Ring detector circuit, transistors not fully activating

[galaxywide]

I am attempting to build a ringer interface circuit, and I’m running into problems that are over my head.
The purpose of the circuit is to buffer the output of a central telephone system that goes to several ringers spaced throughout the building. The reason for the circuit is that whenever the line rings, it overloads and needs to be reset, thus this circuit will buffer the line and offer a lot more output capability.

The schematic is below, with a few notable exceptions:



1. The 20v diodes are 18v (what I had on hand)

2. The signal transistors are 2N4401’s

3. Since I need the output to be high when it rings, I have added another 2N4401 after the first, with the same values of resistors

4. Instead of a relay on the output, I am using a TIP32C PNP power transistor (just something I had lying about, it seems to work)

5. Instead of 5V, I am supplying it with 12V (to better trip the relay I was trying to use, prior to learning that SSR’s don’t play nice when switching DC)

6. The power transistor has its own 24V supply, as that is what I am trying to switch. The grounds are tied together however.

7. I will be adding a 555 monostable timer between the emitter of the second signal transistor and the base of the power transistor (when the phone rings, I only get out a ~.2 second chirp instead of a full ring, hopefully this will let me lengthen that pulse)

I can provide schematics of the changes if necessary, I just don’t have software for it and they will be hand drawn

Now, my problem is this: when everything is on, and the line is ringing (4N27 triggered), I only get 12V out , instead of 24. I went through and checked voltages, and it seems like none of the transistors are turning on (or off as the case may be) completely. When I tie the base resistor of the first (closest to the 4N27) signal transistor to ground (turning it 100% off), I then get the correct 24V out.

Also, it exhibits the same problem with a 5V supply, so I don’t think that’s it.

Anyone have any ideas as to what may be causing this? I’m a bit of a novice at circuit design, so any and all help is much appreciated.

 

[ericgibbs]

hi gw,
If you posted a circuit diagram of the actual circuit you are trying to debug it would be helpful.

E.

EDIT:
This is free circuit drawing software.
**broken link removed**

 

[galaxywide]

Thanks for the link, I'd forgotten all about ExpressSCH. I put together a schematic, I know it's not pretty but it's all I can do at the moment

Since it outputs to .bmp, but this forum does not allow .bmp files, I put it into an archive. Just extract and it should open fine

 

[ericgibbs]

hi,
Unzipped OK, thanks, we will look it over, let you know.

BTW: I use this free screen image grabber for most jobs including expresspcb circuits.
Set it for a GIF format output
**broken link removed**

 

[ronv]

The tip32 needs to be an npn with the load in the collector. Right now it looks like it should switch between +24 and +12.

 

[galaxywide]

@ronv, I'm not sure what you mean. As is, the circuit goes from 0v to 12v when the line rings. However, when I had this set up on a breadboard and I connected the base resistor of the TIP32 to ground, I could get 24v out.

As far as I can tell, with what little knowledge I have, it appears that the base of the TIP32 is not being pulled low enough?

Thanks for the replies so far, I really want to get this figured out!

 

[ericgibbs]

hi,
The top of R7 must go to +24V, not 12V. else the TIP will not be fully switched.

E.

EDIT:
The current thru the load resistor for a 12V and a 24V R7 connection.!

 

[galaxywide]

@ericgibbs - Thanks for catching that! However (and I feel very bad about myself now), it seems that I made an error when making the schematic, and you caught it. In the real circuit, R7 is indeed connected to 24V.

Terribly sorry about that, but thanks anyways!

Also, update: hooked up the 555 timer circuit as per the schematic, and all that happens is that the IC gets super hot (too hot to touch comfortably) in a very short time. No output whatsoever. I tried it on the bench, and it triggers reliably with voltage as low as 1.7v when turning on an LED. I measured the voltage at the collector of Q1, and it reads 2.5v unloaded with the line ringing, haven't checked it with any load yet.

I honestly have no idea what is going on here....this was supposed to be just a simple interface circuit!

thanks!

 

[ericgibbs]

hi,
No problem.

I will simulate your circuit in LTSpice, what monostable period timing on the 555 are you needing?

BTW, it not a good idea to DC couple the 555 trigger input if you want the correct timing.

I am assuming your ring voltage is around 50V.??

Whats the resistance of the Load.??

 

[galaxywide]

The period I'm shooting for is ~2 seconds, right now I think it works out to about 1.7 (47uf capacitor, 33K ohm resistor).

The nominal line ring voltage is 90V AC 30Hz, but I believe the spec has quite a bit of variance in it. The load may be rather variable (greater or fewer # of downline ringers), but at the moment it's less then 100mA.

BTW, it not a good idea to DC couple the 555 trigger input if you want the correct timing.

I'm not sure what you mean by this, could you possibly explain it a bit? I know some electronics theory, but the terms mostly escape me.

Thanks!!

 

[galaxywide]

So I've been doing some testing, and I think I found the problem: the opto-isolator is not being turned on completely! I checked it by hooking pin 1 to +12V through a 10K resistor, and pin 2 to ground; I then observed that all the downstream transistors were working correctly and turning on and off like they should.

Next question: how do I fix this? How do I make the AC signal more fully turn on the 4N27?

thanks!

 

[ericgibbs]

hi,
This circuit will give the results you want.

Note the changes to the 10K to 1K and the 470nF's in parallel in the Zener ringer section, also the capacitor coupling to the 555 Trigger. There was insufficient 4N27 LED current.

Check the 0V connection on the 47uF from the Dis/Thres pin.

E.

 

[galaxywide]

Thanks so much! I'll try this and get back to you, but it looks like it will work great

I appreciate the help, and you confirmed my suspicions about not turning the opto all the way on. I knew something didn't seem quite right.

 

[galaxywide]

Quick question: I do not have on hand any 1N914's, however I do have 1N4007's and 1N2069A's.

Will either of those work? It looks like the 1N4007's at least will work, with a 1000V reverse voltage and current capacity of 1A (actually it's massive overkill I think, but it's what I have).

Thanks!

I also found some 1N5380's, not sure if those are better or worse for this application.

 

[ericgibbs]

Quick question: I do not have on hand any 1N914's, however I do have 1N4007's and 1N2069A's.

Will either of those work? It looks like the 1N4007's at least will work, with a 1000V reverse voltage and current capacity of 1A (actually it's massive overkill I think, but it's what I have).

Thanks!

hi
A 1N400x will be OK.

 

[galaxywide]

Oops, didn't see your reply. Thanks, I'll just use those then.
Hey thanks to all,it is now operating. I am a 19 yr old freshman at Mi. tech and really appreciate all of the helpful advice. cheers!