Hello Everyone,
I'm planning to put in a soft start circuit for the class A/B amp I am building now, which has 2 x 350va toroidal transformers and total 80,000uF of smoothing capacitance (20,000uF for each power rail).
I want to go on the safe side and the one by Ron Elliott seems to be a decent and simple enough one to build. (see figure 5 on Soft-Start Circuits)
Ron's notes suggest that the soft start time is 100ms, but I am thinking of increasing that to 500ms to 1sec, but I am not sure which component / value to change to enable a longer soft start time.
Can someone help me with some advice?
Many thanks and have a safe weekend.
I'm planning to put in a soft start circuit for the class A/B amp I am building now, which has 2 x 350va toroidal transformers and total 80,000uF of smoothing capacitance (20,000uF for each power rail).
I want to go on the safe side and the one by Ron Elliott seems to be a decent and simple enough one to build. (see figure 5 on Soft-Start Circuits)
Ron's notes suggest that the soft start time is 100ms, but I am thinking of increasing that to 500ms to 1sec, but I am not sure which component / value to change to enable a longer soft start time.
Can someone help me with some advice?
Many thanks and have a safe weekend.
Last edited:
Hello Bansuri,
I've done some reading and yes you are right.
I'll still to what's stated in the document. Many thanks.
Do you know how should I connect the ground of the soft start circuit?
Those encapsulated auxiliary transformer has just L and N, with no ground nor 0v. Should the ground of the circuit be connected to the chassis ground?
I've done some reading and yes you are right.
I'll still to what's stated in the document. Many thanks.
Do you know how should I connect the ground of the soft start circuit?
Those encapsulated auxiliary transformer has just L and N, with no ground nor 0v. Should the ground of the circuit be connected to the chassis ground?
The same circuit can be used for a speaker protection relay at the output, then your 1+sec makes more sense.
Just parallel from R1 to the relay and use bigger C and R as described.
You do not need to ground the circuit, the transformer and the relay coil are isolated, so the thing can float and there is less danger to make a short by touching an element with a probe ground.
Just parallel from R1 to the relay and use bigger C and R as described.
You do not need to ground the circuit, the transformer and the relay coil are isolated, so the thing can float and there is less danger to make a short by touching an element with a probe ground.
Last edited:
The soft start delay depends on which soft start device (resistor or NTC/inrush limiter) you use. I have a pretty comprehensive analysis available here: The Ultimate Guide to Soft Start Design – Neurochrome
The best way to optimize the soft start is by measuring the current into the transformer primary during startup. You can do that with a simple current probe that you can build for less than $10: The $10 Current Probe - How to Measure Current with an Oscilloscope – Neurochrome
Tom
Actually it is. So are the 2x350 VA transformers. In the soft start article linked to in the previous post I measured 110 A inrush current from a 400 VA transformer. That's just the current needed to magnetize the core. It does not include the current needed to charge the supply capacitors.
Tom
Last edited:
> I measured 110 A inrush current from a 400 VA transformer.
So?
I have 44 Amp surges all day long. Well-pump. It won't wear-out my wires. I do get lamp-dim, mainly because my power drop is so very long and thin.
If a 400VA transformer makes a large surge, how about the 25,000VA transformer at the street? Or the several dozen along my street? How can the power company handle the surge after a power failure?
So?
I have 44 Amp surges all day long. Well-pump. It won't wear-out my wires. I do get lamp-dim, mainly because my power drop is so very long and thin.
If a 400VA transformer makes a large surge, how about the 25,000VA transformer at the street? Or the several dozen along my street? How can the power company handle the surge after a power failure?
The surge only happens when switching on durimg the first cycle when the core is unmagnetized, maybe every second time. A humming noise is heard for a moment.
All the caps won't saturate the core and the peak current is much lower than before.
Anything in front of the primary is also bound to fail, the resistor burns out, the relay contacts get welded together, so reliability is less.
Actually when nothing will happen without it (blown fuse) I would save the money.
All the caps won't saturate the core and the peak current is much lower than before.
Anything in front of the primary is also bound to fail, the resistor burns out, the relay contacts get welded together, so reliability is less.
Actually when nothing will happen without it (blown fuse) I would save the money.
So? Do you use a fuse in your DIY amps or do you just have a rusty nail in a fuse holder? Do you prefer to specify a fuse that will actually be protective if something goes wrong within the amp? If your answer to this last questions is 'yes' I recommend adding a soft start.
First off, 110 > 44, no? Secondly, I never said going without a soft start would wear out your household wiring. I'm curious where you got that from.
Over time you can wear through the enamel in a toroidal power transformer due to the mechanical stresses on startup. What do you think generates the hum? You don't have to believe it. Your belief. Your problem. One of my electronics teachers in the 1980s warned me about this back then. He did eventually have to replace the 500 VA toroids in his (then 25+ year-old) DIY power amps as the primary was shorting to the core. He didn't include a soft start.
That sounds like a great question to ask a power engineer. Also, how often do you turn your amps on/off? Once or twice a day? How often do you have a power outage? In most areas of the western world hardly ever. You think that might make a difference?
Oh, and by the way: Next time you need to take a leak, do so in your bathroom. There's no need to pee on me.
Tom
True. That hum is caused by the primary winding stretching. This causes the primary to vibrate and rub against the core. There's insulation there and it takes a long time for this to cause the transformer to fail, but it will eventually fail. Unless you leave your amp on all the time, of course.
The transformer is bound to fail without a soft start too. And there are ways to prevent the relay contacts from burning out and ways to extend the life of the relay by lowering the power dissipation in the relay coil once the relay has switched on.
I would too **IF** the fuse can survive without having to be grossly oversized. That's how I've arrived at the ~200 VA limit for going without a soft start. With the smaller cores, you can just double the ampacity of the fuse and be fine. But for the larger cores, you need to significantly oversize the fuse to prevent a nuisance blow. The oversizing in turn makes the fuse less protective.
I agree.
Tom
Last edited:
1. Elliot says above 600W
2. I have seen many amps fail, but 95% were transistors or diodes and 5% resistors and caps. Never had a transformer failure. Even the OP seem to recycle some old stock.
Transformers usually overheat and fail, but not from a single cycle.
3. The mains switches fail when switching off by arcing and inductive kickback. There are rugged versions available tho.
2. I have seen many amps fail, but 95% were transistors or diodes and 5% resistors and caps. Never had a transformer failure. Even the OP seem to recycle some old stock.
Transformers usually overheat and fail, but not from a single cycle.
3. The mains switches fail when switching off by arcing and inductive kickback. There are rugged versions available tho.
Toroidal mains transformers are usually manufactured to maximize VA/size (=$
) Working at the magnetic limits of the core and with no gap, there may be a surge, especially if it happens to connect at the peak of the mains voltage. Slow start circuit will help start up toroids >400 or so VA. Since E I cores effectively have a small gap, slow start at primary is usually not required. Rectifier diodes is usually the weak link with regard to soothing cap surge current.
) Working at the magnetic limits of the core and with no gap, there may be a surge, especially if it happens to connect at the peak of the mains voltage. Slow start circuit will help start up toroids >400 or so VA. Since E I cores effectively have a small gap, slow start at primary is usually not required. Rectifier diodes is usually the weak link with regard to soothing cap surge current.- possibly, not "at the peak" but at the zero (or closer to the zero). Because inrush current is lagging on the primary voltage.
Here is a good article.
And Rod Elliot investigation says about this fact too. (Look at Figure 3A - Measured Transformer Inrush Current)
Last edited: