Micro UK101 Build

digidice · Post by **digidice** » Fri Apr 08, 2011 12:40 am

I just purchased a Rigol DS1052E (about a month ago) and am Extremely happy with it, Being a digital scope it can record the signals so you can freeze them. small lightweight and wasn't too expensive. I still have a BK 2120 dual trace but haven't even turned it on since I got the Rigol. Granted it is rated up to 50Mhz but there is a mod on the net that shows how to take out the front end filter and make it go to a claimed 400Mhz.. I haven't done the mod as it suits its purpose right now.

I rarely have to look at any signals faster than it goes, for that I pull out my HP service monitor and plug that beast in.

micro_brain · Post by **micro_brain** » Fri Apr 08, 2011 10:45 am

OK. Have built the NOP generator and plugged it into the ROM. With a piezo sounder at pin 15, there is no buzzing at all, but it does click when first connected. Does this indicate anything in particular? I tried several other address lines but same thing, a click. Almost like the lines are high all the time.

Have to check all connections again...

BigEd · Post by **BigEd** » Fri Apr 08, 2011 2:51 pm

Is that pin15 of the 6502? Is that going to be toggling at an audio frequency? It might be worth trying some higher address lines.

micro_brain · Post by **micro_brain** » Sat Apr 09, 2011 12:43 pm

BigEd wrote:

Is that pin15 of the 6502? Is that going to be toggling at an audio frequency? It might be worth trying some higher address lines.

Sorry, I meant A15 but I tried various othr address lines too, all the same result. According to Lee's page where he discusses the NOP generator, I should be able to hear buzzing on A15.

Maybe not all the time though- what happens when the CPU reaches the end of memory 0xFFFF? Does it wrap round to zero?

BigEd · Post by **BigEd** » Sat Apr 09, 2011 1:18 pm

Yes, it will keep going around. So this isn't a good sign. First stop is perhaps the clock, and then the RDY input, and the RESET IRQ and NMI inputs. They should all be high for normal operation, unless I'm terribly confused.

If that's all OK, then you'd fail to send NOPs in if your databus had contention, which would be the case if you had other devices driving the bus - time to check the chip selects and address decode, perhaps.

Cheers
Ed

GARTHWILSON · Post by **GARTHWILSON** » Sat Apr 09, 2011 7:05 pm

Quote:

First stop is perhaps the clock, and then the RDY input, and the RESET, IRQ, and NMI inputs. They should all be high for normal operation

BE and ABORT\ also. Looking at the schematic now (I apologize for not taking time earlier), I don't see anything pulling these up (except ABORT\, and RST\ on the D1813-10 which I assume is a complete reset circuit).

micro_brain · Post by **micro_brain** » Mon Apr 11, 2011 7:24 pm

Hi Garth,

D1813-10 - What is it? The push button is the reset switch (has no effect on my board so far as I can see). I am sure the clock circuit is doing something as it is causing LW interference with a nearby radio when turned on. I was not successful in buying the scope from eBay - got outbid whilst on holiday with a very poor mobile broadband signal. Apart from that I should buy a logic probe. And recheck EVERYTHING.

Incidentally, the CPU is a Rockwell R6502AP - I have assumed that this is compatible. I bought it from an eBay seller in the Netherlands. Th elisting states:

1 PC. Rockwell R6502AP 8-bit CPU / MPU R6502P 6502 DIP/DIL

NMOS CPU/MPU 8-bit microprocessor

Brand: Rockwell
Technology: NMOS
Type: 8-bit Microprocessor
Clock frequency: 2MHz
Temperature range: 0 to +70°C
Package: Plastic DIP/DIL 40

Thanks all

JonB

GARTHWILSON · Post by **GARTHWILSON** » Mon Apr 11, 2011 8:15 pm

I mentioned the ABORT\ input, but that's on the '816 but not the '02. WDC's 65c02 has the BE input which I also mentioned, but you don't have that if you have a Rockwell part. I see now I was confusing digidice (who posted this schematic on page 2 of this topic) with micro_brain. My apologies-- I'll have to try harder to get our new members sorted out.

So are the inputs that Ed mentioned getting pulled up now, as they should be? And is the RST\ line high also after the initial quick low pulse? This has to be high also for the processor to run.

Side note: I know "Gareth" (with an e in it) is a common name in the Brittish isles, but my name has no e.

micro_brain · Post by **micro_brain** » Mon Apr 11, 2011 8:32 pm

Duly noted, Garth - and post edited (I know how you feel - people always spell my name - Jon - with an 'h'.)

I will have to look at the circuit again when I have some time. I reviewed the thread and discovered that the NOP gen should be installed without the ACIA and could go in the RAM socket too (maybe that's best?). Not sure what the thing will do if it's in ROM and the RAM is empty (or vice versa). Will it cycle?

Karatorian · Post by **Karatorian** » Tue Apr 12, 2011 5:10 am

micro_brain wrote:

Not sure what the thing will do if it's in ROM and the RAM is empty (or vice versa). Will it cycle?

Well, the first thing it will do is go to the reset vector 0xEAEA. Then, if that's in the ROM area, it'll start executing NOPs until it reaches 0xFFFF. At that point, the program counter wraps around and it'll try to execute what's at the bottom of the RAM, doing who knows what. So no, it won't really work.

However, there's a chance (but no guarantee, floating inputs being iffy), that it'll find a BRK (0x00) at the bottom of RAM and go to the IRQ vector (0xEAEA again). So it might sorta cycle.

Furthermore, if there's any space in the memory map for IO (which I assume there is), a NOP generator (even in both ROM and RAM), won't cycle either. Instead, once it hits the IO space, it'll start trying to execute whatever it finds there. So unless you can hack the address decode select either the RAM or ROM instead of IO, it won't work.

leeeeee · Post by **leeeeee** » Tue Apr 12, 2011 3:37 pm

Quote:

... the NOP gen should be installed without the ACIA and could go in the RAM socket too (maybe that's best?).

The RAM, ROM and ACIA should all be removed and just the NOP generator installed.

Quote:

Will it cycle?

Yes, it cycles continuously through the entire address range.

Lee.

micro_brain · Post by **micro_brain** » Tue Apr 12, 2011 7:09 pm

leeeeee wrote:

Quote:

... the NOP gen should be installed without the ACIA and could go in the RAM socket too (maybe that's best?).

The RAM, ROM and ACIA should all be removed and just the NOP generator installed.

Quote:

Will it cycle?

Yes, it cycles continuously through the entire address range.

Lee.

And the discreet logic chips too?

Dr Jefyll · Post by **Dr Jefyll** » Tue Apr 12, 2011 8:30 pm

micro_brain wrote:

And the discreet logic chips too?

No, just remove the chips (except the CPU) that connect to the data bus.

Normally the data bus is a shared resource, and the RAM, ROM and ACIA wait for a cue -- an address -- from the CPU before they try to put anything on the data bus. The NOP generator breaks the rules; it ignores addresses and never shuts up, "hogging" the bus. That's why you remove the others; they can't talk over the NOP generator, and during this experiment you don't need them anyway. Btw in this case the CPU listens to the bus but doesn't talk -- that's why it's OK to leave it in. Good luck, and have fun!

-- Jeff

BigDumbDinosaur · Post by **BigDumbDinosaur** » Tue Apr 12, 2011 8:53 pm

I don't recall reading it anywhere in the discussion, but have you used a logic probe to:

1) Verify you have a clock pulse at the MPU's Ø2 input?

2) Verify that you have a clock pulse at the MPU's Ø2 output, and if using it, the Ø1 output?

3) Verify that /RESET is high after you've reset the unit?

4) Verify that /RDY and /BE are both high? If either is low or floating the MPU will not do anything for you.

Other things to look for include accidental cross-connects of bus leads, accidental grounds and open circuits, etc. Last but not least, do you know for certain that your MPU is good? If all of the above are good then the thing should run with the NOP generator. You also verified that you correctly wired your NOP generator, right?

micro_brain · Post by **micro_brain** » Wed Apr 13, 2011 7:54 am

I rechecked the NOPgen, it had a miswire. Darn it. Also, I used the wrong resistors in the XTAL circuit. Dagnammit again.

So, fixed NOPgen & clock circuit. Verified that all data lines are showing correct hi / lo values to spell out NOP value. Ran board with radio nearby and can hear the interference from the clock circuit as before. Still no tones from any of the addr lines.

Will check the lines as suggested (using a multimeter).

In the meantime, have ordered a logic probe, which should be here by the weekend.

Incidentally, the MAX232 output is something like 2.5v on Tx line (assuming it is idle).

I've no way to test the CPU