speaker protection - fail again - help please

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi

This is the third sp protection I 've been trying but I never had any one working fine. This is from :

http://users.otenet.gr/~athsam/protection_2_EN.htm

It is a modification of protection section of the PA300 with two slight difference :
1. an extra cap. C5 which is not present in PA300
2. R11 – 120k instead of 47R in PA300

I was not able to get the KTY81-122 so I did a slight alteration also :

I had the R3 and RTH(a NTC-1k instead of a PTC) swapped and I had a VR in place of R2 and R1 so that I may be able to fiddle around the voltages feeding into +- input of IC1A and hence the cut-off temperature. I tried to warm the NTC with my hair dryer and D5 did go on, signifying that the as the temperature goes on, IC1a out will go low but I don’t have the exact temperature. I think I can adjust it but I will do it when I had the other parts working also.

I measure only 11.5-12V on the zener (1n4744 which is suppose to be 15V) but I think this should be fine, isn’t it?

D6 is always on, so I had pull out R15 and R16, isolating the output of IC1b. I measure the voltage across C4 and it’s only about 0.65V where as –ve of IC1B reads about 1.2V. I try changing R13 to 1k thus lowering –ve of IC1B to about 0.7-0.8V. I change C4 to about 100u and it read about .68V when power on for about 30 sec. What can I do now? Keep making C4 larger? The time of charging will be getting longer even if the voltage can get over the –ve of IC1B, right?

Can somebody give me some help on this please. I need to have a sp protection circuit for the power amp. I’ve built (3 already and the next one is coming up)

Tks
 
Here is a simple circuit that does work:

www.darkmatter.myby.co.uk/dcprotect.png

L1 is the relay, can be any 12V type. Signal from speaker output enters at R1. You can probably use any small signal transistors for Q1-Q3 (such as 2N5551), and replace Q4/Q5 with a Darlington transistor.

The R5/R6 and C3 arrangement implements a simple 2/3 second power on delay. If not needed you can remove R5, R6, R7, C3 and Q3, and connect R4 directly to +12v and the detector should still work.

I will probably rework this circuit to run from a bipolar supply, perhaps +/-15V so that it will work from the preamp supply.
 
Unless you want to start with circuit # 4 (and the problems THAT will give you), I would disconnect all the stuff having to do with the DC protection (Q1, Q2 etc).
Since the basic temp stuff seems to work, work your way from the output of IC1a to the relay stage to see where the bug is. It is pretty useless to swap or change components at random if you don't know what the problem is.
This is not a very complex circuit, but you need to be methodical and do some thinking.


Jan Didden
 
janneman said:
Unless you want to start with circuit # 4 (and the problems THAT will give you)

Are you referring to the schematic I posted?

edit: didn't notice the original poster wanted a temperature protection. This is easy enough to add, using any old opamp (a 741 will do) configured in comparative mode, with a thermistor on one input and a fixed resistor/preset on the other. Connect it up so that when the temperature point is exceeded it pulls the output low. Connect this to the base of Q5 via a 10K resistor.
 
jaycee said:


Are you referring to the schematic I posted?

edit: didn't notice the original poster wanted a temperature protection. This is easy enough to add, using any old opamp (a 741 will do) configured in comparative mode, with a thermistor on one input and a fixed resistor/preset on the other. Connect it up so that when the temperature point is exceeded it pulls the output low. Connect this to the base of Q5 via a 10K resistor.

I have not looked at your schematic, that's not the point. If you give up on three after a problem, chances are large that you give up on #4 after a problem. Unless you sit it out and fix it.

Jan Didden
 
Hi janneman,

I totally agree with you : start another circuit, get into other or even more problems. I'll stay with this one (for I have spent quite a bit of time, putting them togher and even lay a PCB for it), Now that I've pulled out Q1, Q2, thus leaving the input of IC1B to the output of IC1A and the delay circuit. I still don;t get a voltage big enough to 'turn on' IC1B (it should be at least 1.4V according to the shcematic because at -ve input of IC1b, the volage is 1/11 x 15V = 1.36V). How to do so? Should I take out D1 also (I don't think it's necessary, but what's your opinion?) to isolate IC1B from IC1A?

I've made amp. from 1 pair of o/p's to 4 paiirs. I had been all the way successful but drowned in simple circuits like this!!!!!!!
 
This circuit has three functional inputs: the temp through IC1A, then the offset through Q1, Q2 and then the stuff through D2. You need to tackle each in turn to be sure it does what it needs to do in case of catastrophic conditions. You already took out Q1, Q2. Now still IC1B doesn't switch, my hunch is that Q3 is hard on. Take it out.
Now go back to the beginning:
Verify that V+/IC1B = 15V, and V-/IC1B = abouth 1.5V.
Heat the NTC, put a voltmeter on IC1A to make sure it switches low. If Vo/IC1A goes down towards zero, it should pull V+/IC1B down through D1 to below 1.5V, and Vo/IC1B should change state. Check that.

BTW, if your zener only gives 11.5-12V, that indicates that the series R from the supply is too high. So R17 may be too high, but that depends on your supply. Try lowering R17 to 2.2k or something to see if you can get a stable +15V supply. This is important, because other wise the nominal +15 varies with the circuit actions and that can upset the bias voltages in the opamp inputs.

Jan Didden
 
I've changed R17 to about 2.2k and now the voltage across zener is about 14.85V. I desolder 'C' of Q3 and when I throw the switch on, there is a delay of about 2 sec before o/IC1B is turned on.(D6 is on at start and stays for about 2 sec). But I was unable to change the delay time (I think 4-5 sec should be more appropiate)

I check the voltage at 'C' of Q3 (desolder from circuit) which is about 2.5mV

NTC is working fine : I heated it and D5 turns on but I don't have the temp. I felt quite hot with the NTC.
 
Seems like Q3 is being turned on - is point B connected to the AC side of the amp's rectifier? - is D2 around the right way? There should be a NEGATIVE voltage at junction of C3 & R9 similar to amp's negative rail. The voltage at the base of Q3 should also be negative, thereby turning it off.

Reducing R11 should increase the turn-on delay - I would suspect the value of 120k is some sort of misprint (given the VERY bad English translation of the text, this would not be surprising).

Cheers
Graeme
 
Hey man, try a fuse, your circuit is wonderfull, but a fuse

Can be some solution.... at least...for a while, when you go studying your problem.

Before the electronics developed into our days sophistication, some basic rules could be obbeyed, as a straight line is the smaller distance we can have between two points.

Well, each one of us with our own preferences...be happy!

regards,

Carlos
 
Tks guys for you help,

Here's what I have (with 'C' of Q3 desolder from circuit) :
junction of C3, R9 = -30V
Base of Q3 = 0.56V

Another problem :
Although D6 was off, the relay was never closed. I have
o/IC1B = 13.3V
Base of Q4 = 0.67V
junciton of R18, D4 = 10V

My DC supply voltage is about 34V. Anything wrong with R16?

I had tried changing value of R11 (I had a 47R to begin with as suggested by shematic of PA300). I had tried using a small (I had a VR in place, too much soldering and desoldering around there whick is starting to peel off the traces) value for it but I guess it is too low for getting any lower. What else can be done with this?
 
I try to test the circuit for driving the relay by doing the following : (hopefully a right way to do it)

I pull R16 out. put a 91k from +15V to B of Q4 and then a VR (50k) to ground. I adjust VR, trying to establish a voltage across B & E to turn Q4 on. But I could never get a BE voltage over 0.67V. I've check that Q4 give some hfe reading. It couldn't be trashed, or could it? What might be happening here? Please help me out. I don't want to start another one without knowing what's going on here.

Tks
 
With -30V at junction C3, R9 you will need to increase resistor R8 to 68k - this will then give slight negative voltage at Q3 base to keep it off until the power is switched off.

The base of Q4 will never go over 0.67V, and at this point Q4 should be fully turned on - measure the collector of Q4, it should be close to 0V. Depending on the voltage and resistance of your relay, R18 may be too large to allow the relay to operate - measure the voltage across the relay when D6 is off, and if it is less than the relay's nominal voltage, decrease R18.

A circuit such as this one will require variations to the values of a number of parts depending on the supply voltage it is working from - it certainly isn't a universal fit.

Cheers
Graeme
 
Hi BigPanda,

I verify the value of R8 and R9. I arrived the same conclusion as Centauri. You can increase R8 as he suggested or reduce R9.

As far as the relay problem is concerned. The base bias curent is about half milli-amp with the current circuit component values. The max collector current will be depending on the beta of the relay transistor. They are normally between 50 - 100. So you should expect a max current of 25 to 50 mA to drive the relay. You said that the voltage at the junction of R18 and D4 is 10V, the power supply rail voltage is 34V and R18 is 1.5 Kohm. So the current thru R18 is 16 mA. Based on these observations, I think your problem could one of the following:

1. D4 is reversed.
2. The relay needs more than 16mA to maintain closure.
-> You need to check the relay spec to see if 16mA is sufficient. Or pull the relay out and check how much current it sucks in across a 24V power supply.
-> Check the polarity of the driving transistor. A simple meter check will do. You may have the C-E reversed. The gain will be very low.

Also, the 15V zenor voltage is only 12V means that the circuit sucks in more current than the designer expected.

Good luck. BTW, I am from HKSAR too..

Kam
 
I put back Q1, Q2 and applied +- voltages to the amp. connector. Everything is fine now. The only thing that's bothering now is the delay time when the power is on. I tried to put a 10R for R11 but doesn't seems to affect the delay too much. Anything I can do to improve this by manipulting R10, R11

Tks again.
 
Hi BigPanda,

I looked at the circuit again. I could not figure out how C4 can provide the time delay. Even before C4 has a chance to charge, voltage at the junction of R10 and R11 is already higher than the voltage at the junction of R12 and R13. The relay will be on after the power is on. It's a simple voltage bridge. Time delay is not possible.

Hope changing the relay resistor to 220 ohm will not over-stress the relay. It should not be the case normally. But I would check to ensure as this is one of the critical protection mechanisms. Too much to take.

You may want to re-check the design of this circuit. Or at least do a thorough check to make sure it functions as expected. A number of designs seem to be kind of odd to me.

For the zener question. My wild guess would ok given it's a 5W. At 15V, it should be able to allow about 330mA thru at 25C. Normal bias current for a zener is about 20mA. You should still have lot of room. A good practice for diode and zener is to keep the lead longer for better heat dissipation. Try to calculate the total current each circuit takes in. You then will have a better idea and good nite sleep.:)


BTW, I am in Calgary.

Good luck with your diy... Kam
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.