| Electronics > Projects, Designs, and Technical Stuff |
| Overvoltage protection for 115 V transformer primary against 230 V mains |
| << < (3/5) > >> |
| fsr:
Maybe you can use a MOV or two different ones to protect the 110v/220v windings? The fuse should blow quickly when the MOV triggers. Still, the secondary side will be at higher than normal voltage until that happens, so you will need to check if your circuit can survive that, or you need another protection at the secondary side. |
| Zero999:
--- Quote from: ampwizard on January 21, 2019, 12:16:22 pm --- --- Quote from: Zero999 on January 21, 2019, 10:30:15 am ---It might be better to monitor the voltage on the DC side and switch a relay on to short the fuses on the primary side. How about designing out the problem? Switched mode power supplies with a universal 100V to 250VAC input are widely available. Noise and interference can be mitigated by careful filtering an additional linear regulators on the secondary side. --- End quote --- Thanks, that could work actually. I've ruled out a SMPS due to expense. --- End quote --- Where are you buying parts from? What output voltage/current do you need? A switched mode power supply is often cheaper than a transformer, rectifier and linear regulator. |
| ampwizard:
--- Quote from: richard.cs on January 21, 2019, 12:39:28 pm ---The transformer will saturate heavily on 230 V and draw lots of primary current, are you sure that won't blow the fuse adequately quickly anyway? If you do want to use thyristors to blow the fuse then you should latch the gate drive so they're on for whole cycles after it's triggered rather than turning off at the crossing and on again as the voltage ramps up. You are going to need large thyristors for this if you want them to outlive the slow-blow fuse, do you have the current-time curve for the fuse in question?* I definitely wouldn't use relay contacts for this, a relay would protect the amp, but likely weld and need replacing along with the fuse. --- End quote --- I'm not sure if the transformer will draw enough current to blow the fuse and I don't know how to work it out. The fuse I'm using needs 10x the rated current to open within 150 ms (worst case). --- Quote ---It would be nicer of course to switch off (or better still, not switch on to avoid having the break high current) at high input voltage, but by the time you're doing that you might as well use one of the various circuits out there for switching the primary automatically, which is another way of avoiding the problem. --- End quote --- I think this is the nicer option. This is what I need to do. I don't want to rely on the user to flick the switch. --- Quote from: Ian.M on January 21, 2019, 01:00:00 pm ---There's no need for an expensive special voltage monitoring relay as you could do the voltage monitoring on the secondary side if you use a relay to implement auto-voltage selection for the primaries. However the relay does need to be good enough to handle a large inductive load reliably at worst-case max 230V mains, so cheap no-name ones are not a good idea. It would power up with the relay off and the primaries configured for 230V. If the secondary voltage was under 59% of the expected value, and stable for a suitable period (e.g. 1 second) the relay would activate, reconfiguring the primaries for 115V and immediately boosting the secondary voltage to its normal operating range. To cover situations like a lost neutral in the USA (which can result in excessive voltage on a 115V outlet), malfunctioning generators and idiots with variacs, a secondary side over-voltage detection circuit would be advisable to turn off the relay until power has been cycled, preferably with a sufficient capacitor reserve to keep the lockout logic state for several seconds so that a seriously unstable supply doesn't cause the relay to repeatedly pull in and drop out with a short cycle time. Depending on your application, you may also need to switch power to your load circuit so it is only powered when the secondary voltage is in its nominal range. A bicolour power indicator LED would be a nice touch if you intend to fit one - red for voltage out of range or still being configured and green for on. If you want a standby indication you could revise that to red for standby and flashing orange (red + green) for voltage out of range. --- End quote --- OK I like this idea. I'm going to have a crack at putting together a design based on this idea. Thanks! |
| ampwizard:
--- Quote from: Zero999 on January 21, 2019, 01:23:14 pm ---Where are you buying parts from? What output voltage/current do you need? A switched mode power supply is often cheaper than a transformer, rectifier and linear regulator. --- End quote --- I've actually been down this road and it really isn't an option for me. I have fairly specific requirements for secondary voltages/currents including a high voltage supply for valves. I also have a relationship with a transformer manufacturer which means I can build a linear supply to my exact spec for a good price. |
| richard.cs:
--- Quote from: ampwizard on January 21, 2019, 01:50:19 pm --- --- Quote from: richard.cs on January 21, 2019, 12:39:28 pm ---The transformer will saturate heavily on 230 V and draw lots of primary current, are you sure that won't blow the fuse adequately quickly anyway? If you do want to use thyristors to blow the fuse then you should latch the gate drive so they're on for whole cycles after it's triggered rather than turning off at the crossing and on again as the voltage ramps up. You are going to need large thyristors for this if you want them to outlive the slow-blow fuse, do you have the current-time curve for the fuse in question?* I definitely wouldn't use relay contacts for this, a relay would protect the amp, but likely weld and need replacing along with the fuse. --- End quote --- I'm not sure if the transformer will draw enough current to blow the fuse and I don't know how to work it out. --- End quote --- It is calculable, but probably easier just to test it. --- Quote from: ampwizard on January 21, 2019, 01:50:19 pm ---The fuse I'm using needs 10x the rated current to open within 150 ms (worst case). --- End quote --- Do you have the curves? If you want a semiconductor to survive it's far more important to know how many milliseconds it will open in at 100x or 1000x. In terms of the heat capacity of a tiny bit of silicon 150 ms might as well be forever. Thyristors are way easier to damage than transformers. --- Quote from: ampwizard on January 21, 2019, 01:50:19 pm ---I think this is the nicer option. This is what I need to do. I don't want to rely on the user to flick the switch. --- End quote --- Agreed, it's both a nicer user experience and more idiot proof. |
| Navigation |
| Message Index |
| Next page |
| Previous page |