15th Apr 2019, 11:53 pm | #61 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 programming interface - MK2
Hmmm - it is the same brand SGS 74LS365BI also marked 9844I
|
16th Apr 2019, 12:09 am | #62 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 programming interface - MK2
To be exact, this one is marked SGS 74LS365ABI 98441 - I think you probably just missed out the 'A' after the '5' on the one you have there.
Anyway, you'll be able to try it for yourself, I will get it on the way back to you ASAP. |
16th Apr 2019, 12:15 am | #63 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 programming interface - MK2
No actually there is no A so maybe a look at the data sheet is needed or these are re topped devices with a printing error... I have 5 more in transit towards the VDU project
|
16th Apr 2019, 12:27 am | #64 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 programming interface - MK2
It does seem slightly suspicious that two devices would have different main device suffixes and the same batch number.
Your mention of 365s being used on the VDU made me leap across to look at mine to see if it had alternative 365s in it that I could try in the MK14 - but of course the devices in the IC9 / IC10 positions in my VDU are... 80L95s. |
16th Apr 2019, 1:06 am | #65 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 programming interface - MK2
Well tried finding the relevant datasheets but, no joy - the general SGS sheets on 74LS series say the suffix should be B for a DIP - I wonder if it is a different manufacturer - maybe a CMOS part or an original 74365 non schotty even. TI parts had an A after them. I am going with a relabelled chip of unknown exact design due to your observation on the BATCH number...
|
21st Apr 2019, 7:50 pm | #66 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 programming interface - MK2
As a quick update - putting the SGS 365ABI back in my machine allows the PI-Programmer to run at full speed fine although still with the longer reset... go figure. I also tried one of the four TI marked 365AN I have bought for the VDU circuit and that works fine as well. There must be some other issue though as it affected SH so badly. The review of the kit in PCW issue 2 (Apr 1978) lists different values for some passive components - perhaps we also need to look at those - I went with the list in the manual maybe SH Issue II has these older values?
|
22nd Apr 2019, 10:24 pm | #67 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 programming interface - MK2
Just catching up as I have been off grid for the whole of the Easter weekend. Those new results with the original 365 are just bizarre. I've just arrived back but I will take a look at the components on mine, especially in the reset area since they seem so different in that particular area. Why can mine be reset so much faster?
Can you just confirm that your reset opto collector resistor really is 120R, not some other value? If it were bigger then obviously it would take a lot longer to discharge the reset capacitor. I've also just checked that mine really is 120R - since I used SM resistors throughout for mine, it would have been all too easy to make the mistake of putting a part marked '120' (12 ohms) in there instead of '121' (120 ohms). I've also managed to wangle a quantity of TI SN74LS365 sourced directly from RS UK, you are welcome to a handful of those (free) if you want them - the least I can do after all the beta testing you've been doing. |
22nd Apr 2019, 10:46 pm | #68 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 programming interface - MK2
Yes just checked reads 119.5 ohms with a meter
Sounds like a good offer - interesting if you get the same result with one of them in which case we need to look further afield - maybe send the chip back to you as well as it has an effect for you if they don’t. On a lighter note I have patched the programmer into an interface on my conversion of Daver’s fpga MK14 onto my Waveshare Spartan board this weekend! It downloads and plays segtris fine but moonland failed (as does daver’s Built in copy so probably a vhdl bug) - message also transferred fine as well - no reset on that interface though so does not add to this problem solution... |
22nd Apr 2019, 11:49 pm | #69 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 programming interface - MK2
I think Daver... (for the benefit of others here on this forum we are referring to a user on the Vintage Computer Forums who has made a virtual MK14 - still a work in progress - in VHDL) ... Daver would be very interested to hear your feedback on that, or might be even more interested if you fixed it first and then told him.
There is one potential problem that I can think of, and that is the way an offset of -128 is treated by some instructions and not others. For pointer instructions an offset of -128 means 'Use the current contents of the E (Extension) register as the offset', whereas with some others - jump instructions? -128 is just an offset like any other possible value. I can't remember the specifics but I think in the case of some emulations there is an assumption that an offset value of -128 always means 'use the value in the E register as the offset for this instruction', in which case any code which happens to include a jump instruction with a -128 offset will fail. I might not be remembering this correctly - Karen is going to be the expert on this one. Last edited by SiriusHardware; 22nd Apr 2019 at 11:56 pm. |
23rd Apr 2019, 9:31 pm | #70 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 programming interface - MK2
The following is a direct copy of the component list which came with my issue II MK14. Where the components actually fitted are different from the parts on the parts list, I have stated the part actually fitted.
All of the components in this list, save one, are the original Science Of Cambridge kit supplied parts. It is probably fatal to say this, but the machine has always worked, so I have never had to replace any parts. The one non-original exception was the INS8154 - that was bought from Tandy shortly after I got the MK14, but it is an original Nat Semi part. The crystal is in with the 'miscellaneous', mainly hardware parts, which I have not included in the list - it was originally a large-can 4.43MHz TV colour crystal, was replaced for a while with a small-can 4.00MHz crystal (the VDU period) and then reverted back to a large-can 4.43Mhz crystal when I started to return the machine to something more like its original condition. Code:
IC1 ISP-8A / 600 (SC/MP) IC2 DM74S571 IC3 DM74S571 IC4 MM 2111-1N Actual component fitted: uPD2111AL-4 (NEC) IC5 MM 2111-1N Actual component fitted: uPD2111AL-4 (NEC) IC6 MM 2111-1N Actual component fitted: uPD2111AL-4 (NEC) IC7 MM 2111-1N Actual component fitted: uPD2111AL-4 (NEC) IC8 INS8154N IC9 DM74LS157 Actual component fitted: SN74LS157N (Texas Instruments) IC10 DM74LS157 Actual component fitted: SN74LS157N (Texas Instruments) IC11 DM 80L95 IC12 DM 74LS173 IC13 DM 7445 IC14 DM 7408 IC15 DM 7408 IC16 DM 74LS08 IC17 DM 74LS00 Actual component fitted: SN74LS00N (Texas Instruments) IC18 DM DM74LS04 IC19 LM340T-5.0 Resistors R1 4K7 R2 2K4 R3 100K Actual component fitted: 150K R4 1K2 R5 2K4 R6 1K2 R7-R10 1K2 'May be any value between 1K and 15K'. Actual values: All 4K7 R11 4K7 R12-R15 1K2 'May be any value between 1K and 15K'. Actual values: All 6K8 Capacitors C1 27p for 33p (sic). Actual value: 33pf C2 1000uF 40V C3 0.01uF C4 0.01uF (There is no C5 either on the list or fitted) C6 22uF 16V I grabbed 5 of those genuine ex-RS SN74LS365 ICs to try in my MK14 in the IC11 position - I'll try that later in the week. Last edited by SiriusHardware; 23rd Apr 2019 at 9:44 pm. |
24th Apr 2019, 9:28 am | #71 |
Octode
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
|
Re: MK14 programming interface - MK2
Is a 4Mhz clock mandatory for the VDU to operate?
|
24th Apr 2019, 12:59 pm | #72 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 programming interface - MK2
It is the frequency from which the syncs and the character bit (pixel) rate, etc are derived. There would be no harm in leaving it at 4.43MHz to try it, but the image would likely be slightly small and offset, and might suffer loss of hold due to the refresh rate being too far from 50Hz.
On my 'VDU demo' project which used a PIC to pretend to be MK14 RAM, I used a four pin 4.00Mhz oscillator module to supply the 4Mhz clock. |
24th Apr 2019, 6:58 pm | #73 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 programming interface - MK2
I've just tried 5-of ex-RS SN74LS365 one after the other in the MK14 in the IC11 (DM80L95) position and it runs absolutely normally even when I have the reset values reduced to my original defaults.
I will also try putting two of these ICs in the IC9 / IC10 positions in the VDU PCB. This will show whether the 74LS365s are suitable replacements for the 80L95s in those positions. |
24th Apr 2019, 9:14 pm | #74 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 programming interface - MK2
Sorry guys I have been offline as my PC went fut - just getting back under way!
Well now to prove that the chip is the cause of the typing issues for you I need to send it back as it seems to work fine for me and see if you still get the issue... but, All my resistors are 1.2k so my pullups R7-R10 on the keypad rows are 1/4 the value of yours... that also means the pullups on D0-D3 are 1/6th of the size. R12-R15 (marked 11-15 on the revision II diagram) in the top corner on end. They give the xxxx1111 on the KB read I assume. So by my view mine should transition faster - I will swap my pullups to 4K7 and put the suspect chip back in! R7-R10 should not affect anything but may account for my funny resets at times as the D0-D3 will take longer (on yours) to get to transition and I wonder if that affects the strobing of the top 4 address lines? I will put them up to your 6K8 and see if that improves. Obviously all this depends on the capacitance of different trace layouts as well and perhaps slightly different spec chips, especially on the D0-D3 side as a lot of chips touch the bus. I may not have time until the weekend to try that now as still some work to do getting my PC fully operating like restoring the raid functions! The VDU test will be useful as I have now had my PCB's from Martin so will be looking to build it and I just assumed I could substitute... |
24th Apr 2019, 9:38 pm | #75 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 programming interface - MK2
By all means send the IC back again but, since it has shown very different behaviour on two occasions in your own machine I'm not sure what conclusion we could draw whether we find it still does or does not slow down the operation of the uploader on my machine. Maybe there was just some very persistent oxidisation on the pins which has taken a few insertions and removals to scrub off.
I meant to do the VDU check just about now but have just discovered that one of the ceramic capacitors has been knocked off - fortunately, I found it in the drawer where the PCB itself has been living for many years. I'll take it to work, where I have much better light, space and soldering / desoldering tools than I have here nowadays. I'd better keep the whole thing in a padded bag from now on. The test will not be a true test, in the sense that my PIC 'VDU Demo' board will be driving it, and the port pins on that have very different (stronger) drive capability than anything on the MK14's buses. But it will be running at full (4MHz) speed. Last edited by SiriusHardware; 24th Apr 2019 at 9:44 pm. |
26th Apr 2019, 11:08 pm | #76 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 programming interface - MK2
Just a cross post in case someone is reading this thread and not the VDU one. I tried one of my MM80C95 chips (CMOS version) in place of IC11 and it works fine so far for the keyboard and the programmer at full speed. Probably because it is connected to the Data Bus so only needs a single fan out drive ability as all the others will not have CS asserted.
|
11th May 2019, 6:56 pm | #77 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 programming interface - MK2
NEWS : Sirus Hardware PI Programmer nearly on Gadget show - realises it cannot interface with MK14MAN 2015 model but, was pleased to see one turn up at PlayExpo Manchester 2019 so had a selfie with it...
https://twitter.com/timbucus/status/1125490199360933889 Here they are in another photograph from the new owner of Colin's machine @desleido. |
11th May 2019, 11:17 pm | #78 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,586
|
Re: MK14 programming interface - MK2
I thought MK14man had made more than one of his 'new' MK14s - I am sure there are photos showing more than one in the same shot.
I had the impression he was intending to make some available to people who wanted one, but I wasn't sure if it ever got to that point. It doesn't appear to have a keypad edge connector - I've never noticed that - a bit of a show stopper. I can see the logic, if you are making the first MK14 ever to have a decent keypad, why would you even need a connection for an external keypad? Last edited by SiriusHardware; 11th May 2019 at 11:24 pm. |
11th May 2019, 11:32 pm | #79 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 programming interface - MK2
Yes he made several I should have said another new owner! I asked and he does hope to make more available - he is finishing his smaller one first though.
|
12th May 2019, 11:49 am | #80 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,364
|
Re: MK14 programming interface - MK2
I also took another look at his schematic and the 16 pin header on the bottom right is an external keyboard connector - it has 5v and GND plus the 8x4 connectors - there are two spare which could be patched to be reset I suppose - so we can make an adapter for it at least.
|