Found these inside a non-working 71B

[Dave Frederickson]

(03-14-2018 03:56 AM)rprosperi Wrote:  

(03-14-2018 03:39 AM)MikeSD Wrote:  Regarding FREE:PORT (0.04) That doesn't work but .05 and .06 do.

Interesting. Perhaps the IL Mailbox (usually hidden port 0.04) must stay at that port address, and the added RAM picks up at the next available port (0.05 & 0.06).

Can you try SHOW PORT (after FREE-porting both 32K banks)? This should confirm the HP-IL ROM is 16KB, type=2, PORT(0.07) which should clarify all remaining unknowns.

The MATH ROM is at Port(.04). Second, unless DOUT from the last 32k module was connected back into the daisy chain, the HP-IL Module, if installed, will go undetected.

[rprosperi]

(03-14-2018 04:06 AM)Dave Frederickson Wrote:  The MATH ROM is at Port(.04).

Thanks Dave, forgot the MATH chip was wired-in before the RAM.

Still, it would be interesting to see the full SHOW PORT results; this would also confirm the IL port is properly wired (if an adapter is inserted).

[Dave Frederickson]

(03-14-2018 12:49 PM)rprosperi Wrote:  

(03-14-2018 04:06 AM)Dave Frederickson Wrote:  The MATH ROM is at Port(.04).

Thanks Dave, forgot the MATH chip was wired-in before the RAM.

Still, it would be interesting to see the full SHOW PORT results; this would also confirm the IL port is properly wired (if an adapter is inserted).

Mike hasn't mentioned whether or not he has an HP-IL module or any peripherals. If not, he can key-in Paul's MEMBUF.

Dave

[rprosperi]

(03-14-2018 02:27 PM)Dave Frederickson Wrote:  Mike hasn't mentioned whether or not he has an HP-IL module or any peripherals. If not, he can key-in Paul's MEMBUF.

Module, yes, it's listed! I had the same thought and sent the MEMBUF listing to enter.

[Paul Berger (Canada)]

(03-13-2018 07:17 PM)J-F Garnier Wrote:  

(03-13-2018 06:19 PM)Paul Berger (Canada) Wrote:  Reading a little more about the memory configuration you could have a RAM module that contains more than 32K bytes the 64K nibble limit is for an individual chip but a module could contain more than one chip, ...

The HHP card reader RAM modules (made of 32K boards) can be configured to create a 64K port from two boards. See here for instance (the link to the file is dead, use this instead).
Unfortunately, it is not possible (or not known) to do the same with 32K CMT modules -well it should be possible since the CMT01 programmer does it.

BTW, I'm still interested by any information on the 1LQ4 chip. Somebody once said that a documentation exists but never shared.

J-F

I would be interested in actual documentation of the chip as well.

I will have to try that 64K mod I have a 128K module in pieces right now because the SRAM was bad on one of the boards, that is fixed now, at least I ran diags for about 18 hours and it did not find a problem. I was not so lucky my bad SRAM was was on the base board in the middle of the stack and had more than one bad chip, it is now a single 32K SRAM. I should look and see if the later modules that used a single 32K SRAM instead of 4 8K have a similar arrangement, if not I can always lift the pin and ground it.

Paul.

Is that your document? I can fill in more of the pinout of the 1L?4 chip if you wish.

[MikeSD]

(03-14-2018 02:27 PM)Dave Frederickson Wrote:  Mike hasn't mentioned whether or not he has an HP-IL module or any peripherals. If not, he can key-in Paul's MEMBUF.

Dave

I have not put the HPIL in yet. I have one and will put it in and test. First the FORTH, then the HPIL. Then I will do a SHOWPORT and see what shows up. I'm feeling good that the problem was just broken wires.

I don't have any way to input programs at present. No reader, or 9114. Got about a dozen prineters but that's no help

[J-F Garnier]

(03-14-2018 04:20 PM)Paul Berger (Canada) Wrote:  I will have to try that 64K mod I have a 128K module in pieces right now because the SRAM was bad on one of the boards, that is fixed now, at least I ran diags for about 18 hours and it did not find a problem. I was not so lucky my bad SRAM was was on the base board in the middle of the stack and had more than one bad chip, it is now a single 32K SRAM. I should look and see if the later modules that used a single 32K SRAM instead of 4 8K have a similar arrangement, if not I can always lift the pin and ground it.

Is that your document? I can fill in more of the pinout of the 1L?4 chip if you wish.

Yes, this document is mine.
Of course I will be interested with the identification of more pins. Especially pins that may define the ROM/RAM type, capacity and organisation (1 x 32K, 4 x 8K, ...).
The pin defining the last-chip-in-chain was easy to find since it is connected to a pad.

Actually, the pictures were taken when I fixed a bad SRAM too !
Does it mean RAM failures in HHP modules are common ? I may suspect a excessive supply voltage, the HP71 power supply may go up to 6.5V, and standard chips have a max operating voltage of 5.5V and sometimes an AMR of 6V !
I didn't see any special circuitry to reduce the supply voltage in the HHP or CMT modules. A simple diode may help.

J-F

[Dave Frederickson]

(03-14-2018 04:22 PM)MikeSD Wrote:  

(03-14-2018 02:27 PM)Dave Frederickson Wrote:  Mike hasn't mentioned whether or not he has an HP-IL module or any peripherals. If not, he can key-in Paul's MEMBUF.

Dave

I have not put the HPIL in yet. I have one and will put it in and test. First the FORTH, then the HPIL. Then I will do a SHOWPORT and see what shows up. I'm feeling good that the problem was just broken wires.

I don't have any way to input programs at present. No reader, or 9114. Got about a dozen prineters but that's no help

Unfortunately the 1BBBB and earlier version of SHOWPORT is crippled and won't show all the ports. To run the new version you'll need some HP-IL I/O or a card reader. If you plan on having more fun with your 71 I highly suggest the acquisition of a PIL-Box from J-F.

Otherwise, keying-in MEMBUF seems to be the only way to do a thorough analysis of your ports.

Dave

[MikeSD]

I'm still puzzled about why the RAM didn't work on my unmodified 71B. I did not have HPIL attached when testing.

But there is something different about the original unit it works in, which might explain why in it and not an unmodified one.

I've attached a pic below of the back of the PCB on the one that works. The U9 (I think) bottom chip, pin 9, has a jumper to the Math ROM (top chip, pin 8), which one might expect, if expanding to that top chip spot.

But that U9-9 also goes to pin 10, which picks up the trace to HPIL, when the top chip slot isn't needed. Not sure why it needs to go to pin 10 at all. If they wanted to also go to the HPIL, they could have done it without going to pin 10.

That said, the trace is broken where it connects to the HPIL trace. So, since U9-9 doesn't go directly to HPIL any longer and the trace is cut, how does HPIL get back in the loop. AND, might this extra wiring and cutting, be the cause for the RAM not working in the unmodified unit.

And now that I think about it, the rest of the chips is how it gets back to HPIL. I think pin 10 is just connected but not used, or possible for some other reason. The chain goes from Math, to RAM, to FORTH, THEN to the HPIL.

Dang! Still don't have an explanation for why RAM didn't work in unmodified unit.

[Paul Berger (Canada)]

(03-14-2018 06:19 PM)MikeSD Wrote:  I'm still puzzled about why the RAM didn't work on my unmodified 71B. I did not have HPIL attached when testing.

But there is something different about the original unit it works in, which might explain why in it and not an unmodified one.

I've attached a pic below of the back of the PCB on the one that works. The U9 (I think) bottom chip, pin 9, has a jumper to the Math ROM (top chip, pin 8), which one might expect, if expanding to that top chip spot.

But that U9-9 also goes to pin 10, which picks up the trace to HPIL, when the top chip slot isn't needed. Not sure why it needs to go to pin 10 at all. If they wanted to also go to the HPIL, they could have done it without going to pin 10.

The only reason I can see for the line to be routed this way is for convenience of laying out the PCB, in any case I would not get too wound up in why it is this way.

(03-14-2018 06:19 PM)MikeSD Wrote:  That said, the trace is broken where it connects to the HPIL trace. So, since U9-9 doesn't go directly to HPIL any longer and the trace is cut, how does HPIL get back in the loop. AND, might this extra wiring and cutting, be the cause for the RAM not working in the unmodified unit.

Ok so now Dout (pin 9) from the Math module is going out to Din (Pin 8) of the first 32K module the first is daisy chained to the second RAM that is hung off the wires. Dout from the second 32K has to be connected somewhere to complete the chain.

If you have the FORTH module connected it would go to Din on the FORTH module and the Dout from the FORTH module would go to the pad on the board where Dout from the module that is in the Math position would normally be connected, That pad should be connected to Din on the connector for the HP-IL.

If you do not have the FORTH module connected then it would be the Dout from the second 32K that would get get connected to that point on the board.

The trace that is cut effectively bypasses the module position where the Math module is installed as far as the daisy chain is concerned, by making that cut you isolate the Dout of U9 from the trace going to the HPIL connector, so that it is only connect to Din for the position where the Math module is, and only Dout for the position above U9 is connected to Din on the HPIL, so the to complete the chain, Dout from the last module before HPIL need to be connected to the Dout pad for the position above U9.

(03-14-2018 06:19 PM)MikeSD Wrote:  And now that I think about it, the rest of the chips is how it gets back to HPIL. I think pin 10 is just connected but not used, or possible for some other reason. The chain goes from Math, to RAM, to FORTH, THEN to the HPIL.

Dang! Still don't have an explanation for why RAM didn't work in unmodified unit.

Yes pins 10,11,and 12 are not connected to anything on memory modules, so the board designer could safely route signals through those connections.

No idea why it would not work in your test 71B if everything was connected correctly, Dout from U9 should have been routed directly to the HPIL so connecting it Din of the first 32K module should have worked.

Paul.

[MikeSD]

Gonna stop for a day and think about this.

I decided to put the RAM in it's final place. Was working good tacked onto the top of one of the RAMs. You can see white wires in the photo above, on right.

Put it in the final place, soldered everything in, tucked the wires away. Looking pretty good. Then I powered it up, and checked the 32K RAMS. Still working great.

But dang! One of the 4K mem chips wasn't configuring now. Total RAM 80,xxx. Wonder if it's the one I was tacked onto. Generally the last thing you touch is where the problem is. I have verified it's the 0.03 bank that won't configure. I can FREE and Claim ports 0-2 but not 3. I need to find a layout so I know which RAM is 0.03. I have lots of 4K RAM, so replacing it, if dead, isn't a problem. But I think it's just a bridge, or bad solder of something I just touched.

I was tacked onto the top RAM (back right) in the photo above.

What Ref is that? And which bank is it? The only 2 devices I touched are the on back right in the photo above, and the back left (Math), in the photo above. No other RAMs were touched.

[Dave Frederickson]

(03-14-2018 09:15 PM)MikeSD Wrote:  What Ref is that? And which bank is it? The only 2 devices I touched are the on back right in the photo above, and the back left (Math), in the photo above. No other RAMs were touched.

Mike, you do have a copy of the schematics, right? They can be found in the H/W IDS and the Service Manual.

To answer your question, U9 is Port(0.3), but how do you know if that's where the issue is? If there were a short on U6 between DIN and DOUT, the Port(.01) RAM, then it may go undetected and U7 would be configured as Port(.01).

I think all you can be certain of is that only three of the four 4k modules were detected.

Dave

[Paul Berger (Canada)]

0.03 will be U9 provided you added on to the end of the internal daisy chain, where did you connect Din? It does not appear like your photo came through for your latest update.

Paul.

[MikeSD]

(03-14-2018 09:54 PM)Paul Berger (Canada) Wrote:  0.03 will be U9 provided you added on to the end of the internal daisy chain, where did you connect Din? It does not appear like your photo came through for your latest update.

Paul.

i didnt include a photo. i was referring to the one aboce. Makes since that would be the one, since thats the area i was working in. However I didn't do any soldering to it. The Math is added it that chip and soldering is questionable there. Might have bad solder joint. So i'll redo that soldering and jopefully resolve the problem. That chip is easier to replace too.

[MikeSD]

(03-14-2018 09:49 PM)Dave Frederickson Wrote:  

(03-14-2018 09:15 PM)MikeSD Wrote:  What Ref is that? And which bank is it? The only 2 devices I touched are the on back right in the photo above, and the back left (Math), in the photo above. No other RAMs were touched.

Mike, you do have a copy of the schematics, right? They can be found in the H/W IDS and the Service Manual.

To answer your question, U9 is Port(0.3), but how do you know if that's where the issue is? If there were a short on U6 between DIN and DOUT, the Port(.01) RAM, then it may go undetected and U7 would be configured as Port(.01).

I think all you can be certain of is that only three of the four 4k modules were detected.

Dave

I have an old schematic in one of my docs. For some reason my IDS don't have a schematic or layout with refs.

I guess I don't know which it is then, except that only the 0.03 can't be configured. I guess it could be any of them then. The only 2 RAMS I touched were the U9 RAM (and I didn't solder directly to it) but the Math ROM was attached to it. And the RAM on the other side of the board, on TOP. Those are the only 2 RAMS touched.

I'll buzz them out and follow the daisy chain, to try and locate where the fault is. It was working before I put everything in the final place, so likely it's related to the last ones touched. U9 and the one on top, on the other side. I'll pick up with that tomorrow and finish the FORTH.

I'm going to bond down all the wires, where they connect to the board, to keep vibrations from breaking these wires. ONCE it's all checked out.

[Didier Lachieze]

(03-14-2018 10:33 PM)MikeSD Wrote:  

(03-14-2018 09:49 PM)Dave Frederickson Wrote:  Mike, you do have a copy of the schematics, right? They can be found in the H/W IDS and the Service Manual.

I have an old schematic in one of my docs. For some reason my IDS don't have a schematic or layout with refs.

Here is a link to a scan of my IDS with schematics and layouts: HP-71_Hardware_IDS.pdf

[MikeSD]

Thanks. i'll look at schematics.

Turns out it's a little more complicated than I thought. I still think the problem is a connection. But it fluctuates between 4K and 2k, missing. And the exact numbers seem odd. i removed the 64k since those are reportin gòod.

The only RAM chip I soldered to is the top chip on other side of board. The only thing that makes sense is if one of the nibble lines is open, to cause it to toggle betseen the two values.

oh well, on to tomorrow.

thanks for schematic. much better than my copy

Update: When I have a problem, I just can't let it go. The only thing that made sense was my unsoldering the 64k RAM from the 4K RAM chip. So, I reflowed the solder on that chip and now its working again. Only FORTH to go, then final button down and test.

[Dave Frederickson]

(03-15-2018 02:22 AM)MikeSD Wrote:  Only FORTH to go, then final button down and test.

Don't forget HP-IL. BTW, you can plug any module into the HP-IL port except the Service Module.

[Dave Frederickson]

U10 you ask? There's no U10 on the schematic.

No, but it's clearly on the PCB, right where Mike installed the Math ROM there is a footprint for a 6th internal memory module. The schematic is missing some parts.

It looks like there's some funny stuff going on with the higher numbered pins on that part as there are a number of connections between the pins and also the pins on U9. In addition, the other version of the PCB (with the golden zebra connector) is wired differently!

So what's U10? I think it's for the Debugger ROM mentioned in the IDS.

Dave

[MikeSD]

If I said that, I must have been tired. I think the only place I used 10 is that the U9-9 is routed to the Math ROM-8 and also to pin 10 on U9, for some reason.

I noticed there is a Forth Assembler part number for "with and without" a debugger. My part number is for the with debugger. Wonder how I verify that? Is the debugger a hardware item, or another ROM, or part of the Assembler?

Thinking out of the box
* I have a ton of 4K RAM boards, from salvaged 71Bs.
* And I have 4 unused front ports.
* And, I bet I could wire in some special connectors deep in those ports, to allow a kludged 8K RAM, made of 4K,
* to be inserted in and plugged into a special connector THAT wouldn't interfere with a standard plugin, if the RAM was removed.

Think that's possible? Can 8K RAMS be made, or do only 4K, 32K and 64K work as a group.

OR, I could put as many together as I could, that might fit in the Card Reader Port. Would like to find a good use for the 4K RAMs.