscribed to the Google Groups "SEBHC" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/af098c03-a991-413f-912c-ec0ff29d71f3n%40googlegroups.com.