Back in 1992, the Hewlett Packard Journal published an article which explained the design of HP's legendary test instrument "HP14570A" which performs soft start. Perhaps you would feel better if you copied a design which has been verified by the reliability department of Hewlett Packard. (?)
_
_
Depends.... At full load, the AC requirement is about 10 amps at 120 volts.
If the resistor was 20 ohms, it would limit the primary current to 120/20 = 6 amps.
The power would be 6 x 6 x 20 watts = 720 watts.
That obviously assumes MAX power. So, in the WORST case, if the amp was turned on to FULL power AND the relay did NOT engage, max power dissipation would be 720 watts.
By this time howwever there would be enuf of a voltage drop across it that the amp would not be able to sustain the secondary and collapse.
In any case, I am going for a 20 - 40 ohms resistor (wirewound), at about 100 watts AND FLAME PROOF (just in case)
Thanks
If the resistor was 20 ohms, it would limit the primary current to 120/20 = 6 amps.
The power would be 6 x 6 x 20 watts = 720 watts.
That obviously assumes MAX power. So, in the WORST case, if the amp was turned on to FULL power AND the relay did NOT engage, max power dissipation would be 720 watts.
By this time howwever there would be enuf of a voltage drop across it that the amp would not be able to sustain the secondary and collapse.
In any case, I am going for a 20 - 40 ohms resistor (wirewound), at about 100 watts AND FLAME PROOF (just in case)
Thanks
I was thinking of a tube amp, the only ones I build, forgetting that solid state amps might draw much more mains current. I'd also overlooked the fact you want to turn on your whole system at once.
Years ago I saw a page by someone who did that using one of these. I think his was a GE built system , the italohm was just the first one that came up in my search.
Years ago I saw a page by someone who did that using one of these. I think his was a GE built system , the italohm was just the first one that came up in my search.
Hi cakyol,
I can guarantee that the resistor will fail open. A 5 watt one will resist nuisance burnouts. The Marantz 500 I'm working on uses what looks like a big 50 watt panel mounted unit. The turn-on surge from this amp would pop breakers if it wasn't for the soft-start. Most amplifiers (tube) would be good with a 5 or 10 watt resistor. Remember that the resistor is a lot cheaper than an output transformer or power transformer. You really, really do want this part to fail when it all goes wrong. Something is going to break under fault conditions. May as well choose where it breaks with a cheap resistor. Sometimes fuses cost more than this part.
-Chris
On average, tube amps can easily draw a lot more than a SS amp. That and their power draw is almost constant by comparison.
I can guarantee that the resistor will fail open. A 5 watt one will resist nuisance burnouts. The Marantz 500 I'm working on uses what looks like a big 50 watt panel mounted unit. The turn-on surge from this amp would pop breakers if it wasn't for the soft-start. Most amplifiers (tube) would be good with a 5 or 10 watt resistor. Remember that the resistor is a lot cheaper than an output transformer or power transformer. You really, really do want this part to fail when it all goes wrong. Something is going to break under fault conditions. May as well choose where it breaks with a cheap resistor. Sometimes fuses cost more than this part.
-Chris
On average, tube amps can easily draw a lot more than a SS amp. That and their power draw is almost constant by comparison.
For a solid state amp, I would chose a fixed beefy low ohms, WW current limiting resistor that gets shunted out by a delay relay shortly after turn-on. The reason is that if the amp is idle for a while, it will draw low current the NTC can cool down and limit the performance of the amp for a while if it is suddenly played loudly.
PS, does anyone know of how to make a symmetric power supply from a transformer WITHOUT a center tab. I decided to use a variac, big *** rectifier and big *** electrolytics which would be exactly what I want. I need up to +/- symmetric 100 volts dc at about 5 - 6 amps. I am a bit stuck on how to generate symmetry from a single ac source.
Thanks
Thanks
Here is an answer to your yes_or_no question:
Yes.
I know how to make a bipolar power supply which outputs ±V volts DC from a transformer with a single secondary having two, not three, I/O terminals. Many other people know how to do this too. So I guess the answer is: Emphatically, yes.
But, I hasten to add, every new project involves the purchase of a new transformer, so why not buy a dual secondary transformer from Antek or Plitron or Toroidy and get exactly what you want and exactly what you need?
Yes.
I know how to make a bipolar power supply which outputs ±V volts DC from a transformer with a single secondary having two, not three, I/O terminals. Many other people know how to do this too. So I guess the answer is: Emphatically, yes.
But, I hasten to add, every new project involves the purchase of a new transformer, so why not buy a dual secondary transformer from Antek or Plitron or Toroidy and get exactly what you want and exactly what you need?
Hi cakyol,
I have never seen a 33R resistor used for inrush current purposes. It will either not charge the supply capacitors enough before you must short it to keep it from burning out. Then by shorting it you will create close to he same inrush current you were trying to avoid in the first place. The other possibility is that it may burn out before you can short it anyway. With really high capacitance (a mistake in my view), you need to use the lower values of surge resistance. This is so that you can allow a meaningful charge to develop in your capacitors before you short the anti-surge resistor I would strongly recommend a 2R2 resistance for that application.
For your second question, the answer is yes. However, you won't like the answer I'm about to give you.
Rectify your voltage, assuming a bridge rectifier as your only option. So now you have approximately 200 VDC hanging around doing nothing yet. You are going to have to create a power supply that has an output of 1/2 the total supply voltage, and it has to track quickly. You will "ground" the output voltage of that circuit, creating a false ground. If done correctly you will have + / - 100 VDC that you can use. Now, the downside of this is that this regulator must be able to handle the peak current from both amplifier channels at once, plus some safety margin. This regulator is a bipolar design which is to say that it can pull up or down as needed. It is an amplifier a little more than twice as powerful as the single channels are.
So, I told you you wouldn't like the answer. It would be less expensive to buy the correct power transformer in the first place. That's why you only see this done in low power applications, like signal circuits using op amps. Increasing the voltage and current levels to supply power amplifiers gets real ugly, really fast. Don't do it.
If you don't mind me asking, why the variac? Regulation through these isn't very good, and it's an expensive transformer on top of that. Plus it might make a whole lot of electrical noise as you change the voltage due to the fact that you have a sliding contact selecting different numbers of turns in what amounts to a make before break switch after it has a few years on it.
And why the high capacitance? This is self defeating (Nelson Pass and I would disagree on this point), not that that matters. High capacitance encourages ringing in the rectifier - transformer circuit. The transformer will couple this nicely to the input side and feed hash to your world.
-Chris
I have never seen a 33R resistor used for inrush current purposes. It will either not charge the supply capacitors enough before you must short it to keep it from burning out. Then by shorting it you will create close to he same inrush current you were trying to avoid in the first place. The other possibility is that it may burn out before you can short it anyway. With really high capacitance (a mistake in my view), you need to use the lower values of surge resistance. This is so that you can allow a meaningful charge to develop in your capacitors before you short the anti-surge resistor I would strongly recommend a 2R2 resistance for that application.
For your second question, the answer is yes. However, you won't like the answer I'm about to give you.
Rectify your voltage, assuming a bridge rectifier as your only option. So now you have approximately 200 VDC hanging around doing nothing yet. You are going to have to create a power supply that has an output of 1/2 the total supply voltage, and it has to track quickly. You will "ground" the output voltage of that circuit, creating a false ground. If done correctly you will have + / - 100 VDC that you can use. Now, the downside of this is that this regulator must be able to handle the peak current from both amplifier channels at once, plus some safety margin. This regulator is a bipolar design which is to say that it can pull up or down as needed. It is an amplifier a little more than twice as powerful as the single channels are.
So, I told you you wouldn't like the answer. It would be less expensive to buy the correct power transformer in the first place. That's why you only see this done in low power applications, like signal circuits using op amps. Increasing the voltage and current levels to supply power amplifiers gets real ugly, really fast. Don't do it.
If you don't mind me asking, why the variac? Regulation through these isn't very good, and it's an expensive transformer on top of that. Plus it might make a whole lot of electrical noise as you change the voltage due to the fact that you have a sliding contact selecting different numbers of turns in what amounts to a make before break switch after it has a few years on it.
And why the high capacitance? This is self defeating (Nelson Pass and I would disagree on this point), not that that matters. High capacitance encourages ringing in the rectifier - transformer circuit. The transformer will couple this nicely to the input side and feed hash to your world.
-Chris
First, thanks for all your comments.
Now, let me explain again what I (think) I am trying to do & the reasoning behind it
I basically want to have LAB POWER SUPPLY (symmetric) with ADJUSTABLE VOLTAGE of up to about +/- 75 volts dc at about 5 -6 amps/rail. It does not have to be sooo precise that it is regulated to a few millivolts. I do not need it. This will be my lab power supply, NOT the amplifier power supply, so that I can use this in any future projects.
Nothing like that exists in the market, unless I spend over $2000. Even then, I cannot get the 2-3kWatts I would ideally like in the long run. This will also be used to test DC electric motors (totally different project). So, it is NOT just for the amplifier.
So, basically, what I am doing instead is essentially building an unregulated dc symmetric power supply with a variable dc output voltage. The variability will come from the fact that I will use a variac.
I was going to use the big rectifiers & the caps I have to smooth the supply. Thats it. I already have ALL the components except only the variac. So. it will not cost me that much.
Now the inrush limiter is for the amplifier (altho I think I may also use it for the power supply unless I turn the variac down every time before I switch on the supply).
For the amplifier, I am following approximately Rod Elliott's advice in the choice of resistors: Soft-Start Circuit For Power Amps
If I use 2R2 as you suggest, there will be an inrush of 60 amps at 120 volts and 120 amps at 240 volts supply. That is too high to be acceptable.
I am setting my timer to about 2 seconds with 33 ohms, which pulls 4 amps on the primary and will be approx 8 amps in the secondary (120/50 ratio). I think 2 seconds is long enuf to inject a burst of charge into the capacitors so that when the relay shorts, it will not look like a short circuit.
I also have an 8 second delay timer before the speakers are relayed in anyway, to avoid pops & crackles until the supply reaches the full voltage.
I may play around maybe by reducing the resistor to 20 ohms which will increase the primary inrush to 6 amps and secondary to 12 amps. I think it is pretty good for 2 seconds to charge the caps to a point where they do not look like short circuit. Besides, it is healthy to limit the cap inrush current anyway so they dont get stressed and need to be changed after 2 - 3 years.
Some other amplifier related values are:
- 250-280 watts rms/8 ohms, class AB, needs about +/70 volts dc.
- My toroid is 1800 VA at 2 x 50 volts secondary
- I am using dual rectifiers rated at 90 amps/channel in the amp.
- I have 2 x 47,000 uF 100 volt smoothers PER channel.
- I have ssr protection on the speakers with a start up delay of 8 seconds.
So, that is the summary.
Now, let me explain again what I (think) I am trying to do & the reasoning behind it
I basically want to have LAB POWER SUPPLY (symmetric) with ADJUSTABLE VOLTAGE of up to about +/- 75 volts dc at about 5 -6 amps/rail. It does not have to be sooo precise that it is regulated to a few millivolts. I do not need it. This will be my lab power supply, NOT the amplifier power supply, so that I can use this in any future projects.
Nothing like that exists in the market, unless I spend over $2000. Even then, I cannot get the 2-3kWatts I would ideally like in the long run. This will also be used to test DC electric motors (totally different project). So, it is NOT just for the amplifier.
So, basically, what I am doing instead is essentially building an unregulated dc symmetric power supply with a variable dc output voltage. The variability will come from the fact that I will use a variac.
I was going to use the big rectifiers & the caps I have to smooth the supply. Thats it. I already have ALL the components except only the variac. So. it will not cost me that much.
Now the inrush limiter is for the amplifier (altho I think I may also use it for the power supply unless I turn the variac down every time before I switch on the supply).
For the amplifier, I am following approximately Rod Elliott's advice in the choice of resistors: Soft-Start Circuit For Power Amps
If I use 2R2 as you suggest, there will be an inrush of 60 amps at 120 volts and 120 amps at 240 volts supply. That is too high to be acceptable.
I am setting my timer to about 2 seconds with 33 ohms, which pulls 4 amps on the primary and will be approx 8 amps in the secondary (120/50 ratio). I think 2 seconds is long enuf to inject a burst of charge into the capacitors so that when the relay shorts, it will not look like a short circuit.
I also have an 8 second delay timer before the speakers are relayed in anyway, to avoid pops & crackles until the supply reaches the full voltage.
I may play around maybe by reducing the resistor to 20 ohms which will increase the primary inrush to 6 amps and secondary to 12 amps. I think it is pretty good for 2 seconds to charge the caps to a point where they do not look like short circuit. Besides, it is healthy to limit the cap inrush current anyway so they dont get stressed and need to be changed after 2 - 3 years.
Some other amplifier related values are:
- 250-280 watts rms/8 ohms, class AB, needs about +/70 volts dc.
- My toroid is 1800 VA at 2 x 50 volts secondary
- I am using dual rectifiers rated at 90 amps/channel in the amp.
- I have 2 x 47,000 uF 100 volt smoothers PER channel.
- I have ssr protection on the speakers with a start up delay of 8 seconds.
So, that is the summary.
Hi cakyol,
The inrush current you quoted is peak. It very quickly falls to more sane levels. Given that we are talking about two projects at the same time it gets confusing and you will get advice that doesn't match what is in your mind. Anyway, the surge resistor is typically shorted in 1/2 a second. So you want to charge your supply up to a reasonable level in that time frame. All of a sudden, 2R2 isn't so unreasonable for a large power amplifier. For more common amplifiers, try 4R7 or 5R6 are common values. Again, your peak current exists for a very brief time. You do want to make sure your supplies are charged up to a reasonable level before you short out your surge resistor. There are no hard and fast rules here, just be in the ballpark and understand what the average power in the resistor will be. You should be able to get down to a standard 10 watt or 5 watt resistor, depending on amplifier size.
-Chris
The inrush current you quoted is peak. It very quickly falls to more sane levels. Given that we are talking about two projects at the same time it gets confusing and you will get advice that doesn't match what is in your mind. Anyway, the surge resistor is typically shorted in 1/2 a second. So you want to charge your supply up to a reasonable level in that time frame. All of a sudden, 2R2 isn't so unreasonable for a large power amplifier. For more common amplifiers, try 4R7 or 5R6 are common values. Again, your peak current exists for a very brief time. You do want to make sure your supplies are charged up to a reasonable level before you short out your surge resistor. There are no hard and fast rules here, just be in the ballpark and understand what the average power in the resistor will be. You should be able to get down to a standard 10 watt or 5 watt resistor, depending on amplifier size.
-Chris
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.