VolumeToo

Hi all,

Normally I drive with the climate display window showing, so that I can adjust the fan speed etc. quickly and easily. Today, on restarting the car after a short journey and rest of an hour or so, the audio system started up in radio mode (previous condition was 'audio off') - and I only noticed when I heard the [factory] power antenna go up. I just assumed I'd nudged the on/off volume control with my knee or something, and thought no more of it.

Later today when I started the car, I was slightly unsettled to see all the climate/heater icons greyed out, with the touchscreen totally unresponsive. It was only when I switched modes (map, audio etc.) via the hard buttons round the outside of the display, that the climate touchscreen started functioning normally again.

What I do notice on a cold start is that there is often some ripple on the display (in the form of subtle shaded bars) which do disappear after a few seconds. Again, on first depressing the brake pedal (i.e. the current drain of brake lights etc.) causes a momentary streak in the display.

This looks to me like poor smoothing somewhere in the climate/nav unit. Some months ago I did check all the caps for ESR when I had the thing out to replace the button lamps. And yes, I did change the ones I could see for premium types (it is after all what I do as part of my day job).

I am now wondering however if this is a sign of the dreaded ECU caps starting to fail. Thoughts?

I do hate this touchscreen unit however; it's an ergonomic nightmare and it's inconsistent (for example why some of the fan controls on the touchscreen and others as hard buttons?) Clearly Toyota went for showroom appeal. The '89-'94 unit was way better IMO. (But that's another thing altogether.)

Thanks all, in advance

Rich

Yamae

I don't think it's an ECU issue. As far as I have experienced, all the ECU failures of 98-00 were started the AT shift problem first when QAS capacitors began to fail.
It seems to me that the problem lies at the power supply circuit or the control circuit in the touch screen display unit.
What brand electrolytic capacitors did you use?
I suggest you to check the ripple level of the regulated DC lines using an oscilloscope. Sometimes a spectrum analyser is useful to check when an oscillation at high frequency is happening.

VolumeToo

Hi Yamae,

Thank you for your reply. Today it behaved perfectly, of course!

I only changed the caps on the top board; I didn't take the whole unit apart. They would have been Rubycon ZLG/ZLH Series (which of that brand have the lowest ESR and among the highest ripple rating other than maybe the RX30 Series). At the time I didn't notice any improvement, anyhow.

Yes, I suppose I should check the supply ripple. (But where is the supply decoupling actually done?)

Kind regards,

Rich

Yamae

I'd say Rubycon manufactures very good products and the ZLG/ZLH series are good when you think about the ESR and the ripple rating points of view. But too low ESR capacitor often causes the oscillation depending on the circuit. A circuit which is not designed to use very low ESR capacitors tend to be unstable and oscillate when those are used.

In order to use low ESR capacitors, the circuit has to have a compensating additional circuit to limit the rotation of the phase or the gain. You can find details of it here. What I worry about is that the power circuit of the touch screen display unit is one of them and causes an oscillation depending on certain conditions such as the temperature, the noise or something triggers. Other items I worry about is that the circuit is partially damaged by the static discharge when you replaced bulbs or some incorrect mounting of the capacitor was done. I mean the lead wires were too long, the polarity was reversed or a solder bridge was there.

The ECU needs an extremely low ripple voltage supply having variety of sensors and many AD converters under that. Thus the low ESR capacitors are needed to minimise the ripple voltage and the circuit is designed to work well with those low ESR capacitors. The original P series capacitors of Nichicon are now not ranked very low ESR but those were ranked so when released in the late 80s. It showed much lower ESR at higher temperature and the circuit side was well thought of the compensation in accordance with the Bode Plot. This may be the reason why the circuit does not oscillate even with the latest very low ESR capacitors. After fixing the ECU using the latest very low ESR capacitors, some people say, "My car runs as good as new or even better". This may be the result of the lower ripple level compared with the original. The very early LS400s had non QAS capacitors manufactured by Marcon but those were changed to Nichicon's QAS capacitors aiming to have better engine performance especially when cold. In other words, the initial 1UZ-FE engines were not working well enough due to the remaining ripples at the DC lines especially when cold.

As far as I know, the touch screen unit does not have any QAS capacitors and it is not designed to use very low ESR capacitors. A few years ago, when I replaced the bulbs and used LEDs instead, I checked some of the electrolytic capacitors and found that they were still OK and I dare did not replace them. As I fix different boards and I only find QAS capacitors in the ECU. Other units do not use them. The only non QAS capacitor I once have replaced was a one used in the 5V regulator circuit in the gauge cluster board. This gets heat from the heat sink located very close to it and loose the capacitance, I can easily assume.

VolumeToo

Thank you again Yamae for your detailed reply. Please allow me time to digest its contents.

I carried out the work to replace bulbs (and took the opportunity to replace some capacitors) about a year ago. Please be assured that I am experienced in through-hole and SMD rework, also understanding fully the need for ESD control. Whilst the display ripple did not improve, it didn't worsen either. (And most people would never notice it - but I am a broadcast engineer and therefore I do!) But in any case, there have been no malfunctions up until now. Therefore, I suspect (and hope) that it was something that I omitted to do, rather than did do.

Without reading further (please give me time) I would agree that a low-ESR cap may cause instability in an AC-coupling application - where the circuit design is otherwise poor - but in a decoupling/bypass application then surely the lower the ESR, the better?

Update: It occurred to me that the symptom I experienced was consistent with the car being in the first key position. So, I am now also suspecting a relay or even the ignition lock barrel switch (which has been known, I think). No more recurrences to-date.

Regards,

Richard