23rd Jan 2021, 8:58 pm | #701 |
Octode
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
|
Re: MK14 schematic revisions
Well I tried creating objects and placing them, and it was taking ages, and I wasn't sure how they'd look when printed because I couldn't get them consistantly anchored. Maybe I should try creating it as a Draw document and inserting it, at the time I couldn't see how to do that... If I get any progress maybe I can email it to someone and see how it works for them when printed out! (I don't have access to a printer at the moment).
|
23rd Jan 2021, 9:03 pm | #702 | |
Octode
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
|
Re: MK14 schematic revisions
Quote:
|
|
23rd Jan 2021, 9:10 pm | #703 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 schematic revisions
Maybe just replace your 'bounciest' bouncy key and see if ithat effects a cure, then you'll know whether it is worth replacing the whole set.
|
23rd Jan 2021, 9:15 pm | #704 |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,298
|
Re: MK14 schematic revisions
I can try printing, I need a set of labels anyway and also interested in using an overlay for the lower profile buttons for the next build.
|
26th Jan 2021, 11:30 pm | #705 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 schematic revisions
You never know, someone somewhere may already be beavering away on a keypad overlay for the issue VI. Stranger things do happen.
Another info update, I've been running my issue VI with the 27S13 PROMs in for most of the night and I'm struck by how much cooler they seem to run than the other manufacturer's types do. They still get warm, but not the '...How is that still working?' kind of heat that you get from a DM74S571. |
26th Jan 2021, 11:43 pm | #706 |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,298
|
Re: MK14 schematic revisions
|
26th Jan 2021, 11:50 pm | #707 |
Octode
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
|
Re: MK14 schematic revisions
|
26th Jan 2021, 11:59 pm | #708 |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,298
|
Re: MK14 schematic revisions
Received the 7445s today, turned out they were nat semi DM7445N.
Tried to verify the key bounce again with the 7442 fitted. At first power up there was no problem with 8 and A keys, but its a bit cool in the basement today. After about 5 minutes the double key returned. Suspect the output drive is a little stronger when the 7442 is cold. Swapped in the DM7445N and no more problems with double keying on 8 and A, even after running for 30 minutes, but don’t have a lot of time to play today. Also received 8MHz and 9MHz crystals and the 8060 seems quite happy to run the SoC monitor at 9 MHz. Is there anything like zxall for the 8060? I may try the 8MHz on the 8073, baud rate at 9600 should be ok. I still find the display brightness variation between digits a bit annoying. I took a quick look with the scope and the digit cathode drive timing seems to depend on the digit displayed and also on which key is pressed. So this might be a limitation of the monitor software rather than the display module. |
27th Jan 2021, 12:06 am | #709 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 schematic revisions
I have to tell you I don't get this at all so I wonder if this is more a characteristic of your display than the hardware driving it. The display I'm using is a Nat Semi NSA1488 harvested from a Texet 880 calculator and the displays are always evenly lit and quite bright. I should say that mine is never overclocked, however.
|
27th Jan 2021, 12:14 am | #710 |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,298
|
Re: MK14 schematic revisions
I may try writing a simple program to drive all segments and all digits with constant timing, just to be certain. Timing of adjacent digits showing the same number didn’t look different enough to explain the brightness variation. There was no change in the weak digit when I changed the 7442 to 7445. Also don’t remember noticing this on my original.
|
27th Jan 2021, 12:19 am | #711 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 schematic revisions
Where did you get your 7408 segment drivers from? I wonder if it would be worth swapping your original pair in from the old lady just for a quick comparison.
|
27th Jan 2021, 12:32 am | #712 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 schematic revisions
If you read way back, Tim had a Russian-sourced display which arrived with a number of dead display cells. I mentioned that I had 'rescued' a dead cell in the display of my Sinclair PFM200 frequency meter by applying some heat to the solder tab of the common connection for that display cell and it sprang to life and has worked ever since.
With nothing to lose, Tim then tried that on his display and managed to get some (but unfortunately not all) of the dead cells on his Russian display working. The nature of the problem suggested that there was some kind of resistance in the common connections of the dead cells (since it affected the whole cell and not just one segment in the cell). If you've made your display unpluggable try unplugging it and testing the individual segments using the 'Diode Test' feature on your DMM - normally the current / voltage available on that range will light the segments dimly so it is a fast way to check out that sort of display, but in your case I would be interested to know whether the common-to-segments forward voltage drop is different on your 'dim' cell to what it is on the others. |
27th Jan 2021, 3:23 am | #713 |
Pentode
Join Date: Jun 2012
Location: Bristol, UK.
Posts: 115
|
Re: MK14 schematic revisions
You've eliminated IC13 as the 'digit dimmer'.
I suggest checking the physical PCB/soldering/etc. With MK14 off, link IC13 actual pin to the actual cathode point on the Display to make a direct parallel connection. Circuit diagram shows IC13 pin 5 and Display 10 is 'dim'. Power up and if the digit stays bad then it's likely the actual Display is at fault. When the display is showing a line of the same digit (usually 0) the Monitor code is in a tight loop, giving each digit exactly the same 'time slot'. There is a *very* slight variation for different digit values because the loop contains a DLY which is fed the segment-pattern by the Accumulator. And a key-press distorts the loop and extends that digit 'slot'. |
27th Jan 2021, 3:42 am | #714 |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,298
|
Re: MK14 schematic revisions
Using DM5408J for segment drivers, sourced as used from Utsource, so maybe not guaranteed genuine, although I haven’t had any problems from Utsource yet.
As its variation between digits, rather than between segments, I did check the outputs of the 7445 but didn’t see any weak pull down of the digit drivers. I probably need to check digit 3 and 4. It probably is the display, maybe just not as bad as Tim’s. I’ll try and capture the cathode and anode voltages on the scope, see how that compares between good and bad digits. |
27th Jan 2021, 8:40 pm | #715 |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,298
|
Re: MK14 schematic revisions
Attaching picture of DM7445N for comparison as requested.
|
27th Jan 2021, 8:48 pm | #716 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 schematic revisions
That looks very realistic to me, bog standard Nat Semi marking with the early 'double wave' logo and the vertical white stripe across the end. Both of mine (In original MK14 / In issue VI replica) are DM7445Ns like that.
Another thing you could try (if your display is removable) is to make a short extension cable and reverse the order of the display commons so that cell 8 displays what cell 1 should, cell 1 displays what cell 8 should etc. (Or at least swap the cathode of the 'dim' display over with that of a 'bright' one). If your 'dim' cell moves, the display driver hardware hardware is the cause, but if your 'dim' cell stays where it is, the display is the cause. |
28th Jan 2021, 7:24 am | #717 |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,298
|
Re: MK14 schematic revisions
I’m sure its not the 7445. I posted the pic because circuitryboy wanted to compare the markings on mine with his, I wasn’t sure how to attach the pic to a PM and it might be a useful reference for someone in the future.
The display is the prime suspect, but still interesting to diagnose in more detail. I did try re-soldering the cathode of digit 5 which seems to be weak, but no change. As the display is connected with header pins I might just displace the connector two pins to the right. It will screw up the segment mapping but give a quick test if the weak digit moves or stays the same digit position. |
28th Jan 2021, 10:03 am | #718 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 schematic revisions
For future reference the forum does not, as far as I know, allow images or other files to be shared via PM. Good idea about displacing the display by two pins - shame it isn't a 9-digit display.
The 'heat the cathode' fix wasn't aimed specifically at the solder joint exposed on the lower edge of the display although that was where it was applied. The trick was to apply heat for long enough so that it travelled up to the little gold wire connecting the pad to the actual display cathode and hopefully 'reactivated' it, which did work for me originally and several times for Tim, although I don't pretend to understand the reason it works, or sometimes works. You'll be aware that common wisdom dictates that you should not apply too much heat for too long to the leads of LEDs or any other semiconductor for that matter, so knowing how much heat is 'too much' is just a matter of fine judgement, unfortunately. If you try it it might be an idea to pop the 'bubble window' off the front of the display in case that suffers heat damage during the attempt. They are generally just held in place by melted 'pips' on the rear side. |
28th Jan 2021, 7:13 pm | #719 | |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 schematic revisions
Quote:
Not had time to do much with it other than fit it. PicoBASIC runs OK with a simple test so the extra Memory is good - not tried the Tape Interface yet as I need to make a lead up. I did see the dreaded Reverse C with the OrtonView but, a Reset took it away and it has not reappeared and all the test programs including MINEFIELD and CHARSET worked. I also tried the CLIVE picture at 0x200 and it worked fine. I did wonder if the memory would give trouble with the original SoC VDU due to the read issue like on the 6561's but, CLIVE appears fine etc - in main memory as I cannot easily change memory address of Video RAM on the SoC VDU. The keyboard frame sits nicely over my keys with the caps left on and really makes it look good and operate well still with a nice click - the yellow matches the colour of my smoothing cap but, that is hidden by the memory board so a RED under would be a nice hint to match the memory board. Obviously black PCB's for the memory board and my Display would be perfection... Any higher than the small CAP I used will stop the board on an MK14. Overall happy so far. |
|
28th Jan 2021, 7:22 pm | #720 |
Octode
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
|
Re: MK14 schematic revisions
I just ordered the key frame - I already have plans in the works re: memory. I exchanged messages with the chap, seems like a nice guy, According to one of his eBay listings he was selling zx80 stuff to finance his MK14 build aparrently - it appears he did it! Said he might pop in here for a chat sometime so keep your eyes peeled
|