TS1500 Diagnostics (No Cursor)

[capndan2112]

Hello all, I've recently acquired a Timex Sinclair 1500 in seemingly good condition. None of my TVs (Everything I own is NTSC, I am in America, so no problems there) could pick up its old RF signal so I set out to do a simple composite mod based off of http://www.zx81stuff.org.uk/zx81/ts1500_video.html. Before I got to work soldering I quickly jumped together the video signal at R20 to the RCA jack center pole just to make sure it worked before committing, and I got a plain white / light gray screen with no K cursor. I'm really hoping it's not a ULA problem as that seems the hardest to remedy out of all the possibilities.

Tests I've already done are:
-Power supply: outputs 11 ish volts, consistent with what I've seen from others online
-Voltage Regulator: Gets the 11ish in one end, outputs 5v out the other just like it should
-Z80 clock pin: my meter reports a 3.25MHz frequency here, all good
-Video signal: I pick up 60Hz at R20 and the modulator, and of course I get that white screen, so doesn't seem to me that video is the problem
-I've also done continuity testing between the ground plate and the ULA pins, only the ground pin and pin 18 (CCW from the dot, pretty sure that's how PLCCs are numbered) are grounded
-Visual inspection: I'm no expert but I couldn't find any obvious shorts or cold/cracked joints or anything like that

What should my next course of action be? I have a multimeter and a soldering iron, but sadly no oscilloscope. I also don't have a ram pack and no means of obtaining one anytime soon.

Some pictures just in case, red wire was to test the composite signal:
https://imgur.com/a/I4328u8

[1024MAK]

Hello capndan2112

Welcome

You may want to introduce yourself in the welcome area

I’ve not looked at your photos yet. But a white screen with nothing else is produced when the ULA is not being fed data by the Z80 microprocessor.
This is normally due to either:

• The Z80 not receiving a clock signal on pin 6 (however you’ve already tested this and it appears fine),
• The reset input on the Z80 (pin 26) being held low, or never pulsing low at power up,
• The loss of either the -5V supply or the loss of the +12V supply to the DRAM chips.
• One (or more) faulty DRAM chips),
• The ROM being faulty.

If the 4116 (or equivalent) DRAM chips have been powered up with the +12V and/or +5V present, but with the -5V missing, one or more may have self destructed

Dry solder joints, broken/cracked tracks or short circuits between signal lines can also cause this, but the latter two are unlikely if the board has not been worked on since manufacture.

There are some useful links in my signature below...

Mark

[bola_dor]

I've had some probles with that simole mod with some TV in the past.
I see from your picture that you kept the modulator internals using the same output. That didn't work to me either before..
I would use a different rca female output for diagnostic at least. Or if you want to keep the RF output...
Have a look to my mod here. Is a different board but it may help https://www.sinclairzxworld.com/viewtop ... f=7&t=3771

[capndan2112]

Thank you for the replies!

• The Z80 not receiving a clock signal on pin 6 (however you’ve already tested this and it appears fine),
• The reset input on the Z80 (pin 26) behind held low, or never pulsing low at power up,
• The loss of either the -5V supply or the loss of the +12V supply to the DRAM chips.
• One (or more) faulty DRAM chips),
• The ROM being faulty.

I tested the z80 reset pin, and it's held high when the power is on, however without an oscilloscope I have no way of knowing if it ever pulses low. I tested all the voltage pins on the DRAM (4116) and all are correct, however Vcc tends to be at around 4.3-4.4V rather than 5, not sure if this is a problem. I also tested the ROM and it too is receiving proper power, and every chip has its ground connected. Is there any pinout for the 1500 ULA so I can start poking around there? Mine is labeled:
TS 1500
(c) Timex
Computer Corp 1983
NCR F816145
F8507 USA

I see from your picture that you kept the modulator internals using the same output. That didn't work to me either before..
I would use a different rca female output for diagnostic at least. Or if you want to keep the RF output...

I've still got the modulator board in there, yes (lost my desolder braid and haven't had time to buy a new one ) however I cut everything off of the RCA center pole so that nothing is contacting it except my wire. Could this still be a problem?

[bola_dor]

I've still got the modulator board in there, yes (lost my desolder braid and haven't had time to buy a new one ) however I cut everything off of the RCA center pole so that nothing is contacting it except my wire. Could this still be a problem?

If you cutted away the +5v to the modulator and any other contact to the center pin out then it should work OK.

To test the reset you must identify the electrolytic capacitor that is between pin 26 and GND. Measure with the multimeter over pin 26 and short the Cap leads. Voltage in pin 26 should drop to cero and then raise slowly to about 5v when shorting is released.. I can't point the capacitor as my board is different but you can test for continuity until you find it..

[1024MAK]

I tested all the voltage pins on the DRAM (4116) and all are correct, however Vcc tends to be at around 4.3-4.4V rather than 5, not sure if this is a problem. I also tested the ROM and it too is receiving proper power, and every chip has its ground connected.

I’m a bit concerned that on the rest of the board the +5V supply is good, but on the DRAM it’s only about 4.3V to 4.4V. This needs following up.

I *think* there is a pin-out of the Timex ‘ULA’ somewhere, but I can’t remember where. So you will have to search for it. In any event, there is at least one schematic of one version of the TS1500 available. See here .

To test the reset circuit without a ‘scope, with the power to the board off, switch your multimeter to a high resistance range, then with the red probe connected to any convenient +5V point, connect the black probe to pin 26 on the Z80. You should get a high resistance reading (it may change over time as capacitors start charging). Now disconnect both probes.

Now connect the black probe to any convenient +0V/GND point, then connect the red probe to pin 26 on the Z80. The resistance value should start low, then quickly rise as the ‘reset’ capacitor charges.

Useful sites:

https://www.timexsinclair.com/

Mark

[bola_dor]

I don't remember were did I download it from.

ts1500_ula_pins_labels.txt

(1.5 KiB) Downloaded 195 times

[capndan2112]

Thanks for the schematics, couldn't find these anywhere else on the web.

To test the reset circuit without a ‘scope, with the power to the board off, switch your multimeter to a high resistance range, then with the red probe connected to any convenient +5V point, connect the black probe to pin 26 on the Z80. You should get a high resistance reading (it may change over time as capacitors start charging). Now disconnect both probes.

Now connect the black probe to any convenient +0V/GND point, then connect the red probe to pin 26 on the Z80. The resistance value should start low, then quickly rise as the ‘reset’ capacitor charges.

In the first test I read 900 ohms, staying the same over about 30 seconds of connection. In the second, it starts at about 2.25 kilohms and within the span of a second stabilizes to 2.3.

To test the reset you must identify the electrolytic capacitor that is between pin 26 and GND. Measure with the multimeter over pin 26 and short the Cap leads. Voltage in pin 26 should drop to cero and then raise slowly to about 5v when shorting is released.. I can't point the capacitor as my board is different but you can test for continuity until you find it..

Using this method I short the leads, voltage drops to zero, release the short, and it raises almost instantly back to about 5.
Is it possible I have a leaky cap? Upon a visual inspection none of them are bulging however they are almost 40 years old after all

As a side note, lately when I plug it in I get some sort of interference pattern with vertical lines rather than a plain white screen.

[bola_dor]

Using this method I short the leads, voltage drops to zero, release the short, and it raises almost instantly back to about 5.
Is it possible I have a leaky cap? Upon a visual inspection none of them are bulging however they are almost 40 years old after all

As a side note, lately when I plug it in I get some sort of interference pattern with vertical lines rather than a plain white screen.

Looks like its OK.. it just has to give time to the rom and ram to be already up when zx80 starts it's job.
If voltages are in doubt do carefully check all tracks and passive components in the way. Measure resistors and exchange electrolitics. Solderings may be bad and if any IC on a socket remove it clean the pins with a water and vinager solution and then with IPA (there may be traces of oxidation) and place them carefully again.
Then I'll check the Z80.. I am sure 1024MAK knows a lot of this but in the mean time yo can check for REFRSH (pin 28) to be pulsing. If you don't have a logic probe or an oscilloscope you can measure voltage and should be around 2.5 or 3v.. if it lower than 2.2v you can be sure is not pulsing enough.. this is not very academic but can tell you if Z80 is not working at all. For further testing you'll need some instruments..
I saw you verified CLK at 3.25MHz was it with a frequencymeter ? Can it tell you if levels are OK?
I am not sure how does your board reacts to a rampack.. but may be a way to rule out internal ram but I don't know if it does over step the internal ram in your TS1500..
To rule out rom I don't know other way tan to replace it wit another .. that is easy if you have it on a socket.. Anyway is no very common to have a faulty rom..

[1024MAK]

Okay, time to admit that I have dropped the ball with my advice on testing pin 26 on the Z80. What I said works on all the Sinclair Research made Z80 based computers, but Timex with the TS1500 have done things differently. So the test I described is not applicable Sorry.

Apart from a multimeter, what test gear do you have?

And which chips on the board are in sockets?

As per bola_dor’s comments, if you do have a Timex or a Sinclair 16K RAM pack, try cleaning the edge connector contacts on the TS1500 then fit the RAM pack and power on. The ULA in the TS1500 can remap it’s internal memory so that the RAM pack can be used. So it’s worth a try.

Mark