See attached pic, on which I have numbered the components that have a number !

 

Can anyone tell me what parts I'd need for a repetitive loop, and how I'd join them to the receiver ?
Thanks,
Bob

 

That's going to be tricky. It's hard enough to reverse engineer circuits when you can clearly see both sides of the board at the same time, read all the chip descriptions, and probe for voltages at will (and I'm probably one of the least qualified people on this forum to try it, although I would certainly enjoy the attempt!) It get's significantly harder doing it remotely, with only a few pictures to go on... and it becomes essentially impossible (if it weren't already) if the images of the components, labels, and traces aren't very clear and easy to follow.

I can just about trace most of your most recent, labelled "Bell Back 2" image, but we don't have anything remotely readable for the front side of that board which connects all the open connections that we can't see from the back side, and I'd also really would like to know what the "too faint to read" chip is.

If that little chip is something vaguely related to a chip like this, which is a self contained sound generator activated by various logic signals, then one of the traces going into that chip might be a simple logic signal connection which could be hacked into relatively easily. I have no idea if that's the case, but it's a nice thought anyway...

If that much were true, which is already wild speculation, then it wouldn't be too hard to rig up an astable 555 timer circuit (easy google search if you're not already familiar with them) which would be activated the first time it received an input signal (the trace we'd cut into) and which would output a signal repeatedly, constantly telling the little sound chip to play over and over again, until a reset switch was triggered.

As is so often the case, I'm out of my depth here, but I thought I'd chime in anyway because I hate to see people left hanging, and I'm also really curious to see what the solution to this ends up being!

 

Hi ebeowulf17,

It's very good of you to spend so much of your time trying to help me, and I'm extremely grateful. I thought it would be a simple matter to just solder a couple of widgets onto the existing PCB, but plainly it isn't !

I've looked at the "too faint to read" thing through my pocket electron microscope. It says TR1894C. Then under that J709A5, with the J being preceded by something that looks like 1F but I can't be certain. Mean anything ?

 

Good work getting those numbers! I can't find a good datasheet so far, and the most promising Google hit was a bad link, but the preview text was:

Melody IC for Doorbell integrated Circuits - Tradeim.com
www.tradeim.com/product_1204065_Me...

melody IC with internal amplifier or transistor than can drive a loud speaker directly--chip on boardTR1894 36 melodies dual toneTR1892 17/25 melodiesTR1889 17/25 melodiesM602 ...

That description seems to fit nicely with how I thought this circuit might be working. I'm not sure how much farther I can get without a datasheet, but it feels tantalizingly close to doable now. I'll keep looking...

 

Here's a link to something like a datasheet :

http://yansifan.vsp.windmsn.com/product1117533.html

 

Ebeowolf17, I think I may have found the answer

http://www.ebay.co.uk/itm/171247862...49&var=470348499258&ssPageName=STRK:MEBIDX:IT

Just add a beeper, and set it to beep continuously until I turn it off ! ! !

 

Looks promising, you may be on to something there!

Having said that, I think with the datasheet you found we can add a 555 to your existing doorbell if you're still interested. It would take a small amount of trial and error experimentation, but I think it's within reach now.

 

Still very interested, because that eBay thing is overly expensive and I'd rather just add a 555 to my existing cheap doorbell !
Also, it would be great to find a simple solution. If we can put a man on the moon, we can surely make a doorbell ring continuously

 

Sorry I've been slow to respond - busy day today and likely to be tomorrow as well.

Before I proceed with this post, let me warn you that I'm making a lot of guesses and assumptions for the following plan and that there is some possibility that if you do what I recommend, you could fry your doorbell. I don't think it's likely, but it's definitely possible - I'm no expert. Having said that, if you're still on board, here are a few things you can try which will help establish what we're dealing with so we know exactly what our timer circuit needs to do:

1) Measure the voltage across the battery terminals
2) Measure the voltage across pins 1 & 2 on the TR1894C chip (top left and 2nd one down on left.)

Hopefully those two voltages are the same, and are both somewhere between 3.0 and 5.1VDC, since that's what the chip specs require. I assume the numbers will be the same, but can't follow all the traces and make sense of the whole circuit for sure, so we just need to be certain. If the numbers are identical and within the desired range, then we can proceed with the following:

3) Using a short bit of wire, paper clip, etc. touch either pin 2 or the positive battery connection with one end and either pin 3 or the top side of the item marked "221" in your drawing, which I suspect connects to pin 3 with a trace that's partially hidden under the chip, with the other end. There a few ways I might expect the chip to behave:
A) as soon as you make the connection, the doorbell sound starts and plays for one full "ring."
B) upon making the connection nothing will happen, but when you break the connection, the doorbell sound starts and plays for one full "ring."
C) the sound plays for as long as you're holding the connection, but breaks as soon as you release it.

If none of these things happens, then I've misinterpreted something and it's back to the drawing board. If A or B happens, then we'll want to rig up a timer circuit whose period covers the amount of time you want from the start of one ring to the start of the next, and the duty cycle of the timer circuit should be irrelevant. If C happens, then we'll still want the same overall time period, but the duty cycle will be important to ensure you get a full ring playback before it cuts off in between rings.

Hopefully these tests all go according to plan. Let me know what you find!

In the meantime, I'll try to brush up on the finer points of 555s again so that I'm ready for phase 2...

 

I don't possess a volt meter, but thanks to eBay I should have one by Monday latest. Luckily, I already have a selection of soldering irons, solder, bent paper clips, and a hammer in case all else fails.

For the record, here's an updated pic showing the numbers on the TR1894C chip, and I've also numbered a tiny black oblong thing (centre left) which is J6, in case it's relevant.

Skipping steps 1 and 2 for the moment and moving to step 3, I've tried A, B and C. No such luck with C, I'm afraid, but what you've predicted for A and B does indeed happen. It seems to be that sometimes it does what you describe for A, and sometimes what you describe for B.

Perhaps not surprising that results differ, what with quivering hands, shaking paper clip, eyeglass popping out, and smoking a cigarette all at the same time. I'll try again later, after 10 expressos, and give you some serious results !

 

Excellent! I prefer a or b cause it makes the timer circuit design just a tiny bit more forgiving.

Next, can you tell me roughly how long the sound it plays lasts and how long of a delay you want between playbacks, or simply how long there should be between the start of one ring and the start of the next?

 

There is a choice of 32 tunes ! The shortest, which I'm using, is a double ding-dong, i.e. ding-dong, ding-dong.

The time between the first ding and the second dong is three seconds, but it's another two seconds before the reverberation fades completely. I don't know if the reverberation is created by the chip, in which case it's part and parcel of the tune, or whether it's just an incidental side-effect and can be ignored.

Ideally, I'd like a one second gap between tunes.

I hope that ding and dong aren't Chinese expletives : -)

 

Cool, so it sounds like 3 seconds between ding and dong, plus 2 seconds of reverb, plus 1 second of silence before the next cycle = a timer with a 6 second period. I'll draw up the proposed circuit as soon as I get a chance and get back to you.

 

Thank you very much.
Bob

 

No problem, it's fun trying to figure out new stuff!

I took a stab at it yesterday, but realized almost as soon as I put pen to paper that my original idea for starting the 555 was all wrong. I think I've got the right concept now, using an scr as a latching circuit and a momentary NC switch to stop the loop. Now I've just got to draw it up and calculate some resistor values... some of which will depend on the voltage we're working with. So, I'll post a rough draft once I think I've got the idea mapped out correctly, but the actual values won't be right until I know our voltage (you said you've got a meter in transit, right?)

 

Yep, volt meter should arrive Monday latest.
I'm planning to run both the transmitter and the receiver with 12v DC, if that helps.
Have a good weekend !

 

Couldn't see an input jack in the photos, nor did I notice an on board voltage regulator. Are you sure it can take 12VDC?

At least one of the chips on that board definitely can not take 12V, so unless there's voltage regulation that I've overlooked (which is entirely possible - I've made sense of portions of the board, but not everything) you wouldn't want to hook 12V up to certain parts of the circuit.

 

I think you're right ! ! !
The transmitter is 12v, but the receiver runs off 2 x 1.5v batteries = 3v.
I have a 3.7v power supply in my box of oddments... do you think that will be OK ?

 

Are you sure that isn't backwards? Looks in the pictures like the transmitter has a few small alkalines, presumably your 3V and the receiver has a larger battery pack, maybe LiPo, maybe 3.7V? Where are you getting the 12V number? Is there an input jack marked for 12V? If so, the device must have the appropriate charging and voltage regulation circuitry, maybe on the side of the board that's not visible in the labelled pic. Have you already successfully run 12V into it?

Anyway, no need for me to speculate further when you've already got a meter on the way. We'll know soon enough!

In the meantime, I've drawn up my proposed circuit, minus most of the resistor values which will be determined based on operating voltage. I've breadboarded an approximation of it on 5V supply, and it works with a momentary 5V input pulse, but I can't say for sure if it'll work off of the receiver trigger signal on the first try. We may need to play with one or two resistor values to get good triggering.



The circuit is built around an astable 555 circuit. The 555 is powered at all times, but produces no output because the Reset Pin (#4) is held low. Once there's an input pulse on the SCR-Gate-In wire sufficient to drive 200uA-1mA through the gate of the SCR (T1,) then the SCR connects the Reset Pin to VCC (positive supply) which allows the 555 to run normally, which it will continue to do until the normally-closed momentary switch is pushed, breaking the flow of current through T1. Proper latching of the T1 SCR is ensured by R3, which maintains 5mA of holding current through T1 at all times. Probably overkill since the 555 is also drawing current, but I wasn't sure how much and took the better-safe-than-sorry approach.

If anyone else reading this thread has thoughts on this circuit, please help me out. I'm very new at all this and I'd hate to lead the OP astray!