Soft start circuit for GainClone

[markiemrboo]

There's a few threads on here about building a soft start circuit (something which seems to have been overlooked seemingly everwhere?), but I was searching google and came across http://www.diyhifi.org/forums/viewtopic.php?t=369

This one looks within my "construction abilities", but I was just wondering if anyone could see anything clearly wrong with it? If I do build it would it be better to have three 270R 5W resistors (= ~90R?) in parallel rather than two 47R 5W resistors (= ~90R?)? I would have thought that in series the first resistor would be taking most of the heat, whereas with parallel it gets spread out evenly between the three? No?

Any input or easy alternatives would be much appreciated!

[AndrewT]

Hi,
whether parallel or series makes no difference when the resistor values and permitted power dissipations are the same.

BUT, in my opinion the thinner wire in a 270r resistor compared to a 30r (if you can find one) will be less tolerant of overload. I would use series connected, to allow the more robust thicker wire.

three //270r 5W = 90r 15W
three series 30r 5W = 90r 15W

[peranders]

You live in 230 V country.

If your wall fuses are 10 A, then the max inrush current should be max 10-15 A under 100-200 ms.

Worst case conditions are (230 x 1.4)/10 = 32 ohms

Short term power is then 3200 W

but....

if you have 100 ohms 4-5 W rating, delay time of 300-600 ms for a 600 VA transformer, the resistor _will_ last. For me it has worked every day since 1988! Rather many of my softstarter are built with 100 ohms and I have never heard about any failure.

You may increase the power rating if you have huge smoothing caps. I have 600 VA, 2 x 63 VDC, 44000 uF per rail and a 100 ohms/4 W resistor.

With 100 ohms you will get 3,2 A max.

It's nothing wrong with higher power rating but 4-5 W is enough according to practical experience.

When it comes to transient capability I'll agree with Andrew, better to connect in series but hardly any problem here.

[markiemrboo]

Thanks both for your input! I never thought about the fact that 270R resistors will have thinner wire (makes sense!). I'll stick with the two 47R's as on the schematic then.

As for delay, how might I figure out what kind of delay that schematic is giving me?


I have also come across something by Rod Elliot: http://sound.westhost.com/project39.htm The first schematic on there looks quite simple too. I'm not sure about either designs at the minute though, as the PCB on Mr Elliots website actually uses a different circuit. I don't suppose anyone here has built Rod's first schematic, the one using the MOSFET?

[munchkin]

If you are using a LM3886, National Semiconductor has you covered already. From the list of features in the spec sheet :

Supply under-voltage protection, not allowing internal
biasing to occur when |VEE| + |VCC| ≤ 12V, thus
eliminating turn-on and turn-off transients

God bless op-amp IC designers

[pinkmouse]

Unless your traffo is over 500VA or so, or you have a huge, (30-40K uF) capacitor bank, then you should be fine without a softstart.

[MikeBettinger]

Originally posted by munchkin
If you are using a LM3886, National Semiconductor has you covered already. From the list of features in the spec sheet :

Has anyone tried using an electrolytic on the mute pin for a soft turn-on for the Lm3886? Seems I saw that in one of Nationals applications notes? I thought it might be good for having the amp slow start after a power outage, which is a semi regular occurance where I live.

I just received my parts and have started building my first chip amps; I was thinking about trying this approach. Does anyone have any experience with it and are there any sonic penalties?

Regards, Mike

[pinkmouse]

I'm trying it at the moment in an active set up to mute the noise of the crossover starting up. The only issue I have found is that with the relatively low rails, (15V), I'm using I have to use a 2200uF cap to get a long enough delay whilst keeping the resistance low enough to sink enough current. I may well go down the route of using a RC delay to drive a transistor and relay simply to make it a little more straightforward.

[pinkmouse]

Here's a pic of my single sided LM4780 PCB I posted here a while back. Note the two holes for the cap just between the two resistors at the bottom, that's the space I have to work with.

[markiemrboo]

Originally posted by pinkmouse
Unless your traffo is over 500VA or so, or you have a huge, (30-40K uF) capacitor bank, then you should be fine without a softstart.

I've heard 300VA and onwards, and the power around here doesn't seem all that brilliant... so I want to be on the safe side.