Delay circuit problem.

Other Hardware General Electronics
1

Good Evening.

I'm working on a project that requires the delayed start-up of a device (an iPod interface in this case).

When the iPod is connected, it provides 3.3V that is used by the iPod interface to determine that the iPod is connected. For reasons that I won't go into just now, I need to delay this 3.3V from reaching the iPod interface for a certain amount of time. I'm still working out the exact delay required, but for now, lets assume it's about 250ms.

With reference to the attached circuit, I'm using an LM3881 power sequencer chip to generate the delay. This is being powered by 5v from another (different) iPod interface. The actual 3.3v switching is achieved using a TS5A3166 analog switch.

Assume everything is powered by 5v, and you plug the iPod in. LM3881 EN pin goes up to 3.3V, and the timer starts. After approx 250ms, FLAG1 goes high, which drives the analog switch IN pin high, and allows the 3.3v to pass through the switch from NO to COM. As it stands, this all works as expected.

If you now unplug the iPod, the 3.3V is removed, and 3.3V_OUT goes low. This is also what I want to happen. My problems begin if you remove the 5v supply while the iPod is still connected. In this case, I also need the 3.3V_OUT pin to go low. Unfortunately it's not.

I think the problem is either (or both) of the IC's latching-up when there's an input (3.3V) above VCC (the now removed 5v). Can anyone suggest an alternative circuit that would allow the delayed turn on, and also remove the 3.3V_OUT if either the 3.3V_IN or 5V is removed ?

If at all possible I'd like to continue using the same devices, due to the fact they're both qualified for automotive use. I could probably do something similar using relays, but I don't really have the space available, and automotive grade relays are generally quite big. I'm going to have to re-make the boards anyway, so adding/removing some components is possible.

Edit: I tried to upload the LM3881 datasheet, but it was too big. Here's a LINK

Many thanks,

Ian.

Delay Circuit.png
Delay Circuit.png1774×986 15 KB

ts5a3166-q1.pdf (955 KB)

2

Maybe a silly idea, but how about running the 3.3V into two TTL NOR gates (1/3rd of a 7404) in series and then to the enable pin. The 7404 output will go off when the 5V supply to it is cut, even if the 3.3V is still on. You may need to reduce the 5V output from the 7404 to 3.3V using a resistor. If I remember correctly, TTL inputs above 2.0 volts are seen as HIGH, so 3.3V should work OK.

3

TTL has been dead for decades, you have CMOS logic today! 74HCT family uses voltage
thresholds compatible with TTL however, so that would the one to choose. You still have
the problem of the nor gates being unpowered when the iPod is plugged in though. Some
solution with diodes is more likely to be feasible?

4

Hi,
What is the interface?
Tom.. :slight_smile:

5

Thanks for the suggestions guys. I'll have a better look at logic gates. It looks like TI do some logic gates that are designed to function if the power is removed while the inputs are still driven.

TomGeorge:
Hi,
What is the interface?
Tom.. :slight_smile:

A bit of history:

I'm using an Alpine KCA-420i iPod interface in my car (BMW 3 series). This is connected to the factory stereo using another interface called an Intravee II. This converts the commands from the BMW iBus (single wire data bus), to Alpine AI-net commands that the Alpine interface understands. This allows the factory stereo (and steering wheel buttons) to control the iPod, and show track/song info on the radio display.

This all works very well, but the audio is still analog, and has to travel up and down the length of the car three times before it finally reaches an aftermarket amplifier. As you can imagine, this does nothing for the sound quality, and I'm a bit fussy in that respect.

I'm now interfacing a Pure-i20 digital iPod dock with the KCA-420i interface. The KCA-420i connects via serial to the iPod for controls and track/song data. The Pure-i20 connects to the iPod via USB for digital audio. Both serial and USB connections are in the 30 pin Apple dock connector. With this arrangement, I can control the iPod as before, but now get high quality digital audio straight into the amplifier (through a very short phono lead).

My next problem is that I've now taken the factory radio out of the audio path, so no longer have control over the volume. In standard form, the factory radio doesn't transmit volume control commands on the iBus, although the steering wheel buttons do. It was never an option on my model of BMW, but I found that if I tricked the radio into thinking it had a DSP amplifier connected, it did start to send volume control messages on the iBus.

This is where the Arduino becomes useful (although it's now a custom PCB). I've built a volume control circuit based around the PGA2311 volume control chip. The Arduino monitors the BMW iBus for volume control commands, and drives the PGA2311 accordingly. The Arduino also sends messages back out onto the iBus to keep the factory radio happy that there's a DSP amplifier connected. Every 10 seconds, the radio asks the DSP if it's still there. If it doesn't get a reply, it drops back to standard and stops sending volume commands. My circuit is also writing the current volume level to the radio display whenever it changes.

Although this is all working very nicely, it's a bit sensitive to the order things are powered up. I need to let the Pure-i20 connect to the iPod before the KCA-420i starts communicating over serial. The KCA-420 doesn't start communicating until it sees 3.3v from the iPod, and this is why I want to delay the 3.3v line going high.

The Arduino shuts everything down after 60 seconds of inactivity on the iBus. This in turn removes power from the Pure-i20, which removes 5v from my delay circuit. At this point, the 3.3v from the iPod is still present, which causes the delay circuit to stay active. If the iPod is left long enough, it will shut down the 3.3v and everything is happy. If there's any activity on the iBus before this happens, the Arduino and Pure-i20 wake back up, but because the delay circuit is still active, the delay never happens, and the KCA-420 starts serial comms before the Pure-i20 has initialised the iPod via USB. When this happens, I have to reset the iPod and power cycle everything to get it talking again.

If I can disconnect the 3.3v from the KCA-420i when the 5v is removed, it will all work fine, but the 3.3v is somehow keeping the circuit active when the 5v is removed.

Sorry that was so long.

Thanks,

Ian.

6

Hi
Fine on the setup.
All for an MP3 player. :o

Tom.... :slight_smile: