If you know of a better one which has these characteristics, I would be glad to hear from you:
- SIMPLE (no software, no processors) circuit whose PCB I can actually buy online.
- latches after a fault condition.
- works with about 35 VDC (rectified from a 24VAC transformer, which is the easiest voltage to find).
- Can have about 8 - 10 seconds startup delay to avoid turn on thumps.
- Can detect loss of AC instantaneously to avoid turn off thumps.
- (Desirable but not crucial) an LED comes ON when the circuit detects a fault. This is available at diyfan site on Didden's circuit by a simple modification & transistor addition.
- SIMPLE (no software, no processors) circuit whose PCB I can actually buy online.
- latches after a fault condition.
- works with about 35 VDC (rectified from a 24VAC transformer, which is the easiest voltage to find).
- Can have about 8 - 10 seconds startup delay to avoid turn on thumps.
- Can detect loss of AC instantaneously to avoid turn off thumps.
- (Desirable but not crucial) an LED comes ON when the circuit detects a fault. This is available at diyfan site on Didden's circuit by a simple modification & transistor addition.
Last edited:
On the input side, I have a simple inrush limiter with a 15 ohm 12A thermistor and a shorting timer delay relay (airotronics MC1004531J).
I will set this to about 4 seconds, which will also limit the inrush capacitor charging on the secondary, and set the spkr protection to about
8 seconds to make sure the caps are charged to a point where the speakers do not thump. I may have experiment with timer values to get it right.
I will set this to about 4 seconds, which will also limit the inrush capacitor charging on the secondary, and set the spkr protection to about
8 seconds to make sure the caps are charged to a point where the speakers do not thump. I may have experiment with timer values to get it right.
Last edited:
So you could use quite a low current rating fuse/CB on the AC feed to your amp, and within the amp on the primary side - eg. down at 120Vac and 4A, and with a fast-blow or standard C-curve characteristic ? That is one of the major benefits of using some active limiting at power on, and provides backup protection if the limiting circuit fails, and much better discrimination for a fault within one of the 500W amps.
With such a powerful stored energy and a powerful transformer, I would pay attention to professional types of switching from the Siemens Sirius series. They have semiconductor contacts for switching inductive loads (at the input of the transformer), and also capacitive loads (after the transformer). Regarding protection, all protection developers are so passionate about them that they no longer care about music and quality. Sometimes their protections resemble fire extinguishing systems and not an amplifier.
I hear you but with speakers worth $10,000+, their protection becomes extremely important
I looked at the siemens contactors. They seem to be a bit of an overkill I think my 25 amp relay will do the job quite adequately.
Last edited:
For the topic of AC mains over-current protection, you have a resistor limiter to totally manage current during any transformer in-rush period, and any bulk DC filter capacitance charge period. Switching the mains AC on is now effectively a resistive load (eg. close to AC1), with peak current limited by the resistor.
If you aim to use a 15 ohm resistor then the initial peak must be less than 8A (assuming 120Vac), and likely significantly less.
If the resistor is an NTC then you really need to estimate the let-through energy to the DC filter capacitors to confirm the NTC won't blow up. If that is too difficult for you then I suggest you identify the NTC datasheet, and confirm your mains ACV and secondary ACV, and I can do the calc based on charging 520,000 uF to 95 volts DC with no load. PSUD2 may also be able to estimate the charge time. The NTC does introduce risk for a hot re-start, and that risk does seem to go against your aim of a reliable setup. A 12A NTC may also not be a good selection unless your amp maintains an AC mains current above circa 2.5A at all times.
If you aim to use a 15 ohm resistor then the initial peak must be less than 8A (assuming 120Vac), and likely significantly less.
If the resistor is an NTC then you really need to estimate the let-through energy to the DC filter capacitors to confirm the NTC won't blow up. If that is too difficult for you then I suggest you identify the NTC datasheet, and confirm your mains ACV and secondary ACV, and I can do the calc based on charging 520,000 uF to 95 volts DC with no load. PSUD2 may also be able to estimate the charge time. The NTC does introduce risk for a hot re-start, and that risk does seem to go against your aim of a reliable setup. A 12A NTC may also not be a good selection unless your amp maintains an AC mains current above circa 2.5A at all times.
no, i am going to use arduino base, i have a small current traffo made up, and at 10 amps produces 3.5 volts, just right for a logic 1.....
Note that the thermistor will be shorted out after 4 seconds. After that it doesnt matter what current value it can sustain
But 4 seconds of excessive let-through energy could kill it - that is why they have a max capacitance type rating in the datasheet.
There is also the scenario that someone turns off the amp after about 3-4 seconds, then back on after about a second or two - or whatever situation that leads to a hot thermistor being asked to again charge up the caps. If you have a soft-start controller, and are using NTC, then I'd recommend you interlock the amp from being allowed to turn-on again for about 3-5 minutes (or at the very least the thermal time-constant in the datasheet).
There is also the scenario that someone turns off the amp after about 3-4 seconds, then back on after about a second or two - or whatever situation that leads to a hot thermistor being asked to again charge up the caps. If you have a soft-start controller, and are using NTC, then I'd recommend you interlock the amp from being allowed to turn-on again for about 3-5 minutes (or at the very least the thermal time-constant in the datasheet).
I need the 4 seconds to also limit the capacitor inrush current on the secondary. There is almost half a farad on the secondary which will is a short circuit as soon as power is applied. By limiting its charge current by controlling the current flow in the primary, I will soften the blow on the capacitors, rectifier and the secondary of the transformer.
Most likely, speakers are not the most expensive item in your home. Your amplifier reminds me of a volumetric blast warhead (slightly weaker than an atomic bomb). Switching the arc inside such ammunition from speaker to ground, as described in this speaker protection article, is not the smartest decision. Accidents in amplifiers and other power devices do not occur from scratch. Conditions for their occurrence systematically, day after day, are created by improper operation of the equipment. As a result, the equipment wears out. For example, for clarity, after each hard start, microcracks appear in the insulation of the transformer wires. Something similar happens in capacitors. The resistor does not guarantee that the transformer will turn on correctly at the correct phase time. Professional equipment ensures reliable operation of such equipment, so that in the future there is no need for a fire extinguishing system at all. For example, in a Millenium amplifier, there is no relay at the output of the amplifier. But there is an effective trigger protection. But this is a low power amplifier with 120 watts per channel. Your amplifier should have a completely different level of protection.I hear you but with speakers worth $10,000+, their protection becomes extremely important
here is a calculator for capacitors charging to the supply rail...Capacitor Charge and Time Constant Calculator - Electrical Engineering & Electronics Tools
This is one feature that I was able to include into my recent design
Splitting the power at entry, fusing each trafo independently, soft starting both trafos simultaneously with CL-90's. I even have the process reversed for poweroff for in the event of a short or other fault so that the shorting relay can release in the event of an overload. Then the relay with the CL-90 can release, being current limited.
Included in the image is the protection circuit that I use. Simple and effective. Uses a latching relay and on it's own, does not need anything other than 5vdc in my case. Other changes could possibly make it work for another voltage. Standalone, when tripped, the relay will short out the enable SSR for the speakers, effectively disabling it and the speakers.
My system uses a Raspberry Pi for controlling the power cycling and additionally will kill the AC to the amps in the event of a DC fault. I had to design my own DAC (not shown) to fit my requirements in the same project. When I get the final finished, I will post photos and info. An engineer over at the RPi foundation has developed a custom device tree overlay specifically for my project and apparently has also created a new functionality for the Pi device at the same time in the process. If successful, I also intend to alter some of the functionality of my own personal design into something for the community that will perform all of the same functions - minus the amp of my own. Essentially, what you would see on this device is the RPi for receiving and playing music data, all of the power cycling functions and speaker protections that I use. This way, what a DIYer could do is connect their power, input and speakers to my device, then the speakers to the device as well. Essentially, this will control soft start, amp and speaker protection all in one, with an added streaming media capability. I don't want to get too deep into talking about it all just yet because there is still some ways to go following my most recent model, so don't hold your breath ever.
You can check it out here...
Amp description - YouTube
S
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.