Page 2 of 4

Re: TS 1000 bad ULA?

Posted: Sat Jul 09, 2016 3:34 pm
by gary2000
1024MAK wrote: Did you record the AC ripple values (using the AC voltage or mV range)?

Mark
No, I'll have to research how (where) to do that.

Re: TS 1000 bad ULA?

Posted: Sun Jul 10, 2016 12:54 am
by 1024MAK
With a direct current (DC) supply that is derived from the mains alternating current (AC) supply, the DC can be crudely processed or finely processed. What do I mean? Well alternating current flows from the power station to the powered device, then the direction of flow reverses and the current flows back the other way. Because current has to flow in a circle, this does not mean that energy is returned to the power station. Instead power is still delivered to the powered device. It is just the direction that the current is flowing is changing every 20mS, at a frequency of 50Hz (UK power network frequency. If you were able to stop time and trace the waveform of the mains, it would be a sine wave. This is because the power generators are normally rotating turbines.

Power supplies for electronic devices like computers have to convert this AC power to a nice steady direct current with no hint of the AC sine wave that it came from. But practical power supplies have limitations due to cost.

With the ZX81, Sinclair actually used a three stage power supply system. In the Sinclair "9V" power supply unit, there is a transformer to convert the mains voltage down to a lower voltage (and provide isolation from the mains voltage). This is the first stage.

Forming the second stage are some diodes and one or two capacitors. The diodes convert the AC output of the transformer to DC. BUT this DC changes voltage value in line with the changing AC voltage from the transformer :(. The capacitor(s) act like a temporary storage tank, saving some of the electricity from the transformer and then when the voltage from the transformer falls, the capacitor(s) supply the DC output for a short time. This is better, but still far from perfect and the voltage will still rise and fall, but not as much. We can detect this voltage change, as it is at double the frequency of the mains AC voltage. We use the AC range on the meter. The meter should have been designed to ignore DC voltage when switched to measure AC voltage.

So to test and confirm that the power supply is working correctly, we need to measure any AC that is still mixed in with the DC power. This is referred to as a ripple voltage. There are two ways of doing this. The best, is to use an oscilloscope. But these are expensive and you need to learn how to use them, as they are complex items of test equipment. On an oscilloscope, you can see the waveform, and it looks a bit like ripples, hence the name.

The cheapest way to see what the approximate amount of ripple is, is to use a multimeter that has AC voltage ranges. The performance of different multimeters varies rather a lot. Generally the more expensive meters have more sensitive AC voltage ranges. However, when testing the output of the Sinclair "9V" power supply unit, most meters will give a useful indication.

Now you may be thinking, where is the third stage? Well it is inside the computer, and is known as a series voltage regulator chip. In the ZX81, it is the 7805 type voltage regulator. This converts the unstable, unregulated, nominal "9V" from the power supply unit, to a nice stable (regulated) +5V DC supply suitable for powering logic circuits including a CPU and memory :D

I hope that helps ;-)

Mark

Re: TS 1000 bad ULA?

Posted: Mon Jul 11, 2016 5:39 pm
by RWAP
I have been pointed to the following on measuring ripple voltages, which may be of some assistance:

http://www.digikey.com/Web%20Export/Sup ... directed=1

Re: TS 1000 bad ULA?

Posted: Mon Jan 23, 2017 2:10 am
by rcade
My TS1000 has this exact problem also. I have swapped power supplies, with no change. If I remove the case, it takes longer but still happens.

I will try to replace my 7805 first, but I am afraid to damage the old keyboard cable.

Did you resolve your problem?

EDIT: My problem was resolved by replacing the transistor I used for the composite mod (7904). To troubleshoot this, I used the RF output and this interference did not appear. Swapped with new one and problem seems to be gone.

Re: TS 1000 bad ULA?

Posted: Fri May 05, 2017 2:15 pm
by amigaman07
I would Temp. Dis-Connect the Composite MOD, Re-Install the RF to see if Issue goes away.

If still Lines and so-on, try Removing and Re-Seating the ULA, sometimes it is a Bad Contact.

If ULA looks a little corroded on the Legs, give it a light Sand and WD40. (Make Better Contact)!

Sometimes it is the Channel Switch on the TX, a Little Silicon Spray in the Switch works Wonders! :)

Darren (amigaman07)

Re: TS 1000 bad ULA?

Posted: Fri Sep 08, 2017 3:55 pm
by gary2000
This turned out to be bad memory, thank goodness. ula is OK :)

Re: TS 1000 bad ULA?

Posted: Fri Sep 29, 2017 4:16 pm
by gary2000
Correction: all this nonsense seems to have been caused by capacitor C2 ( .047UF).

Replaced this and seems stable. At least I was successful at 16 KB internal memory. Now have to order ZX8CCB to add back porch and finish it up.

Re: TS 1000 bad ULA?

Posted: Fri Sep 29, 2017 6:38 pm
by mhudson52
Now would be a good time to put a good heat sink on the ULA. I was fortunate to have found a working replacement ULA so Sir Ian could bring my TS back to life, but they seldom come around.

Re: TS 1000 bad ULA?

Posted: Sat Sep 30, 2017 8:24 pm
by siggi
mhudson52 wrote: Fri Sep 29, 2017 6:38 pm Now would be a good time to put a good heat sink ...
Oh Oh, he wrote the forbidden word :mrgreen:
Over to Moggy (and his dog) ;)

Siggi

Re: TS 1000 bad ULA?

Posted: Sat Sep 30, 2017 9:08 pm
by Moggy
siggi wrote: Sat Sep 30, 2017 8:24 pm
mhudson52 wrote: Fri Sep 29, 2017 6:38 pm Now would be a good time to put a good heat sink ...
Oh Oh, he wrote the forbidden word :mrgreen:
Over to Moggy (and his dog) ;)

Siggi



HEAT SINK ON A ULA INSIDE A CLOSED SYSTEM WITH NO THROUGHPUT OF AIR !!?? YOU MAY AS WELL WRAP AN ELECTRIC BLANKET AROUND IT :evil:



My attack dog, Brenda, has a very special set of skills (plus a 1975 road map ) and is now coming to find you. When she seeks you out she will start by hitting you with sticks, the she will burn you with fire ( and possibly roll over for a tummy tickle depending on her mood) for even considering such a blasphemous act.

( Tongues in cheeks and all that. :lol: )