HRD 4000B

[RickyDean]

Has it done that since the build? If so after you do the IC tests, then I'd start looking for possible shorts under the sockets, starting with the 244's. Take a multimeter and see if you get continuity between two pins then investigate to see if you have a solder blob or a small solder ball ot wire piece shorting out a chip. If you find a possible short, check the same pins on your good 4000 to be sure that it is not doing the same.

Edited April 23 by RickyDean

[NewcomerTI]

Yea the IC’s were all swapped out / no change / going to try to detect the solder condition of perhaps a blob beneath the socket - starting with the 74LS244 sockets.

[NewcomerTI]

Continuity test look good with a few that showed a short but verified with the good 4000B that it was the same so it seems consistent. 
 

There was voltage to pin 7 of U1 which wasn’t on the good 4000B IIRC it was about 2.1 volts. Not sure where this would be coming from (I mean the battery once it was powered off the voltage was not there) but it seems odd - could be the reason for the odd scenario with this board- thanks for the help!

[Gary from OPA]

25 minutes ago, NewcomerTI said:

Continuity test look good with a few that showed a short but verified with the good 4000B that it was the same so it seems consistent. 
 

There was voltage to pin 7 of U1 which wasn’t on the good 4000B IIRC it was about 2.1 volts. Not sure where this would be coming from (I mean the battery once it was powered off the voltage was not there) but it seems odd - could be the reason for the odd scenario with this board- thanks for the help!

try the card without any batteries, you don't need the batteries in it to test the card

[RickyDean]

3 hours ago, NewcomerTI said:

Continuity test look good with a few that showed a short but verified with the good 4000B that it was the same so it seems consistent. 
 

There was voltage to pin 7 of U1 which wasn’t on the good 4000B IIRC it was about 2.1 volts. Not sure where this would be coming from (I mean the battery once it was powered off the voltage was not there) but it seems odd - could be the reason for the odd scenario with this board- thanks for the help!

Pin one of the U1, the 74LS154 ties into the pin 22 "CS" of the M1 memory location. Did you try changing the 154, as it may have a short. I don't have batteries on mine to check it with currently.

 

Edited April 23 by RickyDean
spelling

[NewcomerTI]

I did swap out LS154.  It’s still not configuring.  BTW It lights up the front LED on power up and hangs the whole system - (this is what I found allows the system to boot to E/A cartridge - power cycle the console three times, or turn the 4000B off then back on).  Once it ‘looks’ for the RAM disk it pretty much needs to be turned off and then back on to try to negotiate with the PEB, otherwise the light (LED) remains on; am I overlooking something?

 

[NewcomerTI]

I’m usually better than this on TS&R problems like this - my other two RAM cards work in the PEB. But this one lights when I turn the console on and I think if I can resolve this that the unit will configure without any hesitation. 

 

I might be overlooking something but it’s really being an evasive problem that is not easily traced after swapping out all the other IC’s on the working 4000B, each time hoping that the LED would light briefly and then get past the ability to load a disk to complete this configuration, getting to a scenario where the board stays stable so that when you try to load a ROS that it doesn’t skip past a screen that says press any key makes me think this board is actively sending signals out to try another attempt but it’s not real promising of getting to a resolve yet. 

 

 

[RickyDean]

Here are the schematics of the 4000B card, in this thread. You can start  running traces to see if something is not right in your build. @Ksarul's 4000B build is rock solid and he does a pretty good schematic too. Using the construction manual, these schematics, and the working card, you should be able to trouble shoot this issue. I believe it's a construction issue, solder joint, reversed diode, or something along that line. Just to check are you using this Ramdisk on a Geneve? or on a TI?  Are you using the 32k upgrade at M32 on the Board or leaving that off and using a 32k card internal/external? Are you leaving the the IC at U11 off if you are not using the 32k otion on board? Do you have both transisitor mounted so that the flat side faces toward the bottoms of the board, or if using the can type transistor, being sure the tab is pointed toward the upper right corner of the baord?

 

 

Edited April 24 by RickyDean
spelling, added content

[RickyDean]

I need to correct that the can style transistor should point upper left not right.

[NewcomerTI]

I’m using the 32K option with both chips installed. I have an internal card that I can throw in and remove the optional chips to see if that might make a difference… I’m running on a TI. Both transistors are flat side toward the bottom.  Thank you for the helpful advice and tips on trying before and for the schematic diagrams and manual links.  I’m wondering where the six (Russian style) diodes are in the schematic?  It is not evident or are they elsewhere and not depicted. Also k did get clarification that the schematic calls out 30uF tantalum and that they are indeed 10uF
 

[RickyDean]

19 minutes ago, NewcomerTI said:

I’m using the 32K option with both chips installed. I have an internal card that I can throw in and remove the optional chips to see if that might make a difference… I’m running on a TI. Both transistors are flat side toward the bottom.  Thank you for the helpful advice and tips on trying before and for the schematic diagrams and manual links.  I’m wondering where the six (Russian style) diodes are in the schematic?  It is not evident or are they elsewhere and not depicted. Also k did get clarification that the schematic calls out 30uF tantalum and that they are indeed 10uF
 

Her would be your six diodes in the upper right central of the schematic

[NewcomerTI]

Oh now I went and did it… the MENU for the working 4000B associated numbering on the items is corrupted or something got changed.  There is probably an easy fix rather than reload my whole RAM disk (if necessary I’ll wait until I get the other card sorted out first) but the numbers all look scrambled and the text at the bottom doesn’t look quite right.  Any thoughts?

 

Hopefully this is an easy one to correct.  While the thoughts are pouring in here is some TI midi master ‘classic’ tune to enjoy. 

 

 

[+Ksarul]

Note that the D9K is a suitable sub for the 1N34A. I use it because the D9K is always on-spec for the uses we put it to, unlike pretty much every 1N34A that's been on the market since the early 1990s (definitely not a new problem here).

[RickyDean]

13 minutes ago, NewcomerTI said:

Oh now I went and did it… the MENU for the working 4000B associated numbering on the items is corrupted or something got changed.  There is probably an easy fix rather than reload my whole RAM disk (if necessary I’ll wait until I get the other card sorted out first) but the numbers all look scrambled and the text at the bottom doesn’t look quite right.  Any thoughts?

IMG_3216.mov 26.91 MB · 1 download  

Hopefully this is an easy one to correct.  While the thoughts are pouring in here is some TI midi master ‘classic’ tune to enjoy. 

IMG_3215.mov 157.73 MB · 1 download  

 

You probably corrupted the menu in the 8k IC on U9, don't know if you've lost the data in the other memory IC's, you can try reloading just the ROS without formatting the card again and see if it will recover. But be prepared to reload the complete ramdisk @InsaneMultitasker  and others would have much more knowledge here.

[NewcomerTI]

Reloading the ROS (DSR) worked for the working 4000B card!  It looks as it should. Now back to the other RAM board and trying to configure it. 

 

 

[+InsaneMultitasker]

ROS does contain a character set that is referenced by MENU but if it was corrupted, both the inverted chars and the standard chars should have been corrupted.  It isn't clear to me how a ROS reload fixed the problem, as I would have guessed a MENU problem. Even if you had mixed and matched the ROS, I don't think that would cause the issue. 

 

MENU is stored as a file on the Ramdisk.  ROS is primarily stored in the 8k/32k ram, and is only 'user accessible' via CFG.   

[NewcomerTI]

Thanks for the info - to be clear I did have to run CFG1 and then load the ROS842 file and it asked if you wanted to keep the disk information - but it worked on getting the screen accurate and back to showing characters & numbers.

 

On the problem with the new RAM disk board I decided to ‘Diode’ check the 1N34A’s and found a one reading too far out of range and sure enough it was cracked.  I replaced it with the same type and I’m getting new results.  Not to say I’ve got it resolved but the board moves along further in the Configuration program to allow Disk info to be made - it still won’t allow me to load a ROS because it lights the LED on turning the console on.  It is recognizing ‘phantom’ boards again as GeneveROMpage items - further along or two steps back?

 

I replaced each IC to see if there was an offending chip and none of the many ones replaced fixed the LED staying on. I started with the pair of LS244’s - both at the same time 

 

in the scheme of things it appears I am nearer to trying to get to a resolve with the HRD 4000B, but at the same time it appears to be back to detecting too many non existent Geneve ROMpage items as before. But it gets past the Configure screen with toggling the RAM cards on and off switch so I can access the floppy controller. 

 

 

[RickyDean]

Not knowing how you installed the russian diodes, you are aware that they need to be placed int their locations with the black band reversed from an American diode correct? Not aware if this could be part of the issue, but maybe worth looking at.

[NewcomerTI]

Yes - I got them oriented correctly so it was easy to drop in the 1N34A to match it up with the others in there.  Hope to revisit why this board is behaving so odd today by reexamining all my steps in the construction of the board. 

[+Ksarul]

Note that if you have just one 1N34A in there, it will be installed with the band opposite the direction of the D9Ks. . .I usually run the D9Ks through an ohmmeter to verify the polarity, as that way I'm not forgetting what I'm doing during the installation.

[NewcomerTI]

I misspoke. Both of my RAM boards have the D9Ks in them and I replaced it with a spare D9K. Sorry for the incorrect naming of the device. 

[NewcomerTI]

I need help determining a graphic on the schematic. I’m using the schematic diagram here and I found a lifted pcb feed to one of the 74LS156 pins tied all together so I’m checking other pins but can’t determine what the open circle is at pins 1, 14 & 15.

Thanks for the help, that pin 14 of U7 goes off and should trace to pin 12 of U12 but I’m uncertain about the open circle and my boards don’t seem to trace to that pin.  

 

Edited April 25 by NewcomerTI
Number was stuck out in nowhere land.

[Gary from OPA]

26 minutes ago, NewcomerTI said:

I need help determining a graphic on the schematic. I’m using the schematic diagram here and I found a lifted pcb feed to one of the 74LS156 pins tied all together so I’m checking other pins but can’t determine what the open circle is at pins 1, 14 & 15.

Thanks for the help, that pin 14 of U7 goes off and should trace to pin 12 of U12 but I’m uncertain about the open circle and my boards don’t seem to trace to that pin.  

 

Just follow all the traces in the wiring for the 156. It goes to the 138 and the 259. And is the key in the memory mapping also with the diodes.

 

The 156 along with the 241 that it is connected with as well is important it what gives you the rambo option and allows the ram to appear at the 6000 space in 8k chunks instead of the 5800 space in 2k pages.

 

Also maybe the reason that led lights up right away when you turn on power if that circuit is messed up.

 

As for the little circle it just means the line is active low, enabled, don't worry about that just map out the lines see my screenshot best to view it via a large screen not the mobile or print it out the schematic.

 

Edited April 25 by Gary from OPA
more info

[+Ksarul]

As noted, it means the signal on that line will be an active low. Lines without the circle will be an active high. Note also that sometimes active lows will be tied high voltage-wise to make the circuit respond as desired, and active high lines may be tied low for the same type of reasons. . .the important parts of this are that the circles at the chip do not prevent the line from connecting or being read at any point along the trace it is part of, they just tell us what kind of signal is there.

[NewcomerTI]

21 hours ago, NewcomerTI said:

I need help determining a graphic on the schematic. I’m using the schematic diagram here and I found a lifted pcb feed to one of the 74LS156 pins tied all together so I’m checking other pins but can’t determine what the open circle is at pins 1, 14 & 15.

Thanks for the help, that pin 14 of U7 goes off and should trace to pin 12 of U12 but I’m uncertain about the open circle and my boards don’t seem to trace to that pin.  

 

Yes they trace to that pin so I’m moving forward in attempting to figure another chip not sending the correct signal out that will help me configure this board.