It helps to choose the right part yes 🙂 so a zero crossing version when one wants that function.
Yeh Folks, I had erroneously morphed a triac into an Solid State Relay, and as Jean-Paul noted, most have zero cross detect circuits and do indeed start at a zero crossing.
I get the phase shift thing and how the current is decreased after the first cycle. We'll see about starting mid cycle. - more tests.
Anyway this hole issue got me thinking about designing soft starts. See a new post I will make in the next few days.
The design will:
1. use a resistor in series for the first few cycles.
2. use a triac with zero crossing detect opto driver to bypass the resistor after a few cycles to limit 1st cycle inrush (?), for consistency, and long term reliability.
3. Keep the purchase parts price less than $15.00
4. Try with and without zero cross start and publish results.
Thanks for the participation.
I get the phase shift thing and how the current is decreased after the first cycle. We'll see about starting mid cycle. - more tests.
Anyway this hole issue got me thinking about designing soft starts. See a new post I will make in the next few days.
The design will:
1. use a resistor in series for the first few cycles.
2. use a triac with zero crossing detect opto driver to bypass the resistor after a few cycles to limit 1st cycle inrush (?), for consistency, and long term reliability.
3. Keep the purchase parts price less than $15.00
4. Try with and without zero cross start and publish results.
Thanks for the participation.
Jean-Paul, The triac I have in the amp currently doesn't appear to leak bad enough to cause any issues. I'll check by measurement though.
Back to basics: If the power switching is done with a solid state device, what is the advantage of minimizing the inrush current spike? Better sizing of the primary fuse? Anything else?
In the EU it would be required (transformers equal to or above 500VA). We like a stable 400V/230V grid with as least brown outs/black outs as possible so users must adhere to some rules and requirements.
I think it was the specific 32A SSR that had this phenomenon. When I left that company I did that project out of courtesy as no one wanted to burn fingers. AFAIK leakage was not dangerous but it was too high in the eyes of someone apparently. I heard later that all SSR's were replaced for relays...
I think it was the specific 32A SSR that had this phenomenon. When I left that company I did that project out of courtesy as no one wanted to burn fingers. AFAIK leakage was not dangerous but it was too high in the eyes of someone apparently. I heard later that all SSR's were replaced for relays...
Last edited:
That would be a good approximation, although I have no idea how it could be implemented in practice (for the inrush cycle) since the circuit would not know where the peak is.
Last edited:
Can be done as the circuit could be made to wait for an opportunity the next cycle but it is not in the scope of the OP I think. It is the first cycle that counts.
Last edited:
Can't imagine something better for an audio power amplifier, it is not very simple, though. Works fine for me, over 10 years later.
Auxiliary Circuitry
Auxiliary Circuitry
You can have a permanently-on circuit that monitors the phase or amplitude of the line voltage and fires the active switching device at whatever preset point you choose.
After further investigation, you may find that a small asymmetry of a triak or SSR makes a parasitic effect similar to a (very weak) rectification, thus induces a small DC current into a primary voltage.
That is why I would short the triac/SSR contacts after a few seconds with a good old relay... nothing beats those.
I was always fascinated with the apparent obsession for simplicity in the DIY audio communities. Must be related to the "number of nonlinear junctions the signal passes through", otherwise parts and PCB assets are pennies today.
Hi,
There a two flavors with the SSRS. You can buy then with the zero crossing built into and non zero crossing. If you want to fire the SSR at different AC angle then you use the non zero crossing. The zero crossing will be firing every time it goes to the zero crossing. Depending of the application is the type you will need to use.
There a two flavors with the SSRS. You can buy then with the zero crossing built into and non zero crossing. If you want to fire the SSR at different AC angle then you use the non zero crossing. The zero crossing will be firing every time it goes to the zero crossing. Depending of the application is the type you will need to use.
For a roundy-top power wave, like most utility AC, a differentiator will zero-cross when the wave is near its peak. The zero-cross can be easily sensed and used to poke a triac. It may need an auxiliary power supply or much brain power. Also it will tend to trigger on minor peaks (switching supply racket) so wants a band-pass before and/or after.
These days it may be cheaper to use a $2 microprocessor and keep adding lines of code to handle special cases.
Attachments
That and the less parts there are used the less parts can break down. Operation should be easy as well. I support a few elderly people with technical stuff and I know for sure that making choices will be harder when one is old. So the philosophy is simple appearing devices that make the choices (if at all necessary) automated when possible, as least devices as possible, as least buttons as possible and great quality. Stuff should have a relatively small footprint with as least cabling as possible. One would say invisible equipment would be fine but when hiding away electronics in speakers the reference point of viewing the devices is gone and it does not work out too well in practice.
Last edited:
There's a load of research papers around this I found when looking at PWM controlling inrush that look at magnetisation inrush.
Error - Cookies Turned Off
I get jean-paul's point - the DIYer only has a certain amount of time, thus high count/more complex tends to make it slower progress and more costly to adopt.
If it's not broke, don't fix it? But we have electrical lightbulbs and cars because someone make it simpler and more economical to use them than candles and horses.
Error - Cookies Turned Off
I get jean-paul's point - the DIYer only has a certain amount of time, thus high count/more complex tends to make it slower progress and more costly to adopt.
If it's not broke, don't fix it? But we have electrical lightbulbs and cars because someone make it simpler and more economical to use them than candles and horses.
Yes and simplicity always works. Sometimes things can be improved which of course is better but it may also swing to extremes that do not bring that much. One of the often found extremes being typical "improvements" that are in fact only vendor lock in features.
Yesterday I had a discussion about using DSP's and several amplifiers just because passive filters would be so inefficient. I would never tolerate 6 amplifier channels and six cables, the need to program DSP's etc. if it can be done with passive filters. It also depends on what exactly is "inefficient" 😀 One might understand all that added stuff right now and change characteristics every week because one can, but that may change in a few years. Besides that, all the demos of such complicated setups were not too convincing either as they more seemed an IT affair for those that continuously feel the need to change stuff.
The install-and-forget setup and simply use well designed good sounding equipment is rather efficient IMO.
On topic: if I could do it with a mechanical mains switch that has 3 positions where 0 is off, position 1 is with series resistor (and not a hard position as it should be switched to position 2) and position 2 is direct connected I would do it. So a 3 position switch with a 2 position feel really with a "in between" resistor for limiting inrush current.
Yesterday I had a discussion about using DSP's and several amplifiers just because passive filters would be so inefficient. I would never tolerate 6 amplifier channels and six cables, the need to program DSP's etc. if it can be done with passive filters. It also depends on what exactly is "inefficient" 😀 One might understand all that added stuff right now and change characteristics every week because one can, but that may change in a few years. Besides that, all the demos of such complicated setups were not too convincing either as they more seemed an IT affair for those that continuously feel the need to change stuff.
The install-and-forget setup and simply use well designed good sounding equipment is rather efficient IMO.
On topic: if I could do it with a mechanical mains switch that has 3 positions where 0 is off, position 1 is with series resistor (and not a hard position as it should be switched to position 2) and position 2 is direct connected I would do it. So a 3 position switch with a 2 position feel really with a "in between" resistor for limiting inrush current.
Last edited:
I looked at the idea of using a ON-OFF-ON(momentary) switch, that way you switch on momentary until the voltage in the meter is up to the set amount as it goes through a power resistor then switch over to on it switches out the resistor.
I broke the thing into two - inrush for transformer(s) with bypassing NTCs, and the other being a secondary side controlling the inrush from capacitors etc. In this case about 450-480J of capacitor energy at a max if about 1.3Arms.
In the end the current method is have four resistances (2K,1K, 500, 100) in series with a switch that by passes each. Using two 50W power resistors in parallel for each step.
The the sequence goes - 3.5K, 1.5K, 500 and then finally 100 before 0R. If you want to be smart you could switch them out toggling the 500K: 3.5K, 3K, 1.5K, 1K, 500, 100, 0R. The spikes are then kept in a zone that the secondary is happy with at around 1.4-2.4A peak before the final settling to the operational load.
Last edited: