Re: Ortonview PCB
 

It strikes me that a more realistic first 'Killer App' for the MK14 + VDU would be a clone of 'Pong', entirely in graphics mode with the score rendered as pixel digits. The noise generator would be useful for randomizing the insertion point and initial angle of travel of the 'ball'.

We just need to get the basic VDU stuff working absolutely flawlessly first, then we can start to have some fun. I doubt we'll ever get ELITE running on the MK14, but we should be able to come up with something vaguely amusing.

Re: Ortonview PCB
 

A circuit query regarding the circuit as given in the project document - I notice the NWDS pullup resistor (R3) is 10K - when Tim and I were looking around for ways and means to observe and cure the corruption problem, one thing we did find was that raising the value of the NWDS pullup to 10K made the corruption problem happen more 'reliably', in order to observe it more easily.

If you (Slothie) are still getting occasional flickers in the display cell one step to the left of the data field (so, third one in from the right) even with the four capacitors fitted on A8-A11, try taking that resistor down to 4K7, which is actually the recommended value for the NWDS pullup in the original VDU docs anyway, and the value I had fitted all the time I was playing with my original lash-up version of OrtonView.

220pF capacitors still have not arrived yet, alas. To disable the battery-back feature and just use my 6116 as 'normal' RAM I take it I can either fit a schottky diode (which I happen not to have) in D2 or just link out D2 for now, link pins 2 and 3 of the write protect switch and don't fit the battery or D1. Did I miss anything?

Re: Ortonview PCB
 

Just going through the bits that arrived today to see if I can build mine yet - I only have a standard 6116 as well.

Re: Ortonview PCB
 

My 6116 is an ...LP, but I think Slothie has found the RAM retention feature does not work as he hoped, so he was looking at redesigning that, until we distracted him with talk of QWERTY keyboards.

Once I have the 220pF caps the only thing I will be short of is the correct Phono socket but I have a leaded one which I can use in the meantime.

Am I right about the NWDS pullup resistor? You tended to have yours at 10K, only with the actual intention of causing trouble?

Re: Ortonview PCB
 

No thats about it,

As for the NWDS, when the write protect switch is closed there are 2 10k resistors (R1, R3) in parallel pulling up NWDS, making 5k. I didnt want to make that any lower.

Re: Ortonview PCB
 

Ah, right. Changing the value of that pullup resistance may have unintended consequences, as observed earlier. The value seems important, or to be more precise, based on earlier observation, it ideally needs to be no more than 4K7.

If you remember, the capacitors appear to fix some sort of timing relationship between the rate of change of the states of A8-A11 and NWDS but raising the value of the pullup resistor (thereby slowing the rate of change of NWDS) actually cancels out some of the effect of the capacitors.

Re: Ortonview PCB
 

Mine is 4.7K with a changeover switch to 10K to make the problem appear - my caps are only 47pF on mine (I cant remember if the 10K makes it appear with the the 47pF I will try again) but, I have some 220pF arrived today so will build this board with them - only four are needed right? Links on the others?

I am also not clear what other track to cut before soldering the sockets on - I have the first diagram but, maybe missed something in the later discussion on the battery etc.

Re: Ortonview PCB
 

Well there's no reason not to change R3 to 4.7k, so that the pullup is always at least 4.7k. With the WE switch closed it will only go down to about 3.2k which isn't going to cause a problem. I will change the schematic appropriately.

Re: Ortonview PCB
 

NO links on the others, unless you have a reason to want to short A0-A7 to 0V. Just fit the ones on A8-A11 and only if you are NOT intending to try the buffers as it's either / or. (Buffers + mod to the buffer enable circuit, OR direct connection without buffers and 4 x capacitors on A8-A11)

If you are not going to use the buffers you need to link across each of the 6 gates in each buffer to connect the system address bus directly to the PIC pins as they were in the original prototypes. If you have fitted turned pin sockets for the buffers you can just plug little hoops of solid core wire into the sockets.

Re: Ortonview PCB
 

Tim, I thought I remembered you having raised your caps to a higher value than that - you don't have others fitted in parallel with the 47pFs, maybe on the other side of the PCB?

Re: Ortonview PCB
 

Yes, dont fit load capacitors where not needed, leave the locations empty if not required. I only put in 12 in case I needed to experiment with them, because it was easier to put them all in than ask if they might be useful.

At the moment the buffer option isn't working, there is an idea to see if we can fix it (see the image in this post) but as yet it is untested.

Re: Ortonview PCB
 

I've already done the buffer enable mod (minus one needed capacitor) so I will be trying the buffers option first.

Re: Ortonview PCB
 

I'm trying a different type of Phono (video out) socket. You can find them by searching on this part number.

TruConnect LP-0841-3Y-05A

No idea if they are a good fit for the pinout in Slothie's OV PCB, I just like the idea of having the socket colour coded yellow (for video).

Re: Ortonview PCB
 

Groan, it's back to front isn't it? The single 'foot' needs to be at the back. Back to the internet...

Re: Ortonview PCB
 

From the look of the data sheet it will fit into the holes, but the pinout is different. However the tracks can be cut and re-connected with wire links easily.

Re: Ortonview PCB
 

The "foot" at the front is GND on both the PCB and the connector, the left and right connections are GND on the PCB and signal connections on the connector. You could cut the traces connecting the gnd to the pins and re-route things.
If you're buying from Rapid then the "RVFM Black PCB Skeleton Phono Skt" and the red one will fit. If from Farnell or someone else then the Cliff FC68391 works.

I got mine from eBay, and naturally I looked just now and they are nowhere to be seen and its full of the truconnect style! Looks like I made a poor choice (again),

Re: Ortonview PCB
 

On your Cliff type socket it looks to me as though the centre contact pin is set furthest away from the plug entry hole, whereas on the one I described in #163 I think the centre contact pin is offset towards the plug entry hole. I'll see when it gets here.

I see Farnell etc appear to have your original recommended Cliff part, so I may have to get it from there. In the meantime I have an inline socket dangling on the end of a short bit of slim RF Coax - I can use that.

Edit: Thanks for the further info. Main mistake I made was deciding to order a connector when I do not have the PCB to hand - it is at work at the moment.

Re: Ortonview PCB
 

Yes, on the socket I used the pin furthest from the hole is the signal and the other 3 are ground. The trueconnect one has ground nearest the hole but the other 2 are connected to the signal, one switched the other unswitched. It has no "rear" pin.

Re: Ortonview PCB
 

I did for testing purposes but, when I finally fitted them to the top I dropped back to the lowest value that worked on the basis of simplicity.

Re: Ortonview PCB
 

I can send you one if you want -the minimum order was 10 from Farnell of the cliff non colour coded ones!

Re: Ortonview PCB
 

Hmm, that's a huge diffference from the value (150pF or greater) that I had to use to get to the same point. The optimum value of the capacitors could also vary depending on what else is on the A8-A11 lines. This is why I want to get away from the capacitor bodge if possible, to arrive at something which works for every one, every time.

I thought of another interesting issue as well, specifically relating to Mark. He's in a 60Hz country and the original SOC VDU and Ortonview are both designed to produce a 50Hz framerate. My Philips CM8524 monitor will run at 60Hz although the picture flattens a bit, but I think the ability to tolerate the difference will vary from display to display.

Modern flatscreens will probably cope with it quite well as they essentially rescale whatever they detect to fit on their screen. I imagine most will cope with 50Hz or 60Hz equally well.

One thing Mark could try is to find a crystal (or signal generator) with a frequency of around ((16/5) x 6) MHz = 19.2Mhz, which I think would produce a framerate of around 60Hz. Luckily 19.2MHz is comfortably inside the 20MHz maximum clock frequency of the 877 and it seems to be a commonly available off the peg crystal.

Who knows what issues this might raise, though. This is just one more reason why we need Mark involved in this phase of testing.

Re: Ortonview PCB
 

Oh, yes please.. you have a PM.

Re: Ortonview PCB
 

One problem Mark may find is that the US 60Hz standard only has 525 lines per frame rather than the 625 of the 50Hz standard. It might be possible to tweak the firmware to allow for this however. I'm not sure how it affects the line time either. But it would be unfair if Mark didn't have any challenges to deal with!

Re: Ortonview PCB
 

I think virtually all modern screens will have firmware and chipsets which travel all around the world and are equally happy with the 50Hz / 625 line and 525 line / 60Hz standard, and a lot of the late generation CRT TVs and monitors probably do too.

At least the VDUs (SOC and Ortonview) are mono-only so PAL vs. NTSC is one thing we won't have to worry about.

Re: Ortonview PCB
 

I brought a small Samsung lcd tv to Canada from the Uk, as it runs on 110 to 240v and supports both PAL and NTSC, as most uk tvs do. The advantage of a company paid move was I just brought everything, including boxes of magazines going back to the 60s.

Unfortunately Canadian TVs don’t work on PAL, so my old LG DVD player only worked in monochrome on the big TV even though I had the region unlock code, but it stopped working last year and I haven’t tried to get it working again yet. Suspect it might be bit rot, but could just be battery backup.

I use the small TV for experimenting with the TMS9x18 and was planning to try with the 29 version at some point. I’ll use it for the Ortonview but I can also try on an NTSC only TV later.

Re: Ortonview PCB
 

Briefly OT, does your DVD player have RGB-out and the TV have RGB-In? You can bypass any PAL / NTSC colour encoding problems that way.

Unfortunately I think 'Component' (YPBPR) video has always been more popular than RGB stateside and much less commonly available on European video equipment, so you could find that while the DVD player only has RGB-out the TV only has 'Component' in.

Re: Ortonview PCB
 

It has composite, s-video and scart. I’m not sure what features are supported, I think I tried RGB, but not sure now if that was with UK or Canadian TV, I think I blamed the SCART cable I picked up cheap in a pound shop one time I was visiting the UK. Maybe it was RGB to Component issue. It doesn’t power up correctly anymore, after about 30 second delay with blank display from the standby light changing from red to green the display shows all segments on. Its an LG DVD-3200E, if anyone has any idea what the fault could be.

Re: Ortonview PCB
 

Probably one for the 'Modern Technology' section, that's where the people who still work on current and recent tech mostly hang out. A very helpful bunch.

'Component' normally uses phono output and input connections colour coded Red, Green, Blue (which is what you would expect RGB to use, but never mind) - as opposed to the Yellow / Red / White of "AV" (Composite video +stereo sound).

That explains how your MK14 went along for the ride. I've lost count of the number of times I've heard that people lost or discarded them in house moves, so yours was extremely lucky to make it past that point.

Re: Ortonview PCB
 

I wouldn’t of parted with the MK14, though it had been stored in less than ideal conditions, a jiffy bag in a damp garage for about ten years, it survived in remarkably good condition. I did unfortunately get rid of a few things that I wish I had kept, including an old phillips MDCR that probably only really needed a belt, an old dot matrix printer that I couldn’t find print ribbon for and an old tektronix D10 scope.

Re: Ortonview PCB
 

I found some SMD 220pF capacitors, so I soldered one in place in the buffer enable delay mod and brought the board home. I decided to try it straight off, with buffers and extra RAM fitted. Unfortunately it doesn't look good on first try, there is a lot of noise on the displays between the address and data fields and the VDU itself is producing an abormal - height screen.

So I removed the PIC and the buffers leaving only the 6116 and 7402 address decoder in place, powered up to find a normal MK14 display and found that the memory by itself works fine, with any data entered over the range 0200-07FF 'sticking' as it should.

Next, I'll try powering the fully populated OrtonView from 5V-only - in those circumstances it should just display the random start-up content of the 6116 but should otherwise give me a normal 16 x 32 character VDU display output.

Re: Ortonview PCB
 

Powering the populated board from just 5V, things look a bit more normal. Next step will be to try it in V1 Ortonview configuration, with buffers linked out and four capacitors fitted. I'm expecting the through-hole 220pFs to arrive tomorrow, so I will continue then.

Re: Ortonview PCB
 

Hmm seems the buffer idea needs a lot more work then.

Re: Ortonview PCB
 

Unfortunately it affects the MK14 so badly when connected that I would be reluctant to try to faultfind that way. As said, I'll make sure it still works in the original way with capacitor fix just so I know that nothing is broken.

The 'normal' MK14 switch on prompt is there at switch on just with a lot of noise on the displays, but it crashes within a second or so. It might just be that the buffer enable delay time needs to be a little bit longer.

Re: Ortonview PCB
 

I've just tried manually 'slugging' another capacitor across the existing 220pF - the MK14 keeps running, albeit still with a lot of noise on the displays, and I can step through addresses with the 'Mem' key. However the addresses sometimes increase in large steps by themselves each time I press 'Mem'. I think there's still a chance that the right delay could fix it, but it might have to be more precisely timed, ie, by the PIC rather than by external timing.

Re: Ortonview PCB
 

Well its an improvement over having no delay at all, the display was badly corrupted and sometimes there was nothing on the display at all. The PIC doesn't attempt to use the bus for some time after asserting NENIN so their certainly is scope to increase the capacitance (delay) considerably.

Re: Ortonview PCB
 

How much time is 'some time', do you remember?

Re: Ortonview PCB
 

What value resistor did you use for the RC delay to the Buffers?

With two 74ls ttl inputs on the output of the RC delay, the maximum resistor value is probably 1 k to make sure it can pull the inputs down to less than 0.8 v. A lower value, say 470 ohm might be a better choice, but then the capacitor needs to be a higher value.

I was never much good at calculating RC timing values, I think you want to aim for about 1.5 to 2 us. Based on 8060 memory cycle time at 4MHz.

Re: Ortonview PCB
 

It was 3.75uS according to my calculations (15 machine cycles). I think 1K & 220pF should give a delay of about 220nS so 1nF should give you 1mS?

Re: Ortonview PCB
 

I suggested a 1k resistor, so C should be 1.5-2nF, looks like I suggested a value far too small.

Re: Ortonview PCB
 

Slothie suggested 1K + 220pF so that's what it is at the moment. I stuck an additional capacitor marked '470' across the 220- it's an old British capacitor so the chances are that the value really is 470pF and not 47pF (a more modern 470pF capacitor would be marked '471').

I'm wondering if it would be more productive to knock up a proper monostable delay using a 74121 / 74122 with a multiturn pot to allow free adjustment of the delay in very small increments and generate the delay with that. If that gives us a working delay range we can try to fall back to RC generating the same delay.

Re: Ortonview PCB
 

Surely you mean 1uS? (1000nS in a uS?)

Am I looking at more like 2.2nF then, based on Mark's suggestion of 2uS?

Re: Ortonview PCB
 

Well, just tried a '222' (2,200pf / 2.2nF) and the situation is much the same, unstable and crashing.

I really will need to try it in the original mode just to make sure the PIC and memory are still OK. The 365 buffers were brand new from RS or Farnell a while ago and were actually tested in my SOC VDU as 80L95 substitutes.

I'll have to pause until the 220pF through hole caps arrive, hopefully tomorrow.

Re: Ortonview PCB
 

So I think we need to aim for a delay between 2us and 3.75us, an RC on ls ttl inputs might not be that accurate without selecting the values on test.

Maybe safer to use too large a value instead of too small to avoid bus conflict, so perhaps a 3.3nF and see what that gives, then trim with the resistor.

Re: Ortonview PCB
 

Noted, although 2.2nf seems no better than 220pF+470pF - as said, I think I should first verify that it operates normally in the original way with your capacitor fix on A8-A11 before trying to get something new working.

Re: Ortonview PCB
 

You could try isolating U3-6 from R12 and NRDS using a schottky diode and then connect the second gate pins of the buffers to U3-6 instead of using an RC.

Re: Ortonview PCB
 

Noted as well, but I'm still going to go back to basics first to ensure that I still have a fully working PIC, etc. I need to start from a known good datum point, I think.

Just think, Mark, you have all this fun to look forward to, although hopefully we will have it all worked out by the time your board gets to you.

Re: Ortonview PCB
 

Yes I meant uS not mS. Tired typing!

Re: Ortonview PCB
 

Well I confirmed my two PIC16F877 sourced from utsource last year can be erased and programmed with my GQ4x4 programmer, which was also first time I’ve used that too. My phono connectors look like they are the right type. DIP switches are piano type, but should fit ok. Not sure about 16MHz crystal, I might have one, but I do have a few 48MHz third overtone that seem to prefer 16 MHz without fine tuning the notch filter.

Still need to solder a connector to the MK14, probably same as Slothie, female header on the MK14 and male on the Ortonview.

Re: Ortonview PCB
 

Another possible method without cutting any more tracks would be to connect UA11 to the second gate input on the buffers, with a pullup resistor, but then only the 6116 ram can be used for ortonview, but we still need to make sure we are not corrupting the MK14 ram.

Re: Ortonview PCB
 

Mine has a conventional 32 + 32 edge connector soldered to the 'fingers', so that I could (If I chose) connect it to my original MK14 for compatibility testing purposes, although I would need to remove the RAM since my issue II doesn't have the memory hole for it to occupy.

Original MK14s don't have Clock-Out on the rear edge connector but, luckily, OrtonView does not need it. Of course I would also need to route the buses and control signals to the underside of the edge connector where the issue II doesn't even have fingers to solder the signals to. So on the whole, highly unlikely.

Have to say the Slothie PCB based version looks a whole lot neater than the danglesome veroboard memory expansion 'bridge' board and the absolute rat's-nest of an OrtonView prototype I had hanging off that, so even if it only ever works in its original way with the A8-A11 capacitor bodge, I'll be very happy with it.

If you didn't want the VDU feature you could probably saw the rear two thirds off the PCB and the memory expansion would still work like a traditional plug in Sinclair 1.5K 'RAM Pack' for the issue VI PCB.