When turned on, my amp makes a substantial "bonk" sound accompanied by a dimming of the lights. Since it is power cycled about twice a day on average, I would like to reduce the audible and visible effects and reduce wear and tear on the amp's PS. So, I'm building a resistor-&-relay-style soft-start circuit for it, but I am unsure about an appropriate resistor to use. The amp is class AB, has an EI transformer, has 60 V rails, and uses 15000 uF per rail. Line voltage is 120 VAC.
I will be using a slightly modified version of ESP project 39, and was thinking of using a 25 ohm, 25 watt aluminum chasis mount resistor like these ones, that I picked up at a local surplus shop. I was thinking of a switch time of 0.5 to 0.6 seconds.
Have I made any glaring mistakes or does the plan seem sound? Is 25 ohm to little? Is 25 W too small?
I will be using a slightly modified version of ESP project 39, and was thinking of using a 25 ohm, 25 watt aluminum chasis mount resistor like these ones, that I picked up at a local surplus shop. I was thinking of a switch time of 0.5 to 0.6 seconds.
Have I made any glaring mistakes or does the plan seem sound? Is 25 ohm to little? Is 25 W too small?
25 ohms is a little on the low side. Make sure the resistors are rated to full mains AC voltage. If they are not rated enough simply put two in series. 25W is plenty for such a short duration.
Hi,
I am on 240Vac and would use 50ohm. 120Vac will need a little less.
I think you could try 0.5S delay but 1S might be better.
Have you got an analogue Multi meter? Try measuring the volts dip at your power outlet at switch on. If the lights dim it indicates that your fuse board is also losing voltage at switch on. Maybe that's what happens on 120Vac.
I am on 240Vac and would use 50ohm. 120Vac will need a little less.
I think you could try 0.5S delay but 1S might be better.
Have you got an analogue Multi meter? Try measuring the volts dip at your power outlet at switch on. If the lights dim it indicates that your fuse board is also losing voltage at switch on. Maybe that's what happens on 120Vac.
I have commonly seen 50 to 100 ohm in 240 V soft start circuits. That corresponds to 12.5 to 25 ohm at 120 V. The ESP project uses 50 ohm (3 x 150 ohm, 5 watt, in parallel) and is for 240 V. This will result in the same inrush current but twice the wattage as my proposed value.
The resistor is rated 520 V with 2250 VAC dielectric strength.
The lights dim because they are (unfortunately) on the same circuit as the amp. The dimming is momentary (< 1 sec.) and is only due to the inrush current, which the soft-start is intended to fix.
Hi,
maybe the rules here in the UK are different but we are not allowed to wire anything else into a lighting circuit.
The safety reason is to prevent a fault knocking out the circuit breaker and killing all the lights.
Yes I would expect the dimming to be all over within a second or less. That is why you can turn off the soft start resistor so quickly.
maybe the rules here in the UK are different but we are not allowed to wire anything else into a lighting circuit.
The safety reason is to prevent a fault knocking out the circuit breaker and killing all the lights.
Yes I would expect the dimming to be all over within a second or less. That is why you can turn off the soft start resistor so quickly.
That's a very rational code to have, and I honestly do not know if we have a similar code here or not. In this case the room is lit (mainly) by lamps plugged into wall outlets. These outlets are on the same circuit as the outlet that the amp plugs into.
In any case, I'll give it a go with the 25R 25W resistor, and a ~1 second switch time.
25 ohms will work but is a bit larger than needed. Manufacturers designs commonly use 3 ohm to 10 ohm resistors, usually 10 to 20 watts. Consider that even a resistor as small as 3 ohms will reduce the inrush current by a huge amount... I generally use 7.5 ohm resistors.
You don't need a timer for the relay, but you can certainly go that route if you like, such as the project you mentioned. I personally go for the simple route by powering the relay from a supply rail, and delay the closing of the relay with a capacitor, usually 1000µf to 2200µf, depending on the value of the current limiting resistor for the coil. Less parts, less complexity, less fuss.
BTW, you will need a saftey cap across the relay contacts, or they will destroy themselves little by little from arcing.
YMMV...
You don't need a timer for the relay, but you can certainly go that route if you like, such as the project you mentioned. I personally go for the simple route by powering the relay from a supply rail, and delay the closing of the relay with a capacitor, usually 1000µf to 2200µf, depending on the value of the current limiting resistor for the coil. Less parts, less complexity, less fuss.
BTW, you will need a saftey cap across the relay contacts, or they will destroy themselves little by little from arcing.
YMMV...
An externally hosted image should be here but it was not working when we last tested it.
relay saftey cap
What should the value of this saftey cap be? I have seen caps across relays before and always wondered why they were there. Does the value change for low power signal relays as opposed to high power ones? Do tell.
What should the value of this saftey cap be? I have seen caps across relays before and always wondered why they were there. Does the value change for low power signal relays as opposed to high power ones? Do tell.
.01µf seems to be the popular value on power switches and relays (10,000pf). In the Digikey catalog, it is the largest saftey cap they sell from Panasonic.
I agree totally with EchoWars. I have seen from 2R2 to 10R in this application. 10 to 25W. The resistors are only in circuit a short time and the relay contacts tend to last in excess of 20 yrs. Same for the power switch.
If you get an output short, the relay will tend to drop out and burn out this resistor. Good saftey feature as long as the resistor is a no burn.
-Chris
If you get an output short, the relay will tend to drop out and burn out this resistor. Good saftey feature as long as the resistor is a no burn.
-Chris
Some manufacturers put a thermal fuse in contact with the resistor to save it. Usually as much of a PITA to change the fuse as the resistor...
Hi EchoWars,
I agree. I'd rather just replace the resistor along with whatever else. Easier to find than a thermal fuse sometimes.
-Chris
I agree. I'd rather just replace the resistor along with whatever else. Easier to find than a thermal fuse sometimes.
-Chris
Re: relay saftey cap
Sorry for the thread hijack macboy...I'll go away now...
Guess I didn't answer the 'Q'. If the relay is actually carrying an audio signal, I don't think you'd want a cap on it.lgreen said:
Sorry for the thread hijack macboy...I'll go away now...
Inrush limiter circuit
It seems that the circuit of post #7 is missing the "peak reverse" diode across the DC relay coil...
It seems that the circuit of post #7 is missing the "peak reverse" diode across the DC relay coil...
Not needed...but add it if it gives you a warm fuzzy. 
You could also use a zener across the coil.
You could also use a zener across the coil.
Re: Inrush limiter circuit
Think about this: What exactly will a diode across the coil do?
There is no need for that because the relay coil circuit hasn't got any breaking element but if you feel better it will not do any harm either.fab said:
Think about this: What exactly will a diode across the coil do?
di/dt isn't high enough to generate back emf. Even if it does, it's highly damped by the power supply. I would think the cap across the coil enough for the job.
-Chris
-Chris
I've prototyped and tested the soft-start and was impressed enough by it to permanantly install it. I used a 24R, 10W coffin resistor. It does exactly what it is intended to do, and even after repeated starts, it does not heat up.
The circuit is based on the ESP project mentioned earlier, specifically, the "PCB version" at the end. I have modified it to operate on a constant 12 VDC power supply (the pre-existing standby supply in the amp), along with a logic-high trigger signal. The existing power relay is powered from the same supply and activated by the same signal. I found that the suggested component values gave a very fast switch time, so I increseased R1 to 33k, giving about 0.8 second switch time. I like the fact that this circuit has an extremely fast dropout and consistent turn-on time even after a very brief AC interruption.
The circuit is based on the ESP project mentioned earlier, specifically, the "PCB version" at the end. I have modified it to operate on a constant 12 VDC power supply (the pre-existing standby supply in the amp), along with a logic-high trigger signal. The existing power relay is powered from the same supply and activated by the same signal. I found that the suggested component values gave a very fast switch time, so I increseased R1 to 33k, giving about 0.8 second switch time. I like the fact that this circuit has an extremely fast dropout and consistent turn-on time even after a very brief AC interruption.
- Status
- Not open for further replies.