H17 Refurb progress

286 views
Skip to first unread message

Tim R

unread,
Nov 25, 2024, 10:25:57 PM11/25/24
to SEBHC
Hi all,

With the H8 computer and the H9 terminal up and running, I've now turned my attention to the H17 dual disk drive. I've started disassembly with the intent of cleaning and inspecting the unit. I've already found a bit wrong. 

The first thing I found was a broken fuse holder. A previous owner has used an alligator clip lead to bypass the fuse! I replaced the fuse holder, and after a quick inspection of the power supply board (and unplugging both drives), I decided to do a slow voltage ramp up and see how the PSU section would do. 

All was well for awhile, and by 80VAC input to the PSU, I was already in the proper output range on both +5VDC and +12VDC outputs. As I continued to increase voltage, one of the 2.2uF tantalum caps shorted, and I saw a tiny puff of smoke evacuate out of a tiny hole in the cap (not super spectacular as far as tantalum failures go). I was monitoring the output voltages, and visually watching the PSU board as I was ramping up the AC. Fortunately, I cut power immediately. I was concerned that I might have fried the rectifier diodes, but they test ok. I haven't checked to two transistors yet, fingers crossed on that. I've pulled the board, and have replaced the offending cap (I'm really cursed as far as tantalum caps go).

Also, once I was able to inspect the two large electrolytic caps closer, I noticed that the paper side on one of them looked discolored and appears to have leaked electrolyte. I went ahead and pulled all three electrolytic caps, and the leaky one measured completely out of spec, and the other two had fairly high ESR measurements. So, I guess I'll go ahead and replace them all.

Overall, this H17 unit is not nearly a clean as the H8 and H9 were. I looks like it has been stored in a very dusty environment at the least, or a hot attic at worse. I'm hopeful that the drives are in better shape than the power supply section.

Thats all that's fit to report for now.

Tim

glenn.f...@gmail.com

unread,
Nov 26, 2024, 8:53:07 AM11/26/24
to se...@googlegroups.com

Thanks for the update Tim. Looking forward to hearing more about the condition of those drives!  If my experience is any indication you should be able to get them running…

--
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 visit https://groups.google.com/d/msgid/sebhc/f03b86f0-af32-4de6-a68d-37d200da1fe7n%40googlegroups.com.

Tim R

unread,
Nov 26, 2024, 8:57:23 PM11/26/24
to SEBHC
Power supply done! Started disassembly on second drive. Some of the bearings are frozen/sticky (hub mostly). Started soaking those in oil bath. The motor and worm gear are nice and free moving. While waiting on bearings to soak, I’ll install the controller card in the H8 and run through the memory tests. So far so good I think.

Joseph Travis

unread,
Nov 26, 2024, 9:24:00 PM11/26/24
to se...@googlegroups.com

Tim,

Before you apply power to the drives, you should check / replace the small filter capacitor on the motor speed controller board +12vdc power line.  It is a common point of failure that typically shorts out and causes the inline inductor to burn up.  Good luck!

Joe


--
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.

glenn.f...@gmail.com

unread,
Nov 26, 2024, 9:45:52 PM11/26/24
to se...@googlegroups.com

I think Tim has Wangco 82s?  they don’t have a separate motor control board like the Siemens…  ??

Tim R

unread,
Nov 26, 2024, 11:39:34 PM11/26/24
to SEBHC
Yes, I do have the Wangco 82s. It has a single board, and thankfully no tantalum filter caps.

Tim R

unread,
Nov 27, 2024, 8:05:18 PM11/27/24
to SEBHC
Hi all,

I have a question on the Wangco 82 drives that I hope someone will know the answer to. I have part of the answer from the H17 operations manual, but I'm not able to locate a copy of the illustration guide that the manual frequently refers to.  The question I have is about the configuration for SYS0 and SYS1. 

The manual refers to cutting traces DS1 and DS2 for SYS0. The drives I have use a dip-switch block rather than cutting traces, so I have DS1 and DS2 set open for SYS0 which should be correct.  For SYS1, I have DS1 and DS3 open.  The "MUX" trace is also cut on each drive as per the manual.  

Here is the question that I can't find an answer for in the manual: What are settings T1-T5 for? For SYS 0, T1-T4 are set closed, and T5 is open.  For SYS1, T1-T5 are all closed. I'm not sure if these settings are correct, but I don't see them referenced in any docs I have.

Thanks in advance for any info you can provide.

Tim

Jumper block for SYS1 pictured below...
Wangco82.jpg

glenn.f...@gmail.com

unread,
Nov 27, 2024, 8:36:57 PM11/27/24
to se...@googlegroups.com

Huh.  You’ve got an oldie!  I’ve never seen a Wangco with those labels (later ones have more descriptive labels)

 

Below is the appropriate picture from the illustration guide.  Heath did things “backwards”: DS3 is actually SY0: so you want DS3 connected and DS1 and DS2 open (assuming you’re programming SY0).  From the picture it looks like the other switches (T1..T5) should all be closed as well.  presumably the MUX foil cut has already been made on your drive since it was previously in use. 

 

Easiest way to test it is to set your PC to 030.000 and hit RUN. The drive head should load (solenoid click) and the drive should spin

 

 

From: se...@googlegroups.com <se...@googlegroups.com> On Behalf Of Tim R
Sent: Wednesday, November 27, 2024 8:05 PM
To: SEBHC <se...@googlegroups.com>
Subject: Re: [sebhc] Re: H17 Refurb progress

 

Hi all,

--
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.

image001.png
image002.jpg

glenn.f...@gmail.com

unread,
Nov 27, 2024, 8:46:22 PM11/27/24
to se...@googlegroups.com

Note: that’s for single drive configuration only (where I recommend you start).  Jumpers T1..T5 should only be closed for the *last* drive in the chain; open for others.  An odd way to do things (later versions used a removable resistor pack).  So if you added a second drive you’d open T1..T5 on SY0 and keep them closed on SY1, then close DS2 on SY1 and leave DS1 and DS3 open…  hope that makes sense?

 

I have the manual (and schematic) for this drive and can scan it if you need…

image001.png
image002.jpg

Tim R

unread,
Nov 30, 2024, 9:27:05 PM11/30/24
to SEBHC
Well, things could be going better.  I knew I had a challenge ahead, but this has been a disappointing journey so far with the H17. When I acquired this H8 System, I was warned that it was what was left over after the previous owner assembled their own system. No problem, challenge accepted. Without too much angst, I (with help from all of you) was able to get the H8 computer, and the H9 Terminal back to operating condition. They are actually working well, and I used them today to play Space War when I needed a break from the rather frustrating H17. 

Now on to the H17. As I mentioned in a previous post, the H17 chassis power supply was dead, and the broken fuse holder had been jumped with a clip lead (always a bad idea). So I replaced the fuse holder, two tantalum caps, and one bad electrolytic.  The power supply is perfect now 12.01VDC and 5.08VDC, nothing to complain about there! 

Now on to the drives. Both drives were pretty much mechanically frozen. I disassembled those, cleaned them up, soaked and cleaned the bearings, worm gear, and motor bearings.  They spin smooth as silk, no problem there.  One thing I usually do on any circuit boards where I see tantalum caps is to test for shorts at the power rails (12V and 5V).  These drive boards don't use the little bead tantalums, but instead use the small black cylindrical ones (tubular tantalums).  Sure enough, one of the boards had a short on the 12VDC side due to one of these. I have two drives, so I set that one aside for now.

Time to check out the H17 controller board. I started by installing it in the H8, and running the H8 memory test from the H17's ROM. Worked perfect, no errors. Then I ran the H17 controller RAM test.  Again, everything worked fine. Next I connected the good drive to the H8 and did the drive interface test from the manual and again all worked perfectly. I could start and stop the drive motor, and engage the head solenoid. Next I decided to load and run the speed test routine in the manual, which worked perfectly with a reading of 172. No speed adjustment required.

Now here is where the problems began. I jumped the DS1 on the drive, and set the drive termination (this is required to do the diagnostic tests). I entered the test instructions, and instead of getting the expected output, 037 127 and a beep, I got a single very bright "6" when I then did a reset (don't like overdriving the rare 7 segment displays).  So I'm thinking I might have a ROM issue, but I disregard that idea as the memory tests are on that ROM too, and those worked ok. After multiple tries to complete the test sequence, I get the same result. At this point, I decided to config the drive as SYS0, put in a blank disk, set PC register to 030 000 and hit GO.  The drive indicator came on, and the drive spun up. Then more random bright random characters on the H8 display... RESET!

I decide at this point to disconnect the drive, and rerun the memory tests. This time the tests in the H17 ROM didn't work. I double checked my work, and the test routines that had run perfectly an hour earlier didn't run. I set the MEM pointer to the routine addresses, and sure enough 377 000 377 000 377 000......  I pulled the H17 controller at this point, tested all ICs in the chip tester and reseated them, tested the two 2114 RAM chips, and all was well.  Double checked the voltages on the board, and they were fine. My conclusion is that the ROM (444-19) on the controller died in the middle of testing. 

Any other ideas? I think the drive is fine, and the controller worked for awhile. 

If you made it to the end of this book, thanks for any ideas you have.

Tim

Joseph Travis

unread,
Nov 30, 2024, 9:47:29 PM11/30/24
to se...@googlegroups.com

Tim,

We appreciate the details you have provided however, it would be very helpful if we had closeup photos of your boards and which ones are installed in the chassis.

There are numerous configuration possibilities, not all are compatible.

Regards,
Joe


--
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.

glenn.f...@gmail.com

unread,
Nov 30, 2024, 9:53:17 PM11/30/24
to se...@googlegroups.com

So when you say “jumped the DS1 on the drive and set the drive termination” please clarify. First of all, where in the instructions is that test described?  And how did you “set the drive termination” since you have the older Wangco that doesn’t use terminator resistor packs but has those T1..T5 jumper. 

 

Just trying to follow along in the manual…

 

 


Sent: Saturday, November 30, 2024 9:27 PM
To: SEBHC <se...@googlegroups.com>

--

glenn.f...@gmail.com

unread,
Nov 30, 2024, 9:57:55 PM11/30/24
to se...@googlegroups.com

You say you tried to boot from a “blank disk”.  One of the ones I sent you? May or may not have been blank. If you try to boot from it who knows what random code might be read in and executed. That would explain the odd behavior on the front panel.  I also sent you some formatted bootable disks. Did you try to boot any of those?

 

Seems unlikely that the ROM would fail, especially in the middle of the test…

 

 

 

From: se...@googlegroups.com <se...@googlegroups.com> On Behalf Of Tim R
Sent: Saturday, November 30, 2024 9:27 PM
To: SEBHC <se...@googlegroups.com>
Subject: [sebhc] Re: H17 Refurb progress Ugh!

 

Well, things could be going better.  I knew I had a challenge ahead, but this has been a disappointing journey so far with the H17. When I acquired this H8 System, I was warned that it was what was left over after the previous owner assembled their own system. No problem, challenge accepted. Without too much angst, I (with help from all of you) was able to get the H8 computer, and the H9 Terminal back to operating condition. They are actually working well, and I used them today to play Space War when I needed a break from the rather frustrating H17. 

--

Tim R

unread,
Nov 30, 2024, 11:02:46 PM11/30/24
to SEBHC
Glenn,

When the test asked to insert a blank disk, I used one of the unlabeled ones.  I did not use any of the labeled disks. This was for the test on page 8, the SYS0 READ/WRITE test.  The drives I have, as seen in the previous photos, have a dip switch block in place of the jumper strip.  Where it says to solder bridge the DS1 jumper if it is cut, I closed the switch.  My Wangco drives do have a resistor pack next to the dip switch. With the T1-5 closed, the terminating resistor pack is enabled. This test is done before configuring the DS jumpers.

I though that a ROM failure would be unlikely as well, but I can no longer access any data on the ROM such as the memory test routines. I have checked all the other chips, and there are few passive components on the controller board.  It seems like I was almost to a working drive, but now neither the controller card or the drive are doing anything.  I will post photos tomorrow of both the drive and the controller tomorrow.

Tim

Tim R

unread,
Nov 30, 2024, 11:05:34 PM11/30/24
to SEBHC
Joe,

I'll get photos posted tomorrow (All boards). Hopefully I have not caused an issue (or damage) by combining incompatible components.  I think I should check the CPU ROM version as well.  

Tim

glenn.f...@gmail.com

unread,
Dec 1, 2024, 8:25:24 AM12/1/24
to se...@googlegroups.com

So since you have an ability to load H8T images, one experiment you could do is use BUG8 to dump the H17 ROM contents and see what’s there.

 

Load BUG8 and issue the range command:

 

30000-37377<space>

 

This will dump 2K bytes starting at 030.000 (H17 ROM).  Unfortunately, BUG8 only knows how to display in octal, decimal or ASCII (no hex!) so the default is octal.

 

If you use a PC/Laptop for the console you can capture the output.  Attached are my results.  You could save your results and do a file compare (or post them here and I’ll compare them).

 

For a little more ambitious approach you could write a little BASIC program to PEEK() the values and print them in HEX (I think you’d have to write a little routine to do the hex conversion as EXBASIC doesn’t seem to have one nor does it have any formatted PRINT commands.)  The binary version of the 444-19 ROM is on the SEBHC site:

https://sebhc.github.io/sebhc/software/Roms/H89Roms.zip

 

It may be worth checking that there are no mod’s to your H17 board.  are the capacitor and resistor in the upper right corner installed? If the resistor has been removed and the capacitor shorted then the board was modified for use with the Zero org board and XCON-8 (XCON-8 has its own copy of the H17 routines which it will put into RAM; the mod to the board disables the on-board ROM)…

 

It would also be good to know what ROM version is on your 8080 CPU board…

 

  • Glenn
teraterm.log

Tim R

unread,
Dec 1, 2024, 2:03:05 PM12/1/24
to SEBHC
Glenn and Joe,

I've attached photos of the H17 controller board and the PCB on the Wangco 82.  

Also, the system has the following cards installed:

8080 CPU card running 444-60 ROM
WH8-64 DYNAMIC RAM configured for 56K (8K-64K)
H17 with ROM 444-19
H8-5 Serial/Cassette with default I/O addressing

Glenn, I will run through the routine you suggest (it may be a day or two before I get to it).  

From what I've seen so far, it seems as if the H17 ROM isn't even there. If I go to  MEM 030 000 and step through memory, I just get a series of 000 377 000 377 000 etc, just as if the card was not installed.  It will be interesting to see what output comes from BUG8. 
IMG_4366 copy.jpgIMG_4368.jpg

Joseph Travis

unread,
Dec 1, 2024, 2:29:32 PM12/1/24
to se...@googlegroups.com

I'd like to see every board.


Joseph Travis

unread,
Dec 1, 2024, 2:41:51 PM12/1/24
to se...@googlegroups.com

It is said, "A picture is worth a thousand words".  However, a thousand words only represents what you see when looking at the picture.

glenn.f...@gmail.com

unread,
Dec 1, 2024, 2:55:08 PM12/1/24
to se...@googlegroups.com

So you say “I just get a series of 000 377 000 377 000 etc,”  not sure what you mean but that’s not normal.

 

If you hit

 

RST/0

MEM

0  3  0  0  0  0

 

You should see:

 

0  3  0    0  0  0    x  x  x

 

In the display (where “x  x  x” is whatever the contents are of the location 030.000).

 

Then each time you hit “+” you should see the contents of the next cell, so on my computer I see:

 

0  3  0    0  0  0    3  0  3

0  3  0    0  0  1    0  1  4

0  3  0    0  0  2    0  3  7

0  3  0    0  0  3    0  4  1

0  3  0    0  0  4    3  0  0

0  3  0    0  0  5    3  7  7

0  3  0    0  0  6    0  7  1

0  3  0    0  0  7    3  5  3

0  3  0    0  1  0    0  4  1

 

…. Etc

 

 

 

Are you saying that you get something like the following? :

 

0  3  0    0  0  0    0  0  0

0  3  0    0  0  1    3  7  7

0  3  0    0  0  2    0  0  0

0  3  0    0  0  3    3  7  7

 

Etc?

 

If the ROM were missing you would expect all zeros in every location (no 377s)

 

I’m perpetually suspicious of DRAM boards (perhaps unfairly) only because I’ve had a lot of trouble with them.  I know you say the lower 8K are disabled on your 64K board but if somehow that board were to show up in the lower 8K you could have a competition between the DRAM board and the H17 ROM to see who shows up in that space.  Wish you had another RAM board of some kind… oh well.

 

How did you test the DRAM board?  did you use the RAM test published in the H8 manual?.

 

Joe has requested more close-up pictures of your boards.  I second that, especially for the DRAM board…

 

  • Glenn

 

 

 

From: se...@googlegroups.com <se...@googlegroups.com> On Behalf Of Tim R
Sent: Sunday, December 1, 2024 2:03 PM
To: SEBHC <se...@googlegroups.com>
Subject: Re: [sebhc] Re: H17 Refurb progress Ugh!

 

Glenn and Joe,

 

I've attached photos of the H17 controller board and the PCB on the Wangco 82.  

 

Also, the system has the following cards installed:

 

8080 CPU card running 444-60 ROM

WH8-64 DYNAMIC RAM configured for 56K (8K-64K)

H17 with ROM 444-19

H8-5 Serial/Cassette with default I/O addressing

 

Glenn, I will run through the routine you suggest (it may be a day or two before I get to it).  

 

From what I've seen so far, it seems as if the H17 ROM isn't even there. If I go to  MEM 030 000 and step through memory, I just get a series of 000 377 000 377 000 etc, just as if the card was not installed.  It will be interesting to see what output comes from BUG8. 

image001.jpg
image002.jpg

dwight

unread,
Dec 2, 2024, 2:37:33 PM12/2/24
to se...@googlegroups.com
That might be a typical uninitialized value, depending on the part design. In other words "no data".
Dwight


From: se...@googlegroups.com <se...@googlegroups.com> on behalf of glenn.f...@gmail.com <glenn.f...@gmail.com>
Sent: Sunday, December 1, 2024 11:54 AM
To: se...@googlegroups.com <se...@googlegroups.com>
Subject: RE: [sebhc] Re: H17 Refurb progress Ugh!
 

Tim R

unread,
Dec 2, 2024, 11:24:30 PM12/2/24
to SEBHC
Glenn,

Yes, what I was seeing was something like...

0  3  0    0  0  0    0  0  0

0  3  0    0  0  1    3  7  7

0  3  0    0  0  2    0  0  0

0  3  0    0  0  3    3  7  7


But, that has now changed.  I reseated the ROM in it's socket, and replaced the card.  I'm not sure which of those things helped, but the problem was in part an electrical connection issue.  I now see more reasonable values as I single-step through the H17 ROM memory, but I only stepped a little way through.

There is still an issue with the H17, and I think I'm on a good path to tracking it down.  Here is how the problem behaves. If I dump the ROM via the H8, I get repeated code blocks, for example:

[030000]   303 014 037 041  300 377 071 353  041 100 040 166  042 076 040 066

[030020]   303 014 037 041  300 377 071 353  041 100 040 166  042 076 040 066 (this is incorrect)

[030020]   270 312 046 030  166 000 043 315  216 030 302 036  030 303 032 030

[030040]   270 312 046 030  166 000 043 315  216 030 302 036  030 303 032 030 (this is incorrect)

And so on. 

If I pull the ROM and read it from my T43 ROM reader/burner directly, the data looks fine and exactly matches what is in the 2716-444-19-H17.ROM file you sent me. So the ROM itself is just fine, as I think you suspected.

So, after doing a short analysis on the binary addressing to the ROM, it appears the A4 address line is always held low. I haven't traced the cause of this down yet, but that should be easy enough to find. I'm pretty sure the issue will be isolated to the controller card (or its edge connector) as all other memory addressing with the system seems good. It still loads and runs BH Basic fine as a simple test.

The question still is... what caused this? I was previously able to run the memory check routines from the H17 ROM fine before, as well as operate the drive motor, head solenoid, and indicator light,  so this address line issue was not present then. Perhaps it is gremlins in chip sockets or the backplane connection.  It sure wouldn't be the first time.

Thanks for your suggestions, especially the idea of dumping the ROM. It's good to have that eliminated from the list of possibilities. It probably seems like something I should have obviously done first, but after spending the day on this and getting so close to it working... well, when the rock ran all the way back down the hill, my troubleshooting went with it, lol.

I will also post another message with the other circuit board photos that you and Joe requested. I'm not sure how common the 64K board is that is in the system.  I did test the board using the routine in the H8 operations manual, as well as the H8 memory test routing on the H17 (when it was working) as described on page 4 of the H17 Operations manual.  It will be interesting to see what thoughts or experience you all have with this card.

Thanks again,

Tim

Tim R

unread,
Dec 2, 2024, 11:36:43 PM12/2/24
to SEBHC
Joe,

Very true. I've attached photos of the remaining boards.

H8-panel.jpgH8-CPU.jpgH8-64K.jpgH8-5.jpg

Glenn Roberts

unread,
Dec 3, 2024, 7:35:19 AM12/3/24
to se...@googlegroups.com, SEBHC
So I’m now more comfortable that the 64k board is configured and working properly. Switches and jumper look ok. Curious what you’d get if you pulled the h17 controller and then displayed the contents at 030.000. Presumably you’d get all zeros.

Seems like most likely an addressing problem on the h17 board? I presume you’ve pulled and tested all those chips? Worth a close inspection of the sockets and board traces and a reflow on any suspicious solder joints.

Does your system have gold-plated backplane pins or tinned ones? The tinned connectors were notorious for flaky connections…


Sent from my iPad

On Dec 2, 2024, at 11:24 PM, Tim R <t...@primal-bits.com> wrote:

Glenn,

Glenn Roberts

unread,
Dec 3, 2024, 7:48:05 AM12/3/24
to <sebhc@googlegroups.com>
Just looked at your pictures and I believe I see the tinned backplane connectors?

Worth rearranging boards in the backplane. You may want to bump up the priority on replacing the backplane headers. I use Phoenix enterprises HWS3002 parts for this…



Sent from my iPad

On Dec 3, 2024, at 7:35 AM, Glenn Roberts <glenn.f...@gmail.com> wrote:

So I’m now more comfortable that the 64k board is configured and working properly. Switches and jumper look ok. Curious what you’d get if you pulled the h17 controller and then displayed the contents at 030.000. Presumably you’d get all zeros.

dwight

unread,
Dec 3, 2024, 9:30:07 AM12/3/24
to SEBHC

Repeat blocks cold be an indication of a missing address bit. It still might be a socket problem use an ohm meter and check the ROM socket. Measure from the IC pin to the bottom of the board.
Dwight

From: se...@googlegroups.com <se...@googlegroups.com> on behalf of Tim R <t...@primal-bits.com>
Sent: Monday, December 2, 2024 8:24 PM

Tim R

unread,
Dec 3, 2024, 10:49:00 AM12/3/24
to SEBHC
I did check memory contents at 030 000 with the H17 pulled, and did see all zeros. 

With the exception of the 2114 RAM chips, I have pulled and tested all chips and they all test good. I do have “known good” spares for the 2114s if needed. I’m pretty sure that the problem is local to the H17 and address line A4 on the ROM. Following that path on the schematics this morning. My bets are on a chip socket. 

Yes, I do have the tinned backplane, and plan to upgrade that soon. I can see this being a source of reliability issues going forward.

Joseph Travis

unread,
Dec 3, 2024, 10:54:29 AM12/3/24
to se...@googlegroups.com
Tim -

Thank you for the photos, they tell me a lot more about the system.  Please tell me what appears on the H8 display when you power it on / front panel reset.  Would you also tell me what data appears from address 000 032 to 000 037 (6 bytes)?

Glenn has mentioned replacing the tin pin bus connectors with gold ones to improve reliability.  It's not urgent but is worth consideration.  I will say that one of my H8s has an all tin pin bus and has been reliable for several years.  My H8 test fixture (original Heathkit property) has a mix of tin and gold (also very reliable).  "Darth Vader"  has an all gold bus, sees the most use / abuse and continues to be going strong too.

I will tell you that the white IC sockets on any of the H8 boards can be problematic.  They are subject to oxidation and deterioration.  Oddly, the one saving grace may be the use of TI (Texas Instrument) ICs which have silver plated pins (which turn "black" due to tarnish).  It is a good idea to burnish the contacts by  removing / reinserting ICs in these sockets.  Be careful when doing this.

On another note, if you find the need to replace any of the "black bullet" tantalum capacitors (on your floppy drives for instance), you can replace them with axial lead electrolytics of the same value and (perhaps increased) voltage rating.

One last suggestion, I encourage you to remove the 3 pin connectors going to the voltage regulators mounted to the aluminum bracket and solder the wires directly to the terminals.  I recommend using solid wires and heat shrink tubing.

Regards,
Joe





Glenn Roberts

unread,
Dec 3, 2024, 11:58:39 AM12/3/24
to se...@googlegroups.com

Tim: you're on the right track. I have high confidence that you'll have this working very soon!


Tim R

unread,
Dec 3, 2024, 1:15:03 PM12/3/24
to SEBHC
Joe,

Here is the data:

000 032  040
000 033  046
000 034  030
000 035  303
000 036  043
000 037  000

I will get rid of the regulator plugs as you suggest.  

I've narrowed the search in the 12V side of the second drive to one of two "black bullet" tantalum caps.  Thanks for the proper terminology of those caps. I've seen them referred to as "tubular tantalums" on the Mouser Electronics site, but I like "black bullet" better. It fits well for when they fail shorted and shoot your power supply dead, lol.

Thanks,

Tim
Message has been deleted

Tim R

unread,
Dec 3, 2024, 4:13:01 PM12/3/24
to SEBHC
Tim R
1:09 PM (now) 
to SEBHC
Hi all,

I've found the problem with the A4 line addressing on the H17 ROM.  The edge card connector has failed. The little internal spring contact in pin 34 is gone.  Edge connector 34 goes to the inverter for A4. I also go a few instances of errors on the A1 address line on the H17 ROM, and sure enough one of the little contact tab is bent flat in the edge connector related to A1 (edge connector 31).  Edge connector 29 (RESET) is also bad.

So, three edge connector connections are bad:
#29 (RESET) - both tab in edge socket
#31 (!A1) - unreliable connection with only one of tab bent flat in edge socket
#34 (!A4) - tab missing in edge socket

So, I need to replace the edge connector on the H17 card.  Does anyone have a recommendation for a good quality replacement?

Tim

Glenn Roberts

unread,
Dec 3, 2024, 4:58:29 PM12/3/24
to se...@googlegroups.com
Congrats! I can send you a replacement edge connector. These are still available but in smaller sizes, e.g. you could use a 10 and a 15.

I’ll send you an off list email…


Sent from my iPad

On Dec 3, 2024, at 4:09 PM, Tim R <t...@primal-bits.com> wrote:

Hi all,

I've found the problem with the A4 line addressing on the H17 ROM.  The edge card connector has failed. Neither of the two little contacts in pin 34 are there.  Edge connector 34 goes to the inverter for A4. I also go a few instances of errors on the A1 address line on the H17 ROM, and sure enough one of the little contact tabs is gone at the edge connector related to A1 (edge connector 31).  Edge connector 29 (RESET) is also bad.

So, three edge connector connections are bad:
#29 (RESET) - both tabs missing in edge socket
#31 (!A1) - unreliable connection with only one of two tabs present in edge socket
#34 (!A4) - both tabs missing in edge socket

Tim R

unread,
Dec 3, 2024, 7:04:04 PM12/3/24
to SEBHC
Thanks Glenn,

Until I replace the connectors, I've bodged the faulty edge connector locations using resistor leads to fashion temporary spring tabs.

I can now run both of the memory tests from the H17 controller again (pages 4 &5 of the H17 Operations Manual).

Tim

Joseph Travis

unread,
Dec 3, 2024, 7:28:29 PM12/3/24
to se...@googlegroups.com
Tim -

Congrats on troubleshooting the problems you've encountered!  I'm curious as to what your ultimate goal(s) may be with this system?

BTW - For future reference, the monitor ROM you currently have on your 8080 CPU board is known as the PAMGO (aka PAM8GO) ROM which was the 2nd release.  It added the ability to simply press the 'GO' key (after power on / reset) to boot from the H17 drive.  Prior to this you would have entered the following sequence of keys: REG, PC, ALTER, 030 000, ALTER, GO.

Joe


Glenn Roberts

unread,
Dec 3, 2024, 7:45:05 PM12/3/24
to se...@googlegroups.com, se...@googlegroups.com
The SEBHC website doesn’t show the right part number for PAMGO

Joseph Travis

unread,
Dec 3, 2024, 7:58:43 PM12/3/24
to se...@googlegroups.com
The correct part# for the PAMGO ROM is 444-60.

tim primal-bits.com

unread,
Dec 4, 2024, 12:03:57 AM12/4/24
to se...@googlegroups.com
Joe,

My ultimate goal is to keep this system as original, and complete as possible, with the intent of showing it at retro computer fairs and west coast VCF events. 
If I ever find one, II’d even like to be able to demonstrate the system with a paper tape reader. For these purposes, I plan to stick with HDOS rather than 
CPM, and avoid modernization upgrades

Tim

On Dec 3, 2024, at 4:28 PM, Joseph Travis <jtravi...@gmail.com> wrote:


You received this message because you are subscribed to a topic in the Google Groups "SEBHC" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/sebhc/hcTssNuf5Vo/unsubscribe.
To unsubscribe from this group and all its topics, send an email to sebhc+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/sebhc/CAGQDgBBXZd7Vb2hmxcZ4LOWX3NGgmjcHxQxHzevmAYRB4jduDA%40mail.gmail.com.

Joseph Travis

unread,
Dec 4, 2024, 8:34:25 AM12/4/24
to se...@googlegroups.com

Tim R

unread,
Mar 21, 2025, 4:26:47 PM3/21/25
to SEBHC
Hi all, I have an update of progress with the H17. The short report it that the controller and both of the Wangco 82 drives are now working. On the controller card, one of the backplane connectors was bad, and a replacement (thanks Glenn) put everything back in order with the controller. Also as noted previously, one of the drives had a shorted "black bullet" capacitor on the 12V rail which had to be replaced.  Additionally, several of the power supply caps were bad along with the fuse holder.

On the drives, a lot of the bearings were frozen, so that had to be addressed. After fixing the bearings and rebuilding the drives, I hit a wall getting the first drive (SY0:) working.  Glenn looked at the drive, and discovered the source of the issue. I had reassembled the drive incorrectly so that the "pressure arm" could not move freely. This also caused the felt disk on the pressure arm to become dislodged. Once Glenn shared what the problem was with the first drive, I discovered I had made the same mistake on the other drive (SY1). After making the corrections it is now working too.

So, that is the last piece of the puzzle in the restoration of the H8 system. Everything is now fully functional. The H8 includes the original 8080 board, WH8-64 RAM card, H8-5 I/O card, and H17 disk controller card. The H17 chassis contains two Wangco 82 floppy drives. For a terminal, I'm using an H9. It's really satisfying to see everything working together as none of the equipment was operational when I received it.

Time to dig into the HDOS manual now!

Tim
D2396D88-A149-4A9F-BA0F-494E23C25311_1_105_c.jpeg

Joseph Travis

unread,
Mar 21, 2025, 4:35:13 PM3/21/25
to se...@googlegroups.com

Glenn Roberts

unread,
Mar 21, 2025, 4:41:33 PM3/21/25
to se...@googlegroups.com, SEBHC
Beautiful! Congratulations 

Sent from my iPad

On Mar 21, 2025, at 4:26 PM, Tim R <t...@primal-bits.com> wrote:

Hi all, I have an update of progress with the H17. The short report it that the controller and both of the Wangco 82 drives are now working. On the controller card, one of the backplane connectors was bad, and a replacement (thanks Glenn) put everything back in order with the controller. Also as noted previously, one of the drives had a shorted "black bullet" capacitor on the 12V rail which had to be replaced.  Additionally, several of the power supply caps were bad along with the fuse holder.

On the drives, a lot of the bearings were frozen, so that had to be addressed. After fixing the bearings and rebuilding the drives, I hit a wall getting the first drive (SY0:) working.  Glenn looked at the drive, and discovered the source of the issue. I had reassembled the drive incorrectly so that the "pressure arm" could not move freely. This also caused the felt disk on the pressure arm to become dislodged. Once Glenn shared what the problem was with the first drive, I discovered I had made the same mistake on the other drive (SY1). After making the corrections it is now working too.

So, that is the last piece of the puzzle in the restoration of the H8 system. Everything is now fully functional. The H8 includes the original 8080 board, WH8-64 RAM card, H8-5 I/O card, and H17 disk controller card. The H17 chassis contains two Wangco 82 floppy drives. For a terminal, I'm using an H9. It's really satisfying to see everything working together as none of the equipment was operational when I received it.

Time to dig into the HDOS manual now!

Tim
To view this discussion visit https://groups.google.com/d/msgid/sebhc/aa6e7f8c-80be-4817-bf6e-5d8df2dcbf44n%40googlegroups.com.
<D2396D88-A149-4A9F-BA0F-494E23C25311_1_105_c.jpeg>

Dave McGuire

unread,
Mar 21, 2025, 4:46:53 PM3/21/25
to se...@googlegroups.com
Very nice! H17s are wonderful.

-Dave

--
Dave McGuire, AK4HZ
New Kensington, PA

norberto.collado koyado.com

unread,
Mar 21, 2025, 5:45:19 PM3/21/25
to se...@googlegroups.com

Nicely done!  It feels great when everything is working as designed.

 

Thanks,

Norberto

 

From: se...@googlegroups.com <se...@googlegroups.com> On Behalf Of Tim R
Sent: Friday, March 21, 2025 1:27 PM
To: SEBHC <se...@googlegroups.com>
Subject: Re: [sebhc] Re: H17 Refurb progress Ugh!

 

Hi all, I have an update of progress with the H17. The short report it that the controller and both of the Wangco 82 drives are now working. On the controller card, one of the backplane connectors was bad, and a replacement (thanks Glenn) put everything back in order with the controller. Also as noted previously, one of the drives had a shorted "black bullet" capacitor on the 12V rail which had to be replaced.  Additionally, several of the power supply caps were bad along with the fuse holder.

 

On the drives, a lot of the bearings were frozen, so that had to be addressed. After fixing the bearings and rebuilding the drives, I hit a wall getting the first drive (SY0:) working.  Glenn looked at the drive, and discovered the source of the issue. I had reassembled the drive incorrectly so that the "pressure arm" could not move freely. This also caused the felt disk on the pressure arm to become dislodged. Once Glenn shared what the problem was with the first drive, I discovered I had made the same mistake on the other drive (SY1). After making the corrections it is now working too.

 

So, that is the last piece of the puzzle in the restoration of the H8 system. Everything is now fully functional. The H8 includes the original 8080 board, WH8-64 RAM card, H8-5 I/O card, and H17 disk controller card. The H17 chassis contains two Wangco 82 floppy drives. For a terminal, I'm using an H9. It's really satisfying to see everything working together as none of the equipment was operational when I received it.

 

Time to dig into the HDOS manual now!

 

Tim

Lee Hart

unread,
Mar 28, 2025, 3:18:35 PM3/28/25
to se...@googlegroups.com
I love stories with a happy ending. Congratulations on your successful repairs and a great looking system!

Virus-free.www.avast.com

norberto.collado koyado.com

unread,
Mar 28, 2025, 4:04:52 PM3/28/25
to se...@googlegroups.com
Now with the DOGE team eliminating hundreds of vintage computers that are 30 to 50 years old we shall see a lot of PDP’s 8 and 11 on eBay hopefully. 


From: se...@googlegroups.com <se...@googlegroups.com> on behalf of Lee Hart <electr...@gmail.com>
Sent: Friday, March 28, 2025 12:18:12 PM
To: se...@googlegroups.com <se...@googlegroups.com>

Geo Monroe

unread,
Mar 28, 2025, 4:24:37 PM3/28/25
to se...@googlegroups.com
i cant find any hard sectored hdos diskettes

George


Benjamin Hansen

unread,
Mar 29, 2025, 11:48:34 AM3/29/25
to se...@googlegroups.com
I'd love to get my hands on a working pdp8.  I'll have to keep my eyes open and hope the don't all just go to scrap.

Dave McGuire

unread,
Mar 29, 2025, 12:07:28 PM3/29/25
to se...@googlegroups.com

Gov't surplus has changed radically in the past 15-20 years; a lot of
it goes to scrap but some will hit eBay. But either way, I think the
days of PDP-8s coming out of gov't surplus are long behind hs.

You can always visit LSSM; we have running PDP-8s there. You can't
take them home, but you can spend a day (or two, three..) hacking on them.

-Dave

On 3/29/25 11:48, Benjamin Hansen wrote:
> I'd love to get my hands on a working pdp8.  I'll have to keep my eyes
> open and hope the don't all just go to scrap.
>
> On Fri, Mar 28, 2025, 15:04 norberto.collado koyado.com
> <http://koyado.com> <norberto...@koyado.com
> <mailto:norberto...@koyado.com>> wrote:
>
> Now with the DOGE team eliminating hundreds of vintage computers
> that are 30 to 50 years old we shall see a lot of PDP’s 8 and 11 on
> eBay hopefully.

Michael Rooney

unread,
Mar 29, 2025, 11:59:48 PM3/29/25
to se...@googlegroups.com
I believe I still have one or more DEC tapes somewhere. 

At one point I had a couple of different PDP-11s and one VAX. 

Ah, the good old days. 
 
Regards,
..michael..
Michael Rooney
3047 Centerville Road
Hyde Park VT 05655

image001.jpg

Tim R

unread,
Apr 30, 2025, 11:40:54 PM4/30/25
to SEBHC
Hi all. I made an attempt to document the H17 restoration. The video is a bit rough, but hopefully it covers the journey. Sorry in advance for the music, I was trying something a little different.


Tim

Glenn Roberts

unread,
May 1, 2025, 6:06:28 AM5/1/25
to se...@googlegroups.com, SEBHC
Beautiful job Tim. Great tutorial for anyone else attempting to get their system running.

As for the Y2K compatibility question: HDOS is partially Y2K compatible. Interestingly the internal date format is compatible but the routines that read and write dates are not!

The simple solution for this is to remove the 70 year bias stored in the date. This can be done as a patch and works reasonably well. That is the patch most of us are using I believe. 
The “correct” fix adds slightly more code to the date routines and is a more complicated patch, or can be implemented by reassembling HDOS from the source.

I can provide a more thorough discussion on this if there’s interest…

Again, great job on all your videos. We should provide links to them from the wiki

- Glenn



Sent from my iPad

On Apr 30, 2025, at 11:40 PM, Tim R <t...@primal-bits.com> wrote:

Hi all. I made an attempt to document the H17 restoration. The video is a bit rough, but hopefully it covers the journey. Sorry in advance for the music, I was trying something a little different.

Joseph Travis

unread,
May 1, 2025, 9:17:58 AM5/1/25
to se...@googlegroups.com

Nicely done!  Very comprehensive.  I took the liberty of posting the video on fb.


Reply all
Reply to author
Forward
0 new messages