Quick measurement

[MartinB]

I'm on holiday and away from my Beeb stuff so if anyone has a Model B and a DMM handy and could do a quick couple of measurements for me, I would be most grateful. It's just two power-off passive resistance checks on the analogue (joystick) connector, between the the two fire button inputs and the +5v line. So, as illustrated below, from I0 Pin 13 to +5v Pin 1 and from I1 Pin 10 to +5v Pin 1. It would be useful to measure each twice with the meter leads reversed so four measurements in total please. That's it and thanks!

[VectorEyes]

Might be able to do this, if my multimeter probes will fit into the sockets ... Stupid newbie question, but that diagram is as if you're looking at the analogue socket from outside the case not inside, yes?

[MartinB]

Yes, and I think Pin 1 is top right but there are also probably little moulded numbers in the plastic. A straightened paper clip might fit the sockets....?

( For orders of magnitude btw, I'm thinking about 900 ohms ? )



.

[VectorEyes]

Ok, I've been trying with solder wire but can't get a good connection. Give me a few hours and I'll take the lid off and get to the connections from the inside.

[MartinB]

Please don't go to any trouble, it's certainly not urgent - I was just trying to settle a collection of project thoughts in my head. Anytime in the next few days will be fine...

[IanS]

Do you have some reason to believe it won't be ~10K

io.PNG (37.42 KiB) Viewed 436 times

[MartinB]

Yes, because of the 6522 having what I believe from previous testing to be an internal ~1k pull-up to 5v on the data lines. It's just something that I would check on a few Beebs if I was at home but being away, the thought is niggling me. If it is as I suspect, then the 10k's have actually reduced the overall I0/I1 pull-up resistance to ~900 ohms. No crisis, just interest and relevant to my other projects.



.

[VectorEyes]

DISCLAIMER: I know very little about electronics (I'm having flashbacks to A-level physics!) and I only bought my multimeter so I could check I'd recapped my PSUs properly. That being said, I *think* my approach of finding some wires, jamming one end into the socket and pressing the multimeter probe against the other end works!

I0 Pin 13 to +5v pin 1:

Black probe(COM) of MMD to pin 1, red probe to pin 13: 1.077 kOhm.
Red probe of MMD to pin 1, black to pin 13: 1.06 kOhm.

I1 pin 10 to +5v pin 1:

Black to pin 1, red to pin 10: 1.08 kOhm.
Red to pin 1, black to pin 10: 1.062 kOhm.

So looks like for both pairs it's approximately 1.08 kOhm in one direction and 1.06 kOhm in the other.

[MartinB]

That’s brilliant, thanks Adrian...

So your 6522 has internal pull-ups of roughly 1.2k ohms which is what I think I originally measured on some of mine.

Thanks again

(Any other samples would be interesting if anyone has a bored moment to spare.... )

[tricky]

As I had a DB15 to hand, I just had a quick check of my "to hand" machines and the issue 3 has the same reading as above, while the issue 7 and master both have 0.96x kOhms in both directions.
Caveat, my battery is low in my meter!

[Elminster]

Issue 4 Beeb

1.074 (10) & 1.071 (13) in K

[VectorEyes]

I should have mentioned that mine is an issue 7.

[MartinB]

Thanks also Richard and Duncan, that’s great...

I’ll explain in due course....

[1024MAK]

Martin, some more data for you...

Code: Select all

                                    DT-4000ZC        Fluke       DT-4000ZC       Fluke
Computer   Board Serial number    Pin 10 Pin 10  Pin 10 Pin 10 Pin 13 Pin 13 Pin 13 Pin 13
           issue                  normal rev.    normal rev.   normal rev.   normal rev.
BBC B      ???   01-ANB04-6011871 1.140k 1.193k  1.146k 1.189k 1.121k 1.170k 1.125k 1.167k
BBC B      7     03-ANB01-3088426   980R 1.002k    983R 1.001k   992R 1.013k   994R 1.012k
BBC B+ 64K       01-ANB51-3504713   856R   871R    859R   870R   856R   870R   858R   870R
BBC B      7     01-ANB01-3023119   957R   974R    970R   974R   958R   975R   961R   975R
Master 128 2                         10k    10k     10k    10k    10k    10k    10k    10k  
Electron with Plus 1              2.182k 2.183k  2.184k 2.186k 2.177k 2.178k 2.179k 2.181k
The Beeb where the board issue is shown as ??? may be an issue 4 or an issue 7 (I think).

Everyone above, you did know that the test is only valid if you are using the correct coloured test leads, don't you?

Analogue port fire button resistance testing

Mark

[MartinB]

Wow!

Thanks Mark! I’ll have a digest and comment later....

[Elminster]

Perhaps we should synchronise multimeters.

Mine is an evtech 330.

I have a Maplin gold somewhere but I don’t think I need to confirm my results

[MartinB]

Mark - first off, I also have the excellent DT-4000 and a Fluke (the respected Model 12) and the two always read close enough for the delta not to matter.

All the Beebs look as I would expect, we’re just seeing the tolerance of the external 10k pull-ups working in parallel with the 6522 internal pull-up. The Elk doesn’t use a 6522, the joystick fire buttons are connected to discrete logic gates so we’re simply seeing the 2K2 pull-ups. The Master though does puzzle me... it’s schematic is identical to the Beeb but it’s as if the 6522 in yours doesn’t have internal pull-ups.....

[1024MAK]

I tested the Master 128 and the Plus 1 out of curiosity. I was not surprised by the results for the Plus 1, but was by the results on that Master.
I have some other Masters that I can test later.

Mark

[MartinB]

Ok, thanks Mark. Maybe you’ve previously robbed the 6522 from that Master for another fix because it’s almost as if it doesn’t have one fitted!

[Elminster]

A dead 6522? I will see if I get 5 mins tomorrow to test a master as well.

[1024MAK]

Ok, thanks Mark. Maybe you’ve previously robbed the 6522 from that Master for another fix because it’s almost as if it doesn’t have one fitted!

Hah ha! You should know that pinching bits from a Master 128 is most likely too much like hard work, as in most Masters, everything except the ROMs are soldered in... (and I'm too lazy to go to that much effort when I have some Beebs to borrow bits from).

No, to the best of my knowledge, the Master 128 that I tested and give results for, is fully operational (it's also the only Master that currently has a memory/RTC/CMOS back-up battery fitted...).

So, anyway, I tested two other Masters, taking into account resistor tolerances, they all read the same - that is 10k.

So I investigated the first Master 128. Photo here of IC8:-

G65SC22P-2 in position IC8 in a Master

(G65SC22P-2 in position IC8 in a Master)

So that looks like a CMOS type. Presumably then, it does not have an internal pull-up resistance.

Mark

[1024MAK]

Yeah, it's a CMOS type. Datasheet available from here

Mark

[tricky]

...Everyone above, you did know that the test is only valid if you are using the correct coloured test leads, don't you? ...

So, should we swap the meter end or the clip end when checking the reverse or both

EDIT: Sorry, OT.

PS Is it only issue 7 beebs that have the support in the middle of the vent slot?

[Elminster]

PS Is it only issue 7 beebs that have the support in the middle of the vent slot?

My issue 4 has no support, my b+ does. Issues 7s not to hand.

Results for a b+

13 - 1.275k
10 - 1.278k

Will dig masters out tomorrow

[1024MAK]

Rockwell R6522 Port B

G65SC22 Port B

Mark

[Elminster]

Someone stole the squiggle snake in the cmos one

[MartinB]

Don't be silly Duncan, it's not a squiggle snake, it's clearly a spring....

Thanks again Mark, that of course answers the Master query. The higher effective value isn't actually going to be a problem but it is valuable information for me to be aware of.

No need for any more testing then folks, I have the full story now and minor variations are not significant.

Thanks again all

[Kazzie]

Don't be silly Duncan, it's not a squiggle snake, it's clearly a spring....

Or maybe it's neither, and the wire just does this?

[MartinB]

Heh heh