TS1000 Diagnostic Help

[mrtinb]

If it acts up again, then maybe put some round pin headers into the socket and put the chip into the round pin headers. It's worth a try.

2.54mm Round Female Pin Header
https://a.aliexpress.com/_EGMbRaR

[Moggy]

Great to hear you have had some success at last, long may it continue!

[1024MAK]

Heat and movement/vibration may be a factor where intermittent or poor connections are concerned because either can make a difference between the poor contact being conductive or becoming open circuit or having a high resistance connection. Some contact cleaners help sometimes.

On the subject of removing existing sockets, is it the desoldering that is the problem?

Unless someone is experienced, my recommendation is to use electronic snips (wire / side) cutters to cut the plastic in between each contact. Once the entire socket is chopped up, you can heat up each pin individually with an iron and use long nose pliers to very gently pull that pin out. Never use any force. When the solder is liquid, the pin should come out very easily. Then once all the pins have been removed, you can remove the remaining solder with either a desoldering pump / solder sucker or with desolder braid/wick. Buy good branded from a good supplier, not the cheap rubbish from trading sites.

Then a new socket can be soldered in.

Mark

[bwinkel67]

How can you test if the 16K RAM fully works? I'm having a weird, somewhat intermittent problem with my TS1000 with RAM pack. This little program will fail with either error code 2 or 4:

Code: Select all

10 LET K$=INKEY$
20 IF K$="" THEN GOTO 10

A slightly modified version also fails:

Code: Select all

10 LET K$=INKEY$
15 LET C=CODE K$
20 IF C=0 THEN GOTO 10

They both work fine on EightyOne configured as 16K TS1000.

P.S. One odd note, both on EightyOne and a real TS1000, without RAM pack (i.e. 2K only), both of these programs don't work as C equals 118.

[Edit: fixed code typo where I accidentally used % for $]

[1024MAK]

The 118 is the data in the last key pressed "buffer" (the code of the [ENTER]/[NEW LINE] key you pressed after RUN).

If you delay the key read (INKEY$) by having some other code before it (e.g. PAUSE 10), then the loop until a key is pressed will work correctly.

For testing a 16K RAM pack, see this post

Oh, you are using the INKEY$ keyword and not typing it in as individual letters aren't you? Only you've written INKEY% which is nonsense in BASIC.

Mark

[1024MAK]

Here is a photo of the first program showing the use of a PAUSE 10 so that the INKEY$ works as expected:

Use of PAUSE before using INKEY$

This is on a ZX81 with only 1K of RAM.

Mark

[bwinkel67]

Is it just a timing issue then with 16K (i.e. does the RAM pack slow it down enough so it doesn't need the PAUSE?)

I posted in a new topic about why it fails with error code 2 or 4 when using a RAM pack...turns out its a faulty ROM that still checksums properly.

[1024MAK]

I’ve never looked into why there is a different between a machine with only 1K or 2K of memory and 16K of memory in relation to an INKEY$ returning a new line character when used early on in a program. I know about it because it’s caught me out before.

For reference, the new topic that bwinkel67 mentions is this one.

Mark