I want to make up the following circuit to protect my speakers while powering up my pre-amp.
Would this relay be a suitable candidate (to work both channels)? (I intend to use a 12 volt supply)
An externally hosted image should be here but it was not working when we last tested it.
Would this relay be a suitable candidate (to work both channels)? (I intend to use a 12 volt supply)
An externally hosted image should be here but it was not working when we last tested it.
Would the relay be in series with the output of the preamp or with the output of the amp? If it is the preamp I don't see any major problems. If it is to be in series with the amp output then they are waaaayyy too small, you will need something a lot beefier.
One thing I would add is a diode to prevent the IC from being damaged by the kick back of the relay coil.
Hope this helps!
Sébastien
One thing I would add is a diode to prevent the IC from being damaged by the kick back of the relay coil.
Hope this helps!
Sébastien
Sorry Sébastien,
I didn't explain fully. The idea is to short (to ground) the output of the pre-amp/buffer until the power supply has 'settled'. So the relay would be normally closed, then after a short delay the relay is activated and opened, breaking the short.
So the relay does not have to handle much power.
Thanks for the tip about the diode.
Now that I think of it, the resistor would be better in series right after the output of your buffer. No matter what load you will apply to your buffer, your buffer would always be protected.
As for the value, it greatly depends on the device you are going to use as a buffer and what voltage your supply is going to be.
For something with high drive capabilities (i.e. BUF634) I'd say anything between 75 and 100 ohm would do alright (I personally used 100 ohm because that is what I had on hand with excellent results). With a power supply of 15+15V that would limit the output to 200mA.
For something of a little less powerful, (i.e. OPA2132) you need to limit the output to about 40mA, the resistor should then be (for the same supply) of 375ohm.
All those calculations are made assuming you already have some signal applied to your preamp and that your gain makes it so that the output of the buffer is close to the rails. In practice this will rarely happen, especially at DC where the current would be hard to supply.
So my best advice would be to go for a value low enough to not make your damping factor too low, but high enough to protect your buffer in most cases. 50 to 100 ohm is what I've used in many occasions with good results.
These values allow your buffer output to be limited to a few hundred mA, more than enough to drive even hard loads (which shouldn't happen considering most amps have input impedance in the tens of kilo ohms). And when shorted to ground, the buffer would still be limited thus protected even with quite heavy signal applied.
I hope all of this is clear and will actually help you instead of confusing you more than anything
Sébastien
As for the value, it greatly depends on the device you are going to use as a buffer and what voltage your supply is going to be.
For something with high drive capabilities (i.e. BUF634) I'd say anything between 75 and 100 ohm would do alright (I personally used 100 ohm because that is what I had on hand with excellent results). With a power supply of 15+15V that would limit the output to 200mA.
For something of a little less powerful, (i.e. OPA2132) you need to limit the output to about 40mA, the resistor should then be (for the same supply) of 375ohm.
All those calculations are made assuming you already have some signal applied to your preamp and that your gain makes it so that the output of the buffer is close to the rails. In practice this will rarely happen, especially at DC where the current would be hard to supply.
So my best advice would be to go for a value low enough to not make your damping factor too low, but high enough to protect your buffer in most cases. 50 to 100 ohm is what I've used in many occasions with good results.
These values allow your buffer output to be limited to a few hundred mA, more than enough to drive even hard loads (which shouldn't happen considering most amps have input impedance in the tens of kilo ohms). And when shorted to ground, the buffer would still be limited thus protected even with quite heavy signal applied.
I hope all of this is clear and will actually help you instead of confusing you more than anything
Sébastien
Thanks again Sébastien,
This is one of the buffers that I use.
It already has the 68R in series with the output so would it be OK if the relay shorts the 100K to ground?
G4ME,
The relay contacts are closed to begin with. This shorts the output of the buffer/pre-amp. The circuit 'waits' for a set time and then opens the relay contacts, the short is broken and the signal can travel on to the power amp.
You can order through Farnell internationally and I believe Newarkare in the USA and are part of the same organization. If you can't find that exact relay, one with a simialar spec will do.
This is one of the buffers that I use.
An externally hosted image should be here but it was not working when we last tested it.
It already has the 68R in series with the output so would it be OK if the relay shorts the 100K to ground?
G4ME,
The relay contacts are closed to begin with. This shorts the output of the buffer/pre-amp. The circuit 'waits' for a set time and then opens the relay contacts, the short is broken and the signal can travel on to the power amp.
You can order through Farnell internationally and I believe Newarkare in the USA and are part of the same organization. If you can't find that exact relay, one with a simialar spec will do.
If you are talking about the opamp in the pre-amp then I would gusess it would depending on how quickly the relay reverts to closed mode.
If you are talking about the chip in the amp, then no it wouldn't make any difference and you would need to use something that disconnected the speakers when the power is removed.
Hello again Nuuk,
My guess is that this 68ohm resistor has been placed there for this exact purpose. So yes, I think it would be safe to clamp the output to ground. To make sure of that we need to know the type of transistors you plan on using. Once we know that we'll be able to lookup their SOA for 20V, derate it a bit and figure out how much current they can put out without burning. (The output capacitor will prevent the buffer from emitting DC into the ground so even if your input has some DC component it won't overheat the transistors.)
Hope this helps!
Sébastien
My guess is that this 68ohm resistor has been placed there for this exact purpose. So yes, I think it would be safe to clamp the output to ground. To make sure of that we need to know the type of transistors you plan on using. Once we know that we'll be able to lookup their SOA for 20V, derate it a bit and figure out how much current they can put out without burning. (The output capacitor will prevent the buffer from emitting DC into the ground so even if your input has some DC component it won't overheat the transistors.)
Hope this helps!
Sébastien
Well this is the difference between those of us who mess around, and you experts!
I have downloaded the datasheets for the ZTX653 and BC547. I can't see SOA on the latter but for the ZTX653 the SOA graph is for collector current not output current. Hence I am now bewildered again!
I use both ZTX and BC547's in my buffers and also have some of the BFQ232's that were originally specified for this circuit (yet to try them)!
I have downloaded the datasheets for the ZTX653 and BC547. I can't see SOA on the latter but for the ZTX653 the SOA graph is for collector current not output current. Hence I am now bewildered again!
I use both ZTX and BC547's in my buffers and also have some of the BFQ232's that were originally specified for this circuit (yet to try them)!
OK, I am still a real noob, but....
If you use a resistor clamped to ground like that won't the resistance just be in parallel with the load? The load would still see some signal right? It just seems to me that breaking the circuit to the load would be better, especially if this was for a preamp.
In me previous post about what happens on power down I was referring to a preamp.
If you use a resistor clamped to ground like that won't the resistance just be in parallel with the load? The load would still see some signal right? It just seems to me that breaking the circuit to the load would be better, especially if this was for a preamp.
In me previous post about what happens on power down I was referring to a preamp.
The current will take the line of least resistance, ie through the shorted link (closed relay) to ground.
It is better to use this arrangement because you are not adding another switch in the signal path.
Yes, if the relay closes straight away, the offset produced when the power is withdrawn from the pre-amp circuit will be shorted to ground and you shouldn't hear those noises from your speakers. So you can see that this is quite a neat addition to our systems.
Hi Nuuk,
The collector current is what matters here. How much current are those transistors going to have to sink and at what voltage. For the BC547 you have to be careful because it can only sink about 100mA (MAX). The ZTX643 on the other hand can sink up to 2A (MAX).
Keep in mind that these calculations are worst case scenarios.
The BFQ232 are specified for 300mA (MAX) and the 68ohm limits the output current (considering ±20V PSU) to 295mA thus in the safe zone (not mine but probably ok for the original designer). Following the SOA graph shows that for 20V, the Ic is around 150mA the transistor could potentially be damaged but a 20V swing is never going to happen so with the 68ohm resistor the transistor should survive.
Following this logic, if you use the BC547, you should increase the value of R3 to around 200ohm. This will limit the current sinking to 100mA with ±20V supply. The ZTX643 will probably survive with the current value of 68ohm.
Hope this helps!
Sébastien
The collector current is what matters here. How much current are those transistors going to have to sink and at what voltage. For the BC547 you have to be careful because it can only sink about 100mA (MAX). The ZTX643 on the other hand can sink up to 2A (MAX).
Keep in mind that these calculations are worst case scenarios.
The BFQ232 are specified for 300mA (MAX) and the 68ohm limits the output current (considering ±20V PSU) to 295mA thus in the safe zone (not mine but probably ok for the original designer). Following the SOA graph shows that for 20V, the Ic is around 150mA the transistor could potentially be damaged but a 20V swing is never going to happen so with the 68ohm resistor the transistor should survive.
Following this logic, if you use the BC547, you should increase the value of R3 to around 200ohm. This will limit the current sinking to 100mA with ±20V supply. The ZTX643 will probably survive with the current value of 68ohm.
Hope this helps!
Sébastien
Thanks once again Sébastien.
I can now (almost) go ahead and get the relays and give this a try. Then I will write it all up on DD!
Just one more thing. Looking at the circuit diagram at the start of this thread, it specifies a supply of 5-15 volts. Is the voltage supply for the relay adjusted by the 100-470 preset?
I can now (almost) go ahead and get the relays and give this a try. Then I will write it all up on DD!
Just one more thing. Looking at the circuit diagram at the start of this thread, it specifies a supply of 5-15 volts. Is the voltage supply for the relay adjusted by the 100-470 preset?
Not really, they both get power from the same +5-15V. Only the relay's power is conditional to the state of pin 3 of the NE555. This pin is either low or high depending on the state of the threshold.
I think you should examine the following website:
http://www.uoguelph.ca/~antoon/gadgets/555/555.html
Hope this helps!
Sébastien
P.S. I'm trying to simulate your buffer to see what gives
I think you should examine the following website:
http://www.uoguelph.ca/~antoon/gadgets/555/555.html
Hope this helps!
Sébastien
P.S. I'm trying to simulate your buffer to see what gives
My simulation gives pretty good results. Actually with the values as is, using the BC547, even clamped to ground, the circuit would dissipate just a little heat but nowhere near any dangerous point!
The output capacitor does need to be of good quality, but it could be lowered to 1uF without too much effect on the sound.
Hope this helps!
Sébastien
The output capacitor does need to be of good quality, but it could be lowered to 1uF without too much effect on the sound.
Hope this helps!
Sébastien
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.