Volatile Memory question

[JMACgyver]

Hi guys, back again.

I've got a recordable picture frame I'm giving as a gift (with a custom recording) that would lose the recording every time the recipient had to change the batteries (I've tested it). Would it be as simple as adding a capacitor in parallel to the battery holder to take over the voltage drop during battery replacement? And would it be something i could keep hooked up to it inside and not drain the batteries at the same time? I'm thinking of one of those low-voltage, small form-factor supercaps.

Could it really be that easy? If, not please explain why.

Thanks,
--Electro--
aka David M

[pebe]

David,
Just a thought. Does the frame have an on/off switch. If so, is the recording held until you switch on again?

If so, it probably contains a micro that the switch merely puts into the standby condition. A micro will normally consume very little current in standby and a capacitor across the battery leads would probably be good enough.

[JMACgyver]

[quote="pebe"]David,
Just a thought. Does the frame have an on/off switch. If so, is the recording held until you switch on again?

If so, it probably contains a micro that the switch merely puts into the standby condition. A micro will normally consume very little current in standby and a capacitor across the battery leads would probably be good enough.[/quote]

Peter,

No, it's constant on I believe, using miniscule power when not operating. You push and hold one button to record, one button to play, and one button to erase. The outstanding feature of this item is it is multi-addressable. It gives you 12 seconds total recording time. Press and hold 'record' for a three-second message, and then, when you press and hold it again, you go into the remaining 9 seconds.

(Note: these functions are all preceded with beeps denoting button press and/or memory location)

So, let's say you record three 3-sec messages. to play them, you press and release the 'play' button once. this lets you hear the first message. when you press 'play' the second time, you hear the second recorded message. And so on. When you press 'erase', it erases the first message, press it again, and it erases the next message, and so on successively. I just have two 6-second long messages recorded, that, as denoted in my other post about the multivibrator switch pulse, I want them to play as if being pressed by the momentary push button, but using a lever type switch that will remain in a ON, or OFF, position at any one time. When the item resting on the switch is lifted, the first message is played; when the item is returned to its original place, the second message will be played. Rinse and repeat.

I was hoping this recorder kept the messages without batteries (as most do) but sadly, it does not. This project will be sent cross-country to its recipient, and I cannot let its memory lapse, or the gift will be useless (without the message)

Thanks for any insight,
--Electro--
aka David M.

PS: I've been wracking my brains trying to come up with a small-form-factor, low-friction, reliable, momentary switch activated by linear travel of less than 1 inch. If you can help with that as well, you'd be my hero.
--------------
(reading headlines)
McCroskey: "Passengers Certain to Die?"
Kramer: "Airline Negligent?!"
Johnny: "There's a Sale at Penney's!!"
- 'Airplane!', 1981
--------------

[pebe]

Hi David,

I am trying to visualise how the switch would be operated. Can you explain further?

When you say 'momentary', do you mean on for about 500ms as in your other posting?

[JMACgyver]

Peter,

Ok, I'll try to explain it the best way I can.

First, I mean momentary as in a normal push-button switch. as seen here:
http://www.sparkfun.com/products/97

Second, please forget about the 500ms thing. I only mentioned it as a way for the circuit to avoid being mistaken for a spurious signal filtered out by the original circuit (OC)'s 'debounce' (if any).

Second second, (lol) this is the type of switch I am planning to use (or similar):
http://www.sparkfun.com/products/98

In either case, the OC is expecting the switch to be closed, then opened. The physical 'trigger' of the replacement switch (RS) will only close the switch, not open it again.

Think of it as a talking coaster. You set the coffee cup onto it, the switch gets pressed, and the OC says something. You pick up the coffee cup, the switch gets unpressed, and the OC says something again. In both cases, the OC needs to be tricked into thinking that that switch is being pressed and released.

Therefore, I need the circuit you gave me to interface between the OC and the RS to cause the closure of the RS to appear as a momentary close. I will be building that 4070 circuit this week to test.

That was the gist of my "Odd 'one-shot' requirement " post. However, [i]this[/i] topic involves whether or not a supercap can help keep power to the OC in the event the batteries need changing, without drawing excess current, and decreasing the batteries lives. If the batteries need changing, and there is no backup, the talking coaster will lose the ability to say *anything* when the switch closes, as the recording in it's volatile memory will have disappeared.

I hope this has helped you understand my dilemma, and I do thank you for all your help.
--David M.--

[pebe]

David,
Check your PM inbox.

[JMACgyver]

Cool, thanks Peter.

When I downloaded the datasheet for the 4070, it gave me the pinout for a 4077 also, which looks like an XNOR, which in the same configuration as your original diagram, would output low instead of high.
But your new diagram looks a bit cleaner. Please verify I am right in that the new circuit doesnt use the 10K and 4K7 as the original did?
And thanks for noticing the SPDT nature of the switch. I picked that hoping it would make activating the state change easier.

As to the supercap, I suppose I will have to put a diode in there to keep it from trying to charge the batteries if they get lower than the charge on the cap.

Regards,
--David--

[pebe]

David

Yes, I can confirm that with the 2way microswitch those two resistors are not required, and remember the 4070 can be damaged by static if handled carelessly.

Not sure about the wiring of the supercap.
Can you tell me, 1) the normal battery voltage, 2) the lowest voltage before a battery change is required, and 3) the minimum battery voltage that will retain the recordings in memory?

[JMACgyver]

[quote="pebe"]David

Yes, I can confirm that with the 2way microswitch those two resistors are not required, and remember the 4070 can be damaged by static if handled carelessly.

Not sure about the wiring of the supercap.
Can you tell me, 1) the normal battery voltage, 2) the lowest voltage before a battery change is required, and 3) the minimum battery voltage that will retain the recordings in memory?[/quote]

Yes, I noticed the CMOS nature of the chip, and will take ESD precautions. As far as the rest, the norm V+ is 4.5volts. not sure of the answer to the other questions; I dont have a variable power supply handy. I have tried a few high value caps (1000uF, 1800uF, 2200uF @6.3v; you know, the normal computer mobo ones) and they seem to keep it long enough to hold the RAM to change the batteries. I do think I will be ordering some .5F or 1F ones at 3.6 and 5v though. I will be adding a 1n4007 in series with the neg. battery lead, and since the .7v drop, will be upping the supply voltage to 6V. It seems when it drop to around 3v, the circuit gets a bit wonky.

And a big Thankyouthankyouthankyou! to you for that switch circuit. i was able to breadboard it the other day, and tested it on the o'scope, and it does exactly what i was looking for. I still havent been able to check the current draw yet, but i think it will work for my needs.

Had a question though. I have come across a couple of pocket size voice warpers, and was thinking if I wire them both up to my project, I could get the full memory available for both the up position and the down position. *but* I would need to use your original one-sided circuit. *and* I would need two separate triggers, so they dont activate at the same time. Since your original circuit used only 1/4 of the quad gate, would it be possible to alter it to fit my new requirements? Again, i'd be using a single supply, wiring the power to the boards and trigger circuit together, and also combining the speakers. the trigger outputs would be wired sperately to each board, one for up, one for down.

Thanks so much for your help!

--David--

[pebe]

[quote]Had a question though. I have come across a couple of pocket size voice warpers, and was thinking if I wire them both up to my project, I could get the full memory available for both the up position and the down position. *but* I would need to use your original one-sided circuit. *and* I would need two separate triggers, so they dont activate at the same time. Since your original circuit used only 1/4 of the quad gate, would it be possible to alter it to fit my new requirements? Again, i'd be using a single supply, wiring the power to the boards and trigger circuit together, and also combining the speakers. the trigger outputs would be wired sperately to each board, one for up, one for down.[/quote]
If you mean my circuit of August 12, then you could duplicate the pulser for each of the other three gates. I only used the other three in the later circuit because they were spare and they gave a threefold increase in output power.

[JMACgyver]

Hi pebe.

I don't think I understand enough about how the circuit works to be able to modify it the way i mentioned.

The circuit works GREAT for the recording/playback circuit board I'm using, but I've been requested to build a second one for somebody, and I dont have access to that type of board again. I'm going to need to use two separate "original circuit" boards (one for the 'up' message, one for the 'down' message) so my new request is as follows:

1) I'd like to use the spare 4077's i picked up to simplify getting the negative pulse.
2) I need a pulse from the output to activate only once per on/off cycle.

To explain, the current circuit you graciously came up with for me triggers on both the up and down switch action. I need to get a new circuit to the point that Gate 1 triggers on the up position so Output 1 signals the first board, and Gate 2 triggers of the 'down' action, sending the signal to the 2nd board. If it could still use the SPDT switch, great. but I could go back to the SPST if needed, I'd just use two.

Hmm, just had thought,... is this the type of thing a J-K flipflop is good for?

Whatever works best and easiest to build works for me. Thank you for any help you can give me again.

--ElectroDFW--
aka David M

[pebe]

Hi David,
I'll come back to you in a couple of days.

[pebe]

Hi David,

Here is the circuit for you. It uses one of your existing 4077 ICs for each board. Although I have not shown it, both boards must share a common power supply. Here’s how it works.

For a 4077 gate, if its inputs are both low or are both high, the output will be high. If the inputs are different, its output will be low.

On the top board, with the switch at (B) the inputs will be different and the output of IC1 will be low. C1 is discharged so both inputs of IC2 will be low and its output will be high. The two remaining gates operate as non-inverting buffers so the final output is high.

When the switch is moved to contact (A), IC1 will change state, taking C1 high. The output of IC4 will pulse low until C1 has charged via R2 and taken R3 low.

When the switch is moved to contact (A) again IC1 will change state, but IC2 will stay the same because R3 is already low.

Moving the switch to (B) will give a negative pulse at the output of the bottom board, in the same way.

I hope it’s what you want.

[aadambell]

Yes it would be. The way you have gone about describing your recordable picture frame, your analogy seems to be most appropriate. It is indeed as simple as fitting in a capacitor in parallel to the battery holder to eventually take over the voltage drop while replacing the battery!

[JMACgyver]

Fabulous! Thank you pebe! The circuit works great. And only draws 50uA when idle. Now all that remains is to finish up the hardware portion of the gift in time for Christmas. (No small feat for me..I'm a perfectionist; last time i tried a gift like this, it took me past two of their Christmases and three of their birthdays to complete.)

Thank you so much for your help; couldn't have done it without you.



Regards,
--David--
---------------
Procrastinators Unite! Tomorrow!
--------------