Timex Sinclair 1000 repair help needed.

[jknightandkarr]

I have 2 systems, neither seam to work even on a CRT tv. I composite modded 1 with transistor and resistors, the other with the computer/tv box and adapter to conmect to tv, will check caps, but what are common issues that might cause them to not function? I've heard ULA's can overheat and fail. Should I try there or start with the Z80 inside n connect to my logic analyzer on the data pins n see what happens at power up? Thanks.

 

[Dr. Frankenstein]

I have 2 systems, neither seam to work even on a CRT tv. I composite modded 1 with transistor and resistors, the other with the computer/tv box and adapter to conmect to tv, will check caps, but what are common issues that might cause them to not function? I've heard ULA's can overheat and fail. Should I try there or start with the Z80 inside n connect to my logic analyzer on the data pins n see what happens at power up? Thanks.

Repairing a Timex is relatively easy.

 

[Istarian]

Maybe you should use a multimeter to look for broken traces or bad connections? Checking that the voltages supplied to each chip are at apppropriate levels is also a good idea.

If you have an oscilloscope or logic analyzer, check the clock signal.

 

[daver2]

First off, check the voltage rails.

Second, check that the clock is oscillating at the correct frequency.

Third, check that you have a reset signal.

Dave

 

[jknightandkarr]

Maybe you should use a multimeter to look for broken traces or bad connections? Checking that the voltages supplied to each chip are at apppropriate levels is also a good idea.

If you have an oscilloscope or logic analyzer, check the clock signal.

I've already been doing this with the help of an electronics tech friend of mine, but I haven't checked the traces, will check thanks.

First off, check the voltage rails.

Second, check that the clock is oscillating at the correct frequency.

Third, check that you have a reset signal.

Dave

I've checked the rails with a DMM and my scope, I think we checked the clock signal, but will double check. I know we didn't check the reset, will do thanks.

 

[xp0nentz]

I've already been doing this with the help of an electronics tech friend of mine, but I haven't checked the traces, will check thanks.

I've checked the rails with a DMM and my scope, I think we checked the clock signal, but will double check. I know we didn't check the reset, will do thanks.

Check reset and send your findings here.

 

[jknightandkarr]

Here's all the testing I've done already, forgive photo and links to them please. I have issues with them working time to time, and for some reason they work when clicked.....

https://jknightandkarr.altervista.org/Computers/PowerSupply.jpg
Power Supply at the jack.

https://jknightandkarr.altervista.org/Computers/5v_card.jpg
5V @ the card edge connector. Verified, 5v

https://jknightandkarr.altervista.org/Computers/9v_A.jpg

https://jknightandkarr.altervista.org/Computers/9V_B.jpg
Couple of shots of the 9V at the card edge connector. Kinda messy..

https://jknightandkarr.altervista.org/Computers/M1.jpg
M1 at the card edge connector. Freq = 810 Khz.

https://jknightandkarr.altervista.org/Computers/Pin6.jpg
Pin 6. 3.25 Mhz

https://jknightandkarr.altervista.org/Computers/Pin11.jpg

https://jknightandkarr.altervista.org/Computers/PIN11_scope.jpg
Pin 11 DMM Vs Scope.... UNKNOWN why meter voltage is diff from the Scope measure, which was 5v DC..... DMM showed no AC voltage.

https://jknightandkarr.altervista.org/Computers/7805.jpg
Temp at the voltage regulator

https://jknightandkarr.altervista.org/Computers/Z80.jpg
Z80 Temp

I did a quick test on the Reset pin, on the edge connector, I got 5v on the pin, so this is where I need to start I think.

 

[jknightandkarr]

Misread info. Reset fine, guess start with caps.

 

[Istarian]

The voltage supplied by whatever brick you have plugged into the computer may be stepped down to say 12V AC and therefore unregulated and somewhat load dependent.

If it's a switching power supply then it's regulated, but the output may have some voltage swing under load since the current may not be constant. That may explain the wobble on the 9V measurement.

Neither situation should prevent the linear voltage regulator (e.g. an LM7805 or LM7809) on the mainboard from outputting a steady voltage as long as the capacitors or other supporting components have appropriate values and are functioning properly.

Digital multimeters only sample the measured signal as fast as the ADC used and therefore report just a single instanteous value. The output of an analog multimeter would be bouncing around as fast as the circuitry could respond to the changing signal.

Oscilloscopes are designed to show a continuous measurement over time. An analog scope, usually with a CRT display, simply "renders" the input as it arrives while a digital (storage) oscilloscope will have a much higher sampling rate than your multimeter and display as much of the signal as it can store at any one time.

 

[jknightandkarr]

The voltage supplied by whatever brick you have plugged into the computer may be stepped down to say 12V AC and therefore unregulated and somewhat load dependent.

If it's a switching power supply then it's regulated, but the output may have some voltage swing under load since the current may not be constant. That may explain the wobble on the 9V measurement.

Neither situation should prevent the linear voltage regulator (e.g. an LM7805 or LM7809) on the mainboard from outputting a steady voltage as long as the capacitors or other supporting components have appropriate values and are functioning properly.

Digital multimeters only sample the measured signal as fast as the ADC used and therefore report just a single instanteous value. The output of an analog multimeter would be bouncing around as fast as the circuitry could respond to the changing signal.

Oscilloscopes are designed to show a continuous measurement over time. An analog scope, usually with a CRT display, simply "renders" the input as it arrives while a digital (storage) oscilloscope will have a much higher sampling rate than your multimeter and display as much of the signal as it can store at any one time.

My meter updates pretty quick, I'll have to retest my measurement on it.

I been using my Rigol scope to see what it's doing. Should break out it's logig analyzer kit n see what the data lines are doing on power up.

 

[SiriusHardware]

DMM showed no AC voltage.

The pulsed-DC waveforms you are looking at with both your scope and your DMM are varying between 0V and (about) +5V DC.

The scope will be happy to show / measure this but the DMM, if on AC or frequency mode, may only work if the signal is switching between plus (above 0V) and minus (below 0V) polarity. I've come across at least one, initially baffling case where a DMM on Hz / Frequency mode refused to show the frequency of a 1MHz pulsed-DC signal until the 0V lead of the meter was placed on a point known to be at steady +2.5V. This made the pulsed 0V to +5VDC signal 'look' to the meter as though it was alternating between +2.5V and -2.5V and allowed the meter to indicate the frequency of the signal, which was a (steady-frequency, roughly 50/50) clock signal.

The root cause was probably that there was an internal DC blocking capacitor being switched in in series with the meter input when that particular meter was in Hz / Frequency measurement mode. Many times prior to that case I had used the 'Hz' range on other multimeters to measure the frequency of a Pulsed-DC signal without any problems.

 

[SiriusHardware]

The other thing is, if you use a DC voltage meter to measure the voltage of a Pulsed-DC signal you might expect that the meter reading would be changing between showing '0V' and '5V' so fast that the reading would just be a blur. What actually usually happens is that the meter will just show the average of the two voltages, so if you tried to measure the DC voltage of a steady 50/50 clock signal you would most likely see a voltage of about 2.5V, in reality probably somewhat less as the 'logic 1' voltage on old 5V logic circuits is rarely as high as 5V.

However, this can be a useful quick and rough indicator of whether an expected clock signal is there or not - if you put your DC voltmeter on that 50/50 clock signal line and you see a voltage either close to 0V or 3.5V or more, the chances are that that clock signal is not there, because if it was then the voltage would be more like 2.5V or a little less. Of course, an intermediate voltage is no absolute guarantee that the clock signal is present or the right frequency but an obvious logic '0' or logic '1' voltage on the line would be reason enough to reach for your scope.

 

[jknightandkarr]

The pulsed-DC waveforms you are looking at with both your scope and your DMM are varying between 0V and (about) +5V DC.

The scope will be happy to show / measure this but the DMM, if on AC or frequency mode, may only work if the signal is switching between plus (above 0V) and minus (below 0V) polarity. I've come across at least one, initially baffling case where a DMM on Hz / Frequency mode refused to show the frequency of a 1MHz pulsed-DC signal until the 0V lead of the meter was placed on a point known to be at steady +2.5V. This made the pulsed 0V to +5VDC signal 'look' to the meter as though it was alternating between +2.5V and -2.5V and allowed the meter to indicate the frequency of the signal, which was a (steady-frequency, roughly 50/50) clock signal.

The root cause was probably that there was an internal DC blocking capacitor being switched in in series with the meter input when that particular meter was in Hz / Frequency measurement mode. Many times prior to that case I had used the 'Hz' range on other multimeters to measure the frequency of a Pulsed-DC signal without any problems.

Ok, thanks. The only thing I tried measuring AC on was the 5v rail per my friend's request. On my scope when I was adjusting the screen, I put the lower level of the measuring bar, whatever its called, on 0v, so I know there was no negative voltages at all, and measured the top with the upper of the measuring bar, again no idea what its actually called, not much exp with scopes, just enough so far to get me by on some of the basics, since there are bound to me more I don't understand. I need to watch videos on using them correctly.

The other thing is, if you use a DC voltage meter to measure the voltage of a Pulsed-DC signal you might expect that the meter reading would be changing between showing '0V' and '5V' so fast that the reading would just be a blur. What actually usually happens is that the meter will just show the average of the two voltages, so if you tried to measure the DC voltage of a steady 50/50 clock signal you would most likely see a voltage of about 2.5V, in reality probably somewhat less as the 'logic 1' voltage on old 5V logic circuits is rarely as high as 5V.

However, this can be a useful quick and rough indicator of whether an expected clock signal is there or not - if you put your DC voltmeter on that 50/50 clock signal line and you see a voltage either close to 0V or 3.5V or more, the chances are that that clock signal is not there, because if it was then the voltage would be more like 2.5V or a little less. Of course, an intermediate voltage is no absolute guarantee that the clock signal is present or the right frequency but an obvious logic '0' or logic '1' voltage on the line would be reason enough to reach for your scope.

Didn't try the DMM on anything with any kind of freq, only on the power supply, 9v and 5v rails,, though outta just to see how the meter acts, anyhow, the power supplies I have for the 2 systems are the originals that depend on loads to get to the correct voltage, so other then the caps, not sure why there was a freq on pin 11, which is the 5v rail i think. I noticed that my meter has 3 modes Freq, Pulse Width, and Duty Cycle, just not useful for this as the manual states, it's max input is 999.999 KHz

 

[jknightandkarr]

https://jknightandkarr.altervista.org/Computers/20260128_112338.jpg

Well, Hooked up my Rigol DS1194Z's Logic Analyzer up to D0-D3, saw no need to try all 8 data lines. Anyhow, the Z80 is clearly doing something, so thinking of picking up one of the more modern replacements, such as this: https://www.tindie.com/products/charlieingley/vla81-zx81-ula-replacement/ since from my searching, the originals are prone to failure, so that's what I have so far.

EDIT: That website has some pretty nice stuff, such as the newer ULAs and a Tactile keyboard. I has to think!

 

[jknightandkarr]

https://web.archive.org/web/20250123231729/https://www.bytedelight.com/?page_id=3560

Well...... Pretty sure I figured out my problem, on one of the ZX-81s anyhow... Circuit above is the one I used. Now...... PROVIDING..... I could pay attention to what I was doing, I'd wire it as in the link.....
E-Output
B- Signal
C- 5V+
it MIGHT work..... Me:
E-Signal
B- 5v
C- Output

Need to pay more attention and not work on stuff when tired..... new transistor and will try again.....

 

[jknightandkarr]

Ok, I replaced the transistor, ensured it was wired correctly, found out my mod also didn't have power, fixed that. removed the 33 ohm resistor, and I now have a screen on my tv. It's inverted and very faint, so white is black and black is white, hard to see, but it's there, idk if my transistor is the issue, or if it's the ULA chip. online it says use a 2N3904, but I didn't have anymore, so used a PN3642. Any thoughts? Thinking outta just replace the mod with one bought online. Any suggestions for the best source?

 

[xp0nentz]

Ok, I replaced the transistor, ensured it was wired correctly, found out my mod also didn't have power, fixed that. removed the 33 ohm resistor, and I now have a screen on my tv. It's inverted and very faint, so white is black and black is white, hard to see, but it's there, idk if my transistor is the issue, or if it's the ULA chip. online it says use a 2N3904, but I didn't have anymore, so used a PN3642. Any thoughts? Thinking outta just replace the mod with one bought online. Any suggestions for the best source?

That happened to me with RF. Is it a modern tv or a CRT. If its color, turn the color setting off. That's what worked for me about the invert issue.

 

[tech58761]

Another general purpose NPN-Si transistor is the PN2222A.

 

[jknightandkarr]

That happened to me with RF. Is it a modern tv or a CRT. If its color, turn the color setting off. That's what worked for me about the invert issue.

Modern tv. I can try and turn the color down n see if that helps. Thanks.

Another general purpose NPN-Si transistor is the PN2222A.

Ok, thanks. Sadly I don't have any, but I did use PN3942. Only had 3 diff ones, MPSA06 and 2N5088 I think, but the 2N3904 and the PN2222A aren't among the ones I got

 

[daver2]

The inverted signal (black/white) is probably caused by the video signal being inverted (I state the obvious). The white and black levels are reversed.

Does the monitor input have a 50 or 75 Ohm termination on it? This could account for the lack of brightness if it does.

It would also pay to check the hfe parameter for the original transistor and for the one you are using. Look in the data sheets.

Dave