A TS1500 Resurrection.

[dessony]

A TS1500 Resurrection. Step by step appreciated.

Hi everyone,

I have 2 (two) TS1500. One board have had the 8 (eight) 4116 DRAMs desoldered, removed and -5VDC and +12VDC circuit section removed from the board. And another board have these DRAMS and its related circuit section still intact.

Both TS1500 boards have had 1024MAK's excellent video composite output circuit which I built it on a little Veroboard or a little stripboard. These little cute boards were loosened-fitted within the gutted video modulators. I tested these little cute boards on my 3 (three) TS1000s and a ZX81 with an excellent result!

Both big boards still have its temporary 7805 replacement but not heatsink.

I have an oscilloscope, a multimeter and have enough time to fix these two big boards.

Thank you very much in advance.
Dessony

[1024MAK]

Hi Dessony

I have 2 (two) TS1500. One board have had the 8 (eight) 4116 DRAMs desoldered, removed and -5VDC and +12VDC circuit section removed from the board.

So as there is rather a lot missing, do you want to restore this board back to how it should be, or do you want to use slightly more modern DRAM chips that don't need the -5V and +12V supply rails?
Where the 4116 DRAM chips were fitted, are the tracks and pads on both sides of the board undamaged? If you are not sure, can you post some high quality close in photos please.

And another board have these DRAMS and its related circuit section still intact.

So this board is complete? I presume it does not work. What output does it display on a TV?

Both big boards still have its temporary 7805 replacement but not heatsink.

What do you mean?
If you mean you have fitted a modern DC-DC switching regulator (RECOM or TRACO), these don't need a heatsink.
If you mean 7805 series voltage regulator chips with metal tabs are in use, they do need to have heatsinks fitted. As they reduce the output voltage and current when they get too hot. And they will get hot...

For the second board that is not missing anything, the first step is to use your meter to confirm that the three DC supply rails are okay. The +5V ideally should be between 4.85V and 5.15V and certainly between 4.75V and 5.25V.
The -5V is normally specified to be between -5.5V and -4.5V. The 12V should be between 10.8V and 13.2V.

Next, using your 'scope, check to see if there is a 3.25MHz squarewave (not that it will be very square) on pin 6 of the CPU.

Mark

[dessony]

Hi 1024MAK (Mark),

I want to install slightly more modern DRAM chips or SRAM chip which doesn't request -5V and +12V. I'm submitting a few photos here.

Top side of the TS1500.

DSC00247.JPG (184.01 KiB) Viewed 4033 times

Bottom side of the TS1500.

DSC00248.JPG (165.8 KiB) Viewed 4033 times

Notice that I took parts off, which is producing these -5V and +12V, the board.

Near a Power Jack area.

DSC00249.JPG (176.5 KiB) Viewed 4033 times

Near a Voltage Regulator area.

DSC00250.JPG (174.96 KiB) Viewed 4033 times

I also will type which parts are removed from the board little later.

I need more time on a big board with its DRAM chips intact... However, I will post a few photos of the result on a monitor's CRT and 'scope. I could see a familiar back porch waves horizontally moving somewhat fast with some blurs from a D19 anode.

Fortunately, I have a XP Power SR10S05 ready. I will take 7805 off the board and put the DC-DC switching regulator on the board.

After I replace the V. R., I will measure the board and post a result here.

Thanks,
Dessony

[1024MAK]

Upgrade a TS1500 to use more modern RAM

Options (none of which can do hi-res graphics as far as I can tell):

• Use eight 1 bit x 16k MCM4517P12 DRAM chips - very easy, if you can find the chips... These don't need the +12V or -5V supply rails. Only gives 16k bytes of RAM.
• Use eight 1 bit x 64k 4164 DRAM chips - is the next easiest. System will only see 16k bytes. If RA7 is connected, and an extra logic chip is added, the system may see more memory. See http://zx81.us/ts1500a.txt
• Use two 4 bit x 16k DRAM chips - is a bit tricky. Only gives 16k bytes of RAM.
• Use two 4 bit x 64k DRAM chips - is a bit tricky. Gives 16k bytes. If RA7 is connected, and an extra logic chip is added, the system may see more memory. See http://zx81.us/ts1500a.txt
• Use a SRAM module designed for the ZX Spectrum - electrically it should work, but I have no idea about the mechanical fit. There are two that I know of, this one and this one (3/4 of the way down the page). Only gives 16k bytes.

To have a memory set-up that is compatible with hi-res graphics, we ignore the memory interface of the TS1500 ULA and add a 32k byte SRAM chip controlled by new logic.
The data lines of the new SRAM chip have to connect to either the old 4116 DRAM data lines or the ROM data lines. The address lines A0 to A14 of the new SRAM chip have to connect to the CPU pins. The new SRAM chip and the new logic circuits will have to be mounted on some stripboard or matrix board. This requires a fair amount of work.

I don't have a TS1500, so I can't be certain about all the details, as I would be relying of information that I have found from other people.

What are your thoughts?

Mark

[dessony]

As for the voltages readings off the TS1500; +5V on a C11 + side = 5.06V; -5V between D9 & R23 = -5.17V; +12 (to be determined of its test point) = 12.36V.

I will post a few pictures of o' scope's screens here little later tonight.

Dessony

[dessony]

As I promised all of you that I would post these photos here. Now they are here!

DSC00266.JPG

TS1500's back porch wave at the video output port: 1V AC ,10us, Trig. Mode TVH, Source INT.

(419.99 KiB) Downloaded 233 times

DSC00268.JPG

Only "white screen! No Black Square and white K!

(375.17 KiB) Downloaded 233 times

DSC00269.JPG

TS1500's clock wave on Z80's pin #6: Volt/Div., 1V AC .1us, Trig. Mode Auto, Source INT.

(416.93 KiB) Downloaded 233 times

Thanks all of you for your patience with me.

Cheers ,
Dessony

[1024MAK]

The horizontal sync and back porch, and CPU clock signals are good
So the ULA looks to be working (as far as we can tell).
Next up, I suggest checks/tests on the CPU.

Does the reset pin (pin 26) go from low to high at power-up.

For reference here is a pin-out of the Z80 CPU.


Then test the following CPU signals/pins:

Remember, we are just looking to see which lines are active, don't worry about the actual waveform.

Pin Function
16 /INT (should be pulsing)
17 /NMI (should be pulsing)
18 /HALT (should be pulsing)
19 /MREQ (should be pulsing rapidly)
21 /RD (should be pulsing rapidly)
22 /WR (should be pulsing)
24 /WAIT (should be pulsing)
25 /BUSRQ (should be high)
26 /RESET (should be high, except for a very short time after power is applied)
27 /M1 (should be pulsing)
28 /RFSH (should be pulsing)

Then test address lines A14 and A15. These give an idea of which memory the CPU is accessing and if it is trying to generate a TV display.
You can also test the remaining address lines A0 to A13 and the data lines D0 to D7. A lack of activity is suspicious.

There is no need to take photos of all these signals if they are doing what is expected.

Mark

[dessony]

1024MAK,

Mini update, I decided to replace all of these electrolyte capacitors after I noticed that one (C23) of the capacitors had some very little white crystal "fuzz" at its negative side and had a surface somewhat looked like a cucumber's surface. The C23 is soldered onto the TS1500 board by a ground and a Z80 pin #26.

I will have to buy new caps this week.

Thanks,
Dessony

[1024MAK]

The C23 is soldered onto the TS1500 board by a ground and a Z80 pin #26.

That will be the capacitor in the reset circuit. No low to high reset pulse means a stalled CPU.

Mark