Took Apart My TS1000 - I Have Questions

[DrVenkman]

With the bottom off, I see the nice red finish I'd expect of an Issue 3, like the US-market ZX81 I got a few months ago. However, the layout - despite being machine-routed, looks like an Issue 1. Flip it over and huh, it says "Issue 1" right on the silkscreen, despite being machine-routed. So my first question is, how common is this version board, and why didn't Sinclair call it simply "Issue 2"?

Second, take a look at the power jack. At first I thought maybe the connector had been bent or damaged in use but after looking at the bottom of the board, I'd have to go with nope. Looks like it came this way from the factory (Portugal, per the markings on the bottom of the case). Is this kind of factory fubar a common thing?


had intended to do only minimal changes to this board internally but the stock ULA produces pretty rotten RF. I will at least do a composite mod (using Mark's buffer circuit, as I built for my ZX81). But now I feel like I need to remove this power connector, clear out that through-hole and solder it back in place properly. And since I'm desoldering stuff, maybe I need to go ahead and replace the 7805 with a Traco, as I did with my ZX81. And since the ULA lacks the back porch, maybe I need to buy another vLA81 to install. And since I have the extra SRAMs, maybe I should install at least an internal 16K upgrade, even if I don't get another vLA81 ...

It never ends, does it?

[1024MAK]

With the bottom off, I see the nice red finish I'd expect of an Issue 3, like the US-market ZX81 I got a few months ago. However, the layout - despite being machine-routed, looks like an Issue 1. Flip it over and huh, it says "Issue 1" right on the silkscreen, despite being machine-routed. So my first question is, how common is this version board, and why didn't Sinclair call it simply "Issue 2"?

They are not rare.
We don’t know why it was not called an issue two.

Second, take a look at the power jack. At first I thought maybe the connector had been bent or damaged in use but after looking at the bottom of the board, I'd have to go with nope. Looks like it came this way from the factory (Portugal, per the markings on the bottom of the case). Is this kind of factory fubar a common thing?

I’ve not seen it with a 3.5mm socket before, but it has been seen with ‘drunk’ chips and resistors before…

Which ULA does it have?

It never ends, does it?

Nope

Mark

[DrVenkman]

Which ULA does it have?

The same as in my ZX81, the 2C184E. This on was made in the 10th week of '82 per the date code (that one was made in the 26th week of '82). I swapped ULAs to see if the RF would improve but sadly, it didn't.

So I've already removed and gutted the RF can and have one of your video mod circuits mostly built up on protoboard, like I did with the ZX81 last month. I haven't yet tested it but it follows your schematic just like the one I built before so I am sure it will work great.

Just a test-fit before making some connections and testing tomorrow:

[BarryN]

Which ULA does it have?

The same as in my ZX81, the 2C184E. This on was made in the 10th week of '82 per the date code (that one was made in the 26th week of '82). I swapped ULAs to see if the RF would improve but sadly, it didn't.

So I've already removed and gutted the RF can and have one of your video mod circuits mostly built up on protoboard, like I did with the ZX81 last month. I haven't yet tested it but it follows your schematic just like the one I built before so I am sure it will work great.

Just a test-fit before making some connections and testing tomorrow:

IMG_0762.jpg

Which version ULA was the one you swapped it with?

[DrVenkman]

Which version ULA was the one you swapped it with?

Same as what was in it. I was looking to see if the output quality difference was due to a variation in the specific chips. It wasn't - the RF modulator on the TS1000 just made a worse image than the modulator from the ZX81. That's why I decided to go ahead and build a composite mod board for it.

[1024MAK]

The quality or rather, lack of, of the RF TV output is very much determined by the modulator, and how good the TV is at tuning to the actual carrier frequency produced by the modulator.

These modulators were widely used during the 1980s in various computers. But they are purely analogue designs and don’t use a crystal. Hence the chances of the carrier frequency actually being at the nominal broadcast frequency is low.

This did not matter with analogue CRT televisions, as they would tune to any frequency within their tuning range. But some modern TVs may not be able to be fine tuned.

Also, as the inside of the computer heats up, the carrier frequency will drift. So the picture may degrade before your eyes, unless the TV has an Automatic Frequency Control (AFC).

And of course, these modulators are 35 (or more) years old. And so there is a likelihood of a fault having developed.

It’s interesting that your TS1000 has a 2C184E in it. That blows out the idea that these were all fitted with the 2C210E ULAs, and hence they all have video outputs that include the ‘back porch’.

I was thinking about the USA Issue one board. I don’t have one in my collection, so this is just pure speculation. As it’s hard to compare using only photos. If the functional layout of the USA Issue one board is the same as that of the first ZX81 (hand drafted) issue one board, with no schematic circuit changes. Then that may be why whoever produced it, decided to also call it an issue one. And the PCB design may have been done by Timex (the contract manufacturer) rather than by someone at Sinclair. Although we don’t know who drafted any of the Sinclair PCB layouts from the ZX80 through to the QL.

There are some small schematic circuits changes with the ZX81 issue three boards. The ZX81 issue three boards are also found in later TS1000 models and in European ZX81 computers. Again we don’t know who drafted the PCB design. Was it Timex or Sinclair?

Mark

[BarryN]

It’s interesting that your TS1000 has a 2C184E in it. That blows out the idea that these were all fitted with the 2C210E ULAs, and hence they all have video outputs that include the ‘back porch’.
Mark

My TS1000 came with a 2C184E ULA. I had video issues with it back in the day, dark screen on my TV, etc. When I read an article that they had come out with an improved ULA, I pitched a fit and made them send me a new one. I have used that one (a 210E version) ever since and it still works great.

[DrVenkman]

The quality or rather, lack of, of the RF TV output is very much determined by the modulator, and how good the TV is at tuning to the actual carrier frequency produced by the modulator.
…

Indeed. I’ve got lots of (too much!) experience with crappy or marginal RF from vintage consoles and computers, mostly Atari systems, but also some Commodore stuff. Many Atari modulators have a tuning coil accessible through the can so you can tweak the carrier frequency if it has drifted over the years. Interestingly (if only to me), I noticed when I popped the lid off the RF can on this TS1000 that there’s a hole in the top, covered by the label, to accommodate that kind of adjustable coil even though this modulator doesn’t have one.

Anyway, the modulator board has been extracted and I should have the new composite mod tested and installed later today. Still debating if I’m going to do a simple internal 16K mod (already have the SRAM chip on-hand) and replace the 7805 to lower the heat load inside the computer.

Oh, and I gotta fix that’s crazy power connector.

[Lardo Boffin]

I would definitely replace the 7805 if you are going to use it a lot. It makes a huge difference.

[1024MAK]

An example of a crazy drunk chip is here

Mark