Set the size of the fuse

Hi All

I have been reading the replies here and reading up the thermisters. I really don't want to add a circuit board as the amp is built almost ready to test the sound and finish up. I really don't have room for an extra pcb. Only place I can put is at the back of the front panel. So at this time, I try not to use resistor and start up relay type of circuit. But I would consider it if all else fail.

I experimented, It won't blow a 10A fast blow fuse, I don't have a 7A or 8A fuse, so I can't try it. I studied up the NTC Thermister where it presents higher resistance when it is cold, but getting lower when more current passes. I am thinking about just use two of the NTC thermister like this one:https://www.digikey.com/product-detail/en/ametherm/SL15-5R007/570-1070-ND/1873473

The R at 25deg C is 5ohm, I use two to give me 10ohm when first power up. I would put it in series with one of the 110VAC input line. So the max current is about 11A, then when it settle down, the resistance will be about 2 X 0.09ohm= 0.18ohm. Say max at a little higher temp, the surge current will not be over 20A, I think I can live with that if I get a 7A slow blow fuse.

Bottom line, use two NTC thermisters with a 7A slow blow fuse on the AC line should solve the problem. What do you think?

Thanks
 
Probably just the slow-blo fuse will address the nuisance fuse blowing. It's largely a matter if you actually want inrush limiting, IMO.
 
I just went out and bought a bunch of 6A, 7A and 8A, both fast and Slo-Blo fuses to test. It blew the 6A fast blow, I tried powering up when caps are discharged with 6.3A Slo-Blo, it did not burn the fuse.

I then tried a 7A fast blow and powered up 3 times from total discharged, the fuse survived.

So I know the lower limit, I should be able to use 8A fast blow and 7A Slo-Blo. I think I am still going to get some NTC thermister from Digikey and I think that should do it.

My transformer is spec for 625A, means the max operating current is 6A continuous. 7A fuse is not the end of the world as the transformer has a lot of buffer, it can handle way over 6A for a few minutes for the fuse to blow.

My power amp can easily drives over 50A, so that is never a problem. I have 5A rail fuse for each side, This is after the big filter caps, I have only 1000uF on the output board, so there is not a whole lot of charge on the board that can burn the output transistors. ( remember I have 9 pairs of output transistors).

What do you think?
 
I then tried a 7A fast blow and powered up 3 times from total discharged, the fuse survived.
There's a couple of additional items to consider. Transformer inrush current is a function of line voltage phase at switch-on time, even if there's no load. You need a lot more than three random trials to be reasonably certain of catching the worst case. Also, fuse elements suffer from fatigue over time and become unreliable if they're sized close to the limit.
 
There's a couple of additional items to consider. Transformer inrush current is a function of line voltage phase at switch-on time, even if there's no load. You need a lot more than three random trials to be reasonably certain of catching the worst case. Also, fuse elements suffer from fatigue over time and become unreliable if they're sized close to the limit.
I have been doing it a lot more times by now, both the 7A fast blow and the 6.3A slo-blo survive. I'll be doing a lot of turning on and off the next day or two, we'll see.

Also one important factor, I am going to put the NTC thermistor in series with the fuse also. Another thing is the amp is going to be with me only, so time will tell. I cannot use too big a fuse as the transformer is only 625VA, a 7A fuse already over the limit ( 115V X 7A= 805VA) of the transformer. The transformer is my bottle neck of the amp. I use minimal of 14 gauge wires in the amp, each of the output section is 9 pairs of power transistors and should be able to handle over 50A of output current. Everything else in the amp can handle a much large size fuse.

You think it is safe to use a bigger fuse than 7A for a 625VA transformer?
 
Personally, I think you're looking at it from a continuous power standpoint, but the only time it will see that is on your test bench.

Average power is only going to be something like 1/10 to 1/8th of continuous rated.

So, for normal operation your mains fuse only needs survive inrush (unless you are doing rated power sine testing) and short term peaks. Other than that it has a pretty easy average job.

I keep coming back to the rating of the fuse in my ARC 100.2 of 5A MDQ slow blow. This amp has rating of 100wpc @ 8, 200wpc @ 4, and 360wpc @ 2, with the aforementioned 130,000uf.

I think a fuse along that line (5A slow) would be fine for normal use.

If this was a control circuit transformer, I think NEC allows fusing up to 167% of rated primary current for a transformer this size. However, since you have secondary fuses too, it's different. With secondary fuses allowed up to 125% of rated current, the primary side can be fused up to 250% of rated current. Again though, that's NEC for industrial control, not audio amp. That said, NEC is all about safety.
 
One problem with thermistors on SS amps is the variable resistance they offer. It drops with heat, so basically its high resistance under low load. SS amps in AB tend to have relatively low idle current, so you may find that at idle it doesn't pull enough current to make the thermistor's resistance drop out. Essentially you're creating a higher impedance power supply by doing that. If you can use a resistor and bridge across it with a time delay relay, that would take care of inrush and the impedance issue in one go.

Class A amps will pull enough current at idle to make this a non-issue.
 
One problem with thermistors on SS amps is the variable resistance they offer. It drops with heat, so basically its high resistance under low load. SS amps in AB tend to have relatively low idle current, so you may find that at idle it doesn't pull enough current to make the thermistor's resistance drop out. Essentially you're creating a higher impedance power supply by doing that. If you can use a resistor and bridge across it with a time delay relay, that would take care of inrush and the impedance issue in one go.

Class A amps will pull enough current at idle to make this a non-issue.
I thought of this also. Mine is a high bias with idle bias current of 1.2A. So it is drawing current.

Actually I was thinking the opposite way, the NTC thermistor actually might help to stabilize the rail voltage. Think of this, without the thermistor, whenever you draw more current, the rail sag because of the impedance of the transformer. The NTC thermistor do the opposite, when it is low current, resistance is higher, drops more voltage. But when current increases, resistance decreases and less voltage dropped. This counter act the droop of the transformer!!! Only problem is there is a time delay for the thing to heat up.
 
Personally, I think you're looking at it from a continuous power standpoint, but the only time it will see that is on your test bench.

Average power is only going to be something like 1/10 to 1/8th of continuous rated.

So, for normal operation your mains fuse only needs survive inrush (unless you are doing rated power sine testing) and short term peaks. Other than that it has a pretty easy average job.

I keep coming back to the rating of the fuse in my ARC 100.2 of 5A MDQ slow blow. This amp has rating of 100wpc @ 8, 200wpc @ 4, and 360wpc @ 2, with the aforementioned 130,000uf.

I think a fuse along that line (5A slow) would be fine for normal use.

If this was a control circuit transformer, I think NEC allows fusing up to 167% of rated primary current for a transformer this size. However, since you have secondary fuses too, it's different. With secondary fuses allowed up to 125% of rated current, the primary side can be fused up to 250% of rated current. Again though, that's NEC for industrial control, not audio amp. That said, NEC is all about safety.

If I can fuse at 167% the transformer rating, I am done already!!!:banana:

Even with continuous calculation, 7A and 125VAC(max) is about 900VA, The transformer is rated 625VA. so 900/625 is about 150%.

See, I just don't know this kind of thing, how people fuse the transformer.

I should have bought a 5A slo-blo, then I can try. But if the 6.3A slow blow can survive, I'll just leave it. I have not order the in rush limiter yet, I might not even need it.
 
It doesn't counteract the droop of the transformer, it seems only to introduce droop more across the board rather than only at high output. Personally, I can't see that being a good thing.
 
Remember the purpose of the fuse is not to protect the transformer, it is for the protection of life and property (safety).

Fuse size is chosen to prevent nuisance opening while still protecting again catastrophic failures. The rating of a transformer is likely its maximum continuous rating with its short term surge current rating being higher.

As long as the windings do not act like the wire in a fuse all should be well.

I would be comfortable fusing a transformer at 150 percent of its maximum rating if it prevents nuisance opening of the fuse at turn on.

The rating on your transformer is likely a combination of it thermal and magnetic characteristics.

Generically if one can't fuse a device to reduce nuisance opening from the turn on surge and still provide the safety that is the primary purpose of the fuse, then additional measures need to be taken to deal with the turn on surge.

This is common practice.

Inrush current limiting by the numbers for passive current limiting.


If you would like to put some numbers to your inrush limiting devices.

 
I since connected both amps to the power supply and start testing. Now that I have load to drain the voltage down fast, I have been turning it on and off many times already. The 6.3A slo-blo fuse holds up. Don't think I need to do anything else.

I don't think the amp will cause any safety issue, even if I use 7A fuse, it's still way below the power capability of the wall plug. Other than the transformer that spec 625VA, everything else in the amp can run in much much higher current, and I have rail fuses to protect the amp also.
 
Softstart circuit ? A little light reading on the subject.
I've used Rods boards and circuit. On my MOSFET DIY. with a 1000VA toroidal and 50,000µf per rail.Works great.
http://sound.whsites.net/project39.htm
I was reading this for the fun of it, luckily I don't need the in rush current limiter anymore. I did not think about the over current part.

But one think I question about the article, if I use say a 12A in rush current limiter and a 7A fuse, there is no way the thermister will blow up before the fuse blow. I just don't believe this. It's all about the size of the device as it's power that blow. For example if I use this:https://www.digikey.com/product-detail/en/ametherm/SL225-R012B/570-1268-ND/865094 then with a 7A fuse, I think it's going to work and since it's 5ohm at 25deg C, current is limited to 115VAC/5=23A.
 
I am so glad I put the rail fuse on the output boards, I was testing and I shorted the output or something, sparks fly and the amp stop drawing current. When I troubleshoot, I found the +ve rail fuse blown, change the fuse and everything works again. Can you imagine without the rail fuse?
 
My amp has bare WBT type speaker binding posts so I have shorted them myself, feeling around behind the amp with a banana jack in hand while powered up. Blew my rail fuse. Glad that was the only damage. Smart design feature!
 
Back
Top Bottom