Does anyone know of a soft start board or kit for a low power amplifier (like headphone amplifier) or preamplifier (power requirements around 30W, solid state rectification)?
I know that they don't absolutely need soft start, but I would like to apply one for longevity of rectifiers and heating filaments - so I would like to apply it before transformer on mains. Easiest would be just a single NTC thermistor and a bypass switch to operate it manually. But I'd rather have automatic bypass with relay.
Would commonly used low power thermistors even work at such a low current? Inrush current is below 1A, steady current around 150 mA - would 1A rated thermistor even heat enough to drop resistance? I have 120 Ohm 1A and 33 Ohm 1.3A thermistors - would any of them be of any use in this case?
I know that they don't absolutely need soft start, but I would like to apply one for longevity of rectifiers and heating filaments - so I would like to apply it before transformer on mains. Easiest would be just a single NTC thermistor and a bypass switch to operate it manually. But I'd rather have automatic bypass with relay.
Would commonly used low power thermistors even work at such a low current? Inrush current is below 1A, steady current around 150 mA - would 1A rated thermistor even heat enough to drop resistance? I have 120 Ohm 1A and 33 Ohm 1.3A thermistors - would any of them be of any use in this case?
You don't have anything to worry about. The in rush current from heaters isn't all that much ~4X. Light bulbs have an inrush > 10X, and light bulbs last a good long while before the filament fails open. Even without any sort of soft start, your VTs will go low emission long before there's a heater failure. The only cases where I've seen heaters fail open is when connecting in a series string across the AC mains. In that case, some heaters can be over volted badly before they have a chance to really heat up. Once, this was with an old Admiral TV set where the CRT filament flared white hot on power up. That was the absolute worst VT to over volt.
It can't so any harm, but it isn't doing any good either, just adding complexity and cost to your project.
Mains side inrush current will likely be far greater than 1A - it's typically limited just by DC resistance of the primary, if the transformer saturates during application of initial mains depending on where on the cycle and what residual magnetism is in the core.
You've already identified the classic elements, such as relay bypassed primary series R or NTC thermistor. But IMO the only point to it is if the inrush current causes some problem externally, such as mains trip takeout or inducing spikes on the mains which might be damaging to other equipment. The transformer isn't 'transforming' when the inrush happens, by definition it's in saturation, I shouldn't worry about the secondary side for valve gear.
If you can be sure the transformer doesn't saturate then there are zero crossing point solid state relays which might provide some assurance if you really are concerned about the secondary rise time at switch on. Somewhere in the archive of my memory is that zero switching is the worst case for saturating the core if it is pre-magnetised, and peak switching (absurdly) is the best for avoiding core saturation - can't recall exactly why, but worth thinking about if you go down this route. Peak switching solid state relays minimise transformer inrush current, as I recall. HTH !
Bottlehead Crack had a 0.5A quick blow fuse, now they use 1A quick blow fuse - I think there can't be much more current during start except maybe for a first few cycles. I also read that low power EI transformers start quite softly, so there really shouldn't be any external problems. They really don't need soft start, this is more of an experiment than a necesity.
Wow, 0.5A quick blow suggests there really isn't any notable inrush current - presumably dc resistance of the primary is such that it inrush current self limits and/or the transformer doesn't saturate then, and so no need to be concerned about it I guess.
I spent a happy few months last year persuading a 1kW torroidal not to pop mains trips, the other end of the spectrum !
Correct, with the fundamental understanding that exciting current lags voltage by a full 90 degrees, you can see that at peak voltage your theoretical exciting current will be at zero. This will neither add nor subtract from the remanence, so is the safer time to energize.
If you energize at the zero voltage crossing, current is at its peak. This will either add or subtract from the remanence, so can produce one of two possibilities; maximum inrush or minimum inrush. So due to the random probabilities of remanence magnitude/polarity and energizing polarity, you cannot know if you have the perfect storm or the perfect calm. Safest (but not necessarily the least) inrush is therefore at voltage peak.
I've been trying to find out how the thermistors would behave in my amp, when connected to primary side of transformer...
Data for the headphone amplifier:
Bottlehead Crack, arround 30W of power from the wall:
6080 heater, 2.5A x 6.3V=15.75W
12AU7 heater, 300mA x 6.3V=1.89W
6080 cathode current, 100V<2>/3000ohms=3.33W (X2 triodes=6.66W)
12AU7 plate current, 80V<2>/22.1Kohms=.29W (X2 triodes=.6W)
Power supply bleeder, 170<2>/270kohms=.1W
Power supply dropping resistors 18V<2>/270 ohms=1.2W (X2 resistors=2.4W)
There are also very small amounts of dissipation in the rectifiers and the output load resistors and some losses in the power transformer.
Total current draw using the numbers above= 27.4W
230V, around 200mA constant current draw - 1A quick blow fuse. It has three 220uF capacitors in CRCRC B+ section at 170V.
Thermistors that I have on hand are:
EPCOS / TDK B57153S330M - 1.3A, 33 Ohm. Datasheet
Meritek SCK10-1201 - 1A, 120 Ohm. Datasheet
I wonder if anyone knows how to calculate or estimate what will be the thermistor behaviour - will the 200mA constant power draw even heat the thermistor enough to drop the resistance in any considerable amount?
I'm planing to just add the thermistor and a bypass switch, since I can't find simple enough schematic for relay for my use. I could of course just measure the behaviour, but I'm not that far in my build and I'd like to have at least a bit of theoretical idea before I proceed with this.
http://www.mouser.com/ds/2/400/NTC_ICL_S153-525550.pdf
Data for the headphone amplifier:
Bottlehead Crack, arround 30W of power from the wall:
6080 heater, 2.5A x 6.3V=15.75W
12AU7 heater, 300mA x 6.3V=1.89W
6080 cathode current, 100V<2>/3000ohms=3.33W (X2 triodes=6.66W)
12AU7 plate current, 80V<2>/22.1Kohms=.29W (X2 triodes=.6W)
Power supply bleeder, 170<2>/270kohms=.1W
Power supply dropping resistors 18V<2>/270 ohms=1.2W (X2 resistors=2.4W)
There are also very small amounts of dissipation in the rectifiers and the output load resistors and some losses in the power transformer.
Total current draw using the numbers above= 27.4W
230V, around 200mA constant current draw - 1A quick blow fuse. It has three 220uF capacitors in CRCRC B+ section at 170V.
Thermistors that I have on hand are:
EPCOS / TDK B57153S330M - 1.3A, 33 Ohm. Datasheet
Meritek SCK10-1201 - 1A, 120 Ohm. Datasheet
I wonder if anyone knows how to calculate or estimate what will be the thermistor behaviour - will the 200mA constant power draw even heat the thermistor enough to drop the resistance in any considerable amount?
I'm planing to just add the thermistor and a bypass switch, since I can't find simple enough schematic for relay for my use. I could of course just measure the behaviour, but I'm not that far in my build and I'd like to have at least a bit of theoretical idea before I proceed with this.
http://www.mouser.com/ds/2/400/NTC_ICL_S153-525550.pdf
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.