[Plus / 4B] Power Issue (fried TR4)

happycactus
Posts: 24
Joined: Tue Sep 28, 2021 6:16 pm

[Plus / 4B] Power Issue (fried TR4)

Post by happycactus »

Hello!

So I finally had my first Plus issue 4B, I bought it as "non working" for a dozen Euros.
Indeed when powered on, TR4 was literally frying.
IMG_20211020_091817.jpg
Image

I changed it with a new ZTX651, but a better inspection showed also C80 had exploded and R29 well cooked.
IMG_20211020_092101.jpg
IMG_20211020_091738.jpg
Changed all of them, changed the 7805 with an SPS module, but still only 5V was present.

Changed C44 and C45 with brand new radial Caps, 12V is ok.

I checked TR5, but found something strange in the schematic.
Schermata 2021-10-20 alle 16.08.39.png
As said, it is a issue 4B, but TR5 is a ZTX 315, but as far as I'm able to see, there's no such a transistor. It could be a ZTX313, but this is a NPN transistor while clearly on the schematic it's a PNP transistor. So I guess it's a ZTX213 like in issue 2A. Can someone confirm?


Image


Changed it with a BC547 but still no -5V. I noticed two ram ics, IC8 and IC9, are going very hot. I'm going to remove them and see what happens.

Thanks in advance, and kind regards
User avatar
Tiger
Posts: 88
Joined: Tue Sep 01, 2020 10:37 am

Re: [Plus / 4B] Power Issue (fried TR4)

Post by Tiger »

Seems it is a wrong value in the diagram. In an copy of the original service manual (page 9 of supplement no.1) for issue 4A-6A modifications the value of TR5 is ZTX213 (BC213/BX214/FRB925). I have always replaced it with the ZTX 750 or ZTX 751.
I think the burned resistor was R29 (2K2) not R59. In this case check D19 ( 5V1 Zenerdiode), D18 / D11 (1N4148) and C46 / C79 (1µF/50V, 1µF/16V).

By the way: the blown 2SC2236 at TR4 is a npn-type ZTX213 or ZTX 750 must be a pnp-type! Maybe this caused the burning resistor ...
User avatar
1024MAK
Posts: 5103
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: [Plus / 4B] Power Issue (fried TR4)

Post by 1024MAK »

happycactus wrote: Wed Oct 20, 2021 4:22 pmTR4 was literally frying.

I changed it with a new ZTX651, but a better inspection showed also C80 had exploded and R29 well cooked.

Changed all of them, changed the 7805 with an SPS module, but still only 5V was present.

Changed C44 and C45 with brand new radial Caps, 12V is ok.

I checked TR5, but found something strange in the schematic.

As said, it is a issue 4B, but TR5 is a ZTX 315, but as far as I'm able to see, there's no such a transistor. It could be a ZTX313, but this is a NPN transistor while clearly on the schematic it's a PNP transistor. So I guess it's a ZTX213 like in issue 2A. Can someone confirm?

Changed it with a BC547 but still no -5V. I noticed two ram ics, IC8 and IC9, are going very hot. I'm going to remove them and see what happens.
TR4 being damaged can also cause TR5 to fail. So it’s recommended to renew TR5 if TR4 keeps failing. The normal reasons why TR4 gets damaged, is due to either one or more faulty 4116 (or equivalent) ‘lower’ DRAM chips drawing excessive power from the +12V rail (and sometimes the -5V rail) or edge-connector abuse. Degrading electrolytic capacitors may also contribute to a failure of TR4 and TR5.

There is nothing special about the transistor used for TR5, any general purpose small signal PNP type that has a reasonable gain can be used, as long as you keep in mind that there is no standard lead-out. Hence if using a transistor that does have a different lead-out, it either needs fitting the opposite way round compared to the silk screen, or it may be necessary to bend two of the leads to crossover each other.

The suggested use of a ZTX751 is because it has the same lead out as the ZTX213 and is electrically suitable. There are numerous typos and mistakes in the schematics. TR5 should be shown as a ZTX213 and not ZTX315 or ZTX231.

R29 is actually a 10kΩ resistor used on the Z80A /WAIT input. The resistor I think you are talking about is R79. And yes, the schematic is wrong again. R79 should be a 2.2kΩ resistor. It is used on the -5V rail. In the older version of this circuit, R79 did not exist, with R55 doing a similar job.

My blog here has the detail of Sinclair’s final version (modification) of the DC/DC converter circuitry.

If a 4116 (or equivalent) ‘lower’ DRAM chip is getting unusually hot, it should be removed. Otherwise the excessive power consumption will cause TR4 to fail again.

The -5V supply is critical to the operation and lifespan of the 4116 (or equivalent) ‘lower’ DRAM chips. Don’t operate the computer other than for quickly testing supply rail voltages until the -5V supply has been restored (unless you remove all the 4116 DRAM chips).

If the +12V supply is present with the -5V out of specification, the 4116 DRAM chips may be damaged.

If there is still a problem with the -5V rail, check the rest of the components in this part of the circuitry, as Tiger says.

Mark
ZX81 Variations
ZX81 Chip Pin-outs
ZX81 Video Transistor Buffer Amp

:!: Standby alert :!:
There are four lights!
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb :!:
Looking forward to summer later in the year.
happycactus
Posts: 24
Joined: Tue Sep 28, 2021 6:16 pm

Re: [Plus / 4B] Power Issue (fried TR4)

Post by happycactus »

Hi 1024MAK and Tiger, Thank you for your precious hints.

As you said, R29 is indeed in the \WAIT line but of course, it's also on my schematics!
Schermata 2021-10-20 alle 21.19.22.png
Most likely it was R79 as for issue 4A, unfortunately I can't check in the silkscreen because it's too damaged and unreadable.

Anyway, I changed it with a similar 2R2 1/4W.

After some check, I'm pretty sure the Zener is Ko. The forward voltage is almost correct (~0.5V) but the reverse voltage is not (~0.2V). Also TR4, D19 and R79 seems the lowest resistance current path to -5V. I'm going to change it ASAP.

BC547 is a valid substitute for the ZTX213, as far as I know. VCE0 is higher, Ic is the same and pinout is also the same, when you consider that Zetex marked their transistor in the "back", that is where you expect to have the "curve" of the normal TO92 packages...

I already changed also C44 and C45, Unfortunately I don't have a capacitance meter, but I think they were ok.

Thank you... I keep you posted !


Kind regards.
User avatar
1024MAK
Posts: 5103
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: [Plus / 4B] Power Issue (fried TR4)

Post by 1024MAK »

happycactus wrote: Wed Oct 20, 2021 9:47 pm BC547 is a valid substitute for the ZTX213, as far as I know.
BC547 is an NPN. ZTX213 is a PNP…

If you want a European PNP you want a BC556, BC557, BC558, BC559 or BC560

Mark
ZX81 Variations
ZX81 Chip Pin-outs
ZX81 Video Transistor Buffer Amp

:!: Standby alert :!:
There are four lights!
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb :!:
Looking forward to summer later in the year.
User avatar
PokeMon
Posts: 2264
Joined: Sat Sep 17, 2011 6:48 pm

Re: [Plus / 4B] Power Issue (fried TR4)

Post by PokeMon »

In my eyes there is an error in the drawn schematics of TR5. Emitter and base is on the same potential so the transistor is never conductive. I guess that the emitter has to be connected to the +5V line as in the issue 4A. Or this is a Sinclair trick (you never now) but for now this circuit is not working for me.

.
Bildschirmfoto 2021-10-20 um 22.42.10.png
User avatar
1024MAK
Posts: 5103
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: [Plus / 4B] Power Issue (fried TR4)

Post by 1024MAK »

Yes, that’s another error on the schematic. The emitter should connect to the +5V rail.

Mark
ZX81 Variations
ZX81 Chip Pin-outs
ZX81 Video Transistor Buffer Amp

:!: Standby alert :!:
There are four lights!
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb :!:
Looking forward to summer later in the year.
happycactus
Posts: 24
Joined: Tue Sep 28, 2021 6:16 pm

Re: [Plus / 4B] Power Issue (fried TR4)

Post by happycactus »

1024MAK wrote: Wed Oct 20, 2021 9:54 pm
happycactus wrote: Wed Oct 20, 2021 9:47 pm BC547 is a valid substitute for the ZTX213, as far as I know.
BC547 is an NPN. ZTX213 is a PNP…
Mark
Typo. I meant BC557. I must not write about electronics after 9pm.
happycactus
Posts: 24
Joined: Tue Sep 28, 2021 6:16 pm

Re: [Plus / 4B] Power Issue (fried TR4)

Post by happycactus »

Ok, today first I removed the two failing dram Chips (IC8 and IC9). I didn't tell that checking for continuity in parallel to C44 and C45 showed a shortcut somewhere. After the change the check passes, no more shortcut in C44 or C45.

Then I changed D19 with a huge 5V1 5W zener (the only sample I found). Still No -5V

I then changed the two diodes D11 and D18 (1NB4148) and the -5V was ok!

I tested the input to the RF modulator. The PAL signal was there, though a bit disturbed.

I disconnected the RF modulator, bypassed the connector with 100uF as usual, and checked the signal on my monitor.

What I found is a very noisy image. Sometimes the boards seems to reset -- but the recap of all remaining capacitors (C34, C50, C74, C27, C28, C65, C25) seems to have fixed a bit the reset issue. Still there's a lot of noise.
20211021.05.Disturbi Video 3.jpg
20211021.04.Disturbi Video 2.jpg
20211021.03.Disturbi Video 1.jpg
Next steps are:

- change TR1 and TR2. Check the power lines near the modulator
- remove the old ROM, modify for EEPROM, put a diagnostic ROM

Anything else?

Thank you
happycactus
Posts: 24
Joined: Tue Sep 28, 2021 6:16 pm

Re: [Plus / 4B] Power Issue (fried TR4)

Post by happycactus »

Ok I'm back to this PLUS 4B with the fried TR4.

So after having -5V fixed, I have recapped all the electrolytic capacitors, there was a slightly better behavior, the board seems not to reset anymore but there's a lot of noise in the video image, and it seems colors are completely gone.

What I noticed is the 12V has a lot of noise and it's around 9.2V instead. It's an issue I think, and may explain all the problems I'm seeing, DRAMS requires VSS minimum 10.8V and we are well below.

Indeed the voltage is 9.6 on average, but there's a low frequency oscillation between 9.2 and 10V... not good.

I removed the ROM, modified the board for hosting an EEPROM, put a diagnostic ROM but nothing changes.

Changed TR1 and TR2 with BC549.

There is something I can't catch. The resistors net R58 and R59 is the feedback network for the 12V. The base should be controlled to be at 5V +Vbe(TR5) = 4.3V, Indeed it is at 4.2V. This means that on C44 we should have 4.3V / 0.36 = 12V.

I checked the values of R59 (1k8 2%) and R58 (1k0 2%) and they are ok. This doesn't make sense, because if I have 9.6V on C44 I should have 9.6V * 0.36 = 3.4V... that is Vbe = -1.54V!

I also checked with an Oscilloscope.

In the following images there are the traces of the 12V (output of C44) and the Collector of TR4 / C78.
LabNation_Screenshot3.png
And I have compared with the other 48k Issue 4A that's working perfectly.
LabNation_Screenshot6.png
We note that the pulses are areound 60us but they should be around 80us to have 11.4V.

Any idea?

Thanks in advance!


Federico
Post Reply