what good is blowing a fuse in series with an MOV??
That permits the biggest part of the spike to get through? No?
Or are you saying put the fuse in series with the load and have the MOV blow that??
I think it is not a bad idea to put in a super heavy duty high current MOV, since they do degenerate with pulses over time... but I haven't read anything on their degeneration WRT the size of the MOV vs. the current that hits... that would be good info.
_-_-bear
That permits the biggest part of the spike to get through? No?
Or are you saying put the fuse in series with the load and have the MOV blow that??
I think it is not a bad idea to put in a super heavy duty high current MOV, since they do degenerate with pulses over time... but I haven't read anything on their degeneration WRT the size of the MOV vs. the current that hits... that would be good info.
_-_-bear
Yes. Put the MOVs at the end of the fuses, across the load receptacle. Inline holders work OK. If you're going to supply loads with no ground don't even run that out to the receptacle. It could be just another arc path you don't need. Put a plastic plug down in the hole so no one relies on a connection that isn't there. Put a spark gap across the receptacle too if you want.
If you can get a fast acting type fuse rated slightly above normal running current to hold the inrush for several years, that's the one to use. Nothing gets out of the way of current faster than vapor and the element has to hit that state in a piece of a millisecond to do real well at keeping energy off the arrestors/clamps. Maybe a nice expensive breaker can put a lot of pulse energy into throwing the contacts far away in that short of a time, but I don't think the ones I've seen in a 20 dollar power strip can race a fast acting fuse. Especially not less than a 5A AGC. They might Start opening right away, but it'll hold a low voltage arc for too long.
If you can get a fast acting type fuse rated slightly above normal running current to hold the inrush for several years, that's the one to use. Nothing gets out of the way of current faster than vapor and the element has to hit that state in a piece of a millisecond to do real well at keeping energy off the arrestors/clamps. Maybe a nice expensive breaker can put a lot of pulse energy into throwing the contacts far away in that short of a time, but I don't think the ones I've seen in a 20 dollar power strip can race a fast acting fuse. Especially not less than a 5A AGC. They might Start opening right away, but it'll hold a low voltage arc for too long.
Last edited:
I cheated. The attached file is actually postscript but your PDF reader should open it.
Fusing the neutral might not be legal everywhere, but depending on where the lightning hits your neutral and ground might not be the closest thing to ground anyway, and inside of 10uS there's probably not too much help from it due to inductance. One time I had something smack the phone line and the arc jumped off the modem board (this was a while ago) and onto the bottom of the computer case. From there it put a pin hole on the side of the case to the inside of a steel desk with a shelf at the front underside that the PC was sitting on. It look like someone started a TIG welder on the computer. There was nothing around the desk but air, in a very old house with asbestos tile flooring over tongue and groove hardwood. It must have been the capacitance of and other little ionized paths off the desk that loaded the case. I was in another room and heard a sharp crack. If anything went out the ground wire in the supply it didn't matter. The PC locked up but rebooted OK. The modem was gone.
Fusing the neutral might not be legal everywhere, but depending on where the lightning hits your neutral and ground might not be the closest thing to ground anyway, and inside of 10uS there's probably not too much help from it due to inductance. One time I had something smack the phone line and the arc jumped off the modem board (this was a while ago) and onto the bottom of the computer case. From there it put a pin hole on the side of the case to the inside of a steel desk with a shelf at the front underside that the PC was sitting on. It look like someone started a TIG welder on the computer. There was nothing around the desk but air, in a very old house with asbestos tile flooring over tongue and groove hardwood. It must have been the capacitance of and other little ionized paths off the desk that loaded the case. I was in another room and heard a sharp crack. If anything went out the ground wire in the supply it didn't matter. The PC locked up but rebooted OK. The modem was gone.
Last edited:
doesn't open in 9.0
I still am not sure what you're suggesting.
A fuse is not fast enough to stop a HV pulse from lightning, afaik...
By the time an MOV has shunted enough current to flash a fuse, a whole lot of current has passed, so IF the MOV has shunted it, then the danger is gone, if not, then the fuse won't likely help... but I'd like to see what your pdf shows, maybe I am not "getting it".
_-_-bear
I still am not sure what you're suggesting.
A fuse is not fast enough to stop a HV pulse from lightning, afaik...
By the time an MOV has shunted enough current to flash a fuse, a whole lot of current has passed, so IF the MOV has shunted it, then the danger is gone, if not, then the fuse won't likely help... but I'd like to see what your pdf shows, maybe I am not "getting it".
_-_-bear
If you want dood lightening protection talk to people who operate radio transmitters in Florida. Orlando sees a lightenig storm almost daily every summer afternoon and antenna up on towers are easy targets and they all have big coax cables that lead right to some expensive gear. They can design systems that can take direct hits.
The think that seem to work is to provide a very low impedance path to ground using a conductor that can take lots of current. I use coax feed line and connect the shield to a field of three 6ft groud stakes just outside the house. These in turn are bonded to the building's ground system. The hope is that a direct strike will travel on the outside of the coax and then into the ground stakes. Well at best it can only divide the voltage and I might still get a few KV in the house. The coax then right after it passes through the wall goes into a comercially made lightening arrester. This is basically a gas discharg tube The tube and low value resistor from a voltage divider. the tube normally looks like an open circuit but when the gas ionizes is a good path to ground. There is also an LC network to shunt transsients.
Ths white paper pretty much describes current best practice. The ideas in this paper are well tested and known to work
http://www.dxengineering.com/pdf/Installing%20Lighnting%20Protection.pdf
In the past I owned a sailboat. It is not fun sitting under 65 foot aluminum mast in the thunder storm at sea. You are the tallest thing around for many miles. But it can be safe again the idea is to offer the lightening a very easy path to the water. The boat had a 6,000 pound lead keel that extended about 6 feet under the hull. Connect that to the mast with some solid copper strap and it's as safe as can be. Old timmers would wrap a chain around the mast and put the other end of the chain overboard. This gets most of the current off the boat. Same idea as the tower ground system in the above white paper.
Bottom line is that it's common for electronics to take a direct hit and survive and yu don't have to invent protection scheems
Fuses have limited voltage flash over arcing. A big voltage spike will vaporize the fuse filmament and turn it into ionized gas that conducts current. Breakers are even worse, and have very slow switching times. Thermal breakers take a couple of seconds, a Magnetic breaker is around 500 ms.
They rely on the tower to take the bulk of the lighning strike. They usually protect the attenna and isolate the coax cable from the tower. Gas Discharge tubes are used to divert any surges that are conducted into the coax cable.
http://www.sylvador.com/samples/man_light.pdf
For Lightning, Gas Discharge tubes work very well. For Solar flares, EMP, Gas Discharge tubes are too slow.
Here is an article that describes various Surge protection devices and some of the common circuits for Surge protection:
Multiple Protection Devices Guard Against Transients Page of
It is a violation of the NEC 2008
250.52 (B)
Not permitted for use as Grounding Electrodes
The following systems and materials shall not be used as grounding electrodes:
(1) Metal underground gas piping systems
(2) Aluminum
Bear, Dratt! 8.0 works in linux. I'll post in pdf soon.
A "real" MOV can handle 20kA pulses with 200J energy for under 5 bucks.
Here's just an example.
Digi-Key - 495-3647-ND (Manufacturer - ETFV25K130E4)
You probably couldn't find any glass if an AGC fuse had to try pushing that into it, about the same energy as a small handgun. The fuses would more than share the energy dissipation (that's the point of blowing a fuse with an MOV, anything to keep the pulse from getting out of the clamp and into the equipment), a sacrifical part anyway, may as well go out with a bang. MOVs are sort of like lamp filaments. Several times below their max ratings they can run for a very long time, over the spec kiss it goodbye.
They don't put a good fraction of that much MOV in your typical 20 dollar power strip. If they did we probably wouldn't be having this discussion.
A "real" MOV can handle 20kA pulses with 200J energy for under 5 bucks.
Here's just an example.
Digi-Key - 495-3647-ND (Manufacturer - ETFV25K130E4)
You probably couldn't find any glass if an AGC fuse had to try pushing that into it, about the same energy as a small handgun. The fuses would more than share the energy dissipation (that's the point of blowing a fuse with an MOV, anything to keep the pulse from getting out of the clamp and into the equipment), a sacrifical part anyway, may as well go out with a bang. MOVs are sort of like lamp filaments. Several times below their max ratings they can run for a very long time, over the spec kiss it goodbye.
They don't put a good fraction of that much MOV in your typical 20 dollar power strip. If they did we probably wouldn't be having this discussion.
Last edited:
How realistic are those MOV ratings anyway ? The part specified in the digikey link has 20kA specified for 20uS only!
20uS doesn't seem long enough to me.
I ask because I once built a surge protector using a bank of 20mm varistors (431K20D, rated 6.5kA) and it wasn't of much use. Whenever there was a spike, the varistors would blow up with a very impressive bang even though the current through them was limited to 4A by a fuse. (Which seems quite reasonable given the total device dissipation is specified as 1W)
Last edited:
agreed.
Afaik this is NOT how it is done. The feedline is BONDED to the base of the tower, and the transmitter building is also bonded to the tower's ground system often via a radial.
You can read online what the major commercial transmitter mfrs say about this...
_-_-bear
But only once. In an average device with MOV protection, the MOV are frequently tripped as induction loads (refridge, AC, etc) during the day. Each time the MOV trips it degrades reducing the maximum energy and current it can clamp. Lets say you install a new MOV in a device, then two years later after it's degraded big surge hits. it fails and permits the bulk of the surge to pass through. There could also be instance where multiple rapid surges happen, such as tree branch during a storm striking multiple arc a the pole.
Here's just an example.
Digi-Key - 495-3647-ND (Manufacturer - ETFV25K130E4)
When fuses blow the filament, the metal vaporizes and creates an plasma cloud that conducts electricity. Most AGC fuses have limited voltages within a few hundred volts. a Lightning strike can cause line voltages above 6KV.
A device protected with a single MOV does protect against common mode surges, and MOVs that divert surges to ground can also cause problems as the surge flow from ground into the device and out any other external connections (ie phone line, data cables, etc).
More like a device with thousands of tiny filaments. each time a spike triggers a group the tiny filaments are destroyed, leaving less and less current capacity. The problem with MOVs is that the don't blow like fuses. The can look fine until they suddenly fail. MOV's are inexpensive solution to a difficult problem. They are cost cutting devices to save money, over more traditional surge protection systems.
MOVs aren't cost saving devices in any other sense than they protect against more expensive failure than their own. It's only when you try to cut cost on the cost cutting that you develop problems.
There must be dozens of papers out there that can help you choose an appropriate rating for the MTBF you want.
The OP asked for a line conditioner design, and I just wanted him to be sure he needed a line conditioner, because if he wants very high energy surge protection most line conditioners fall back on transient absorbers of one kind or another anyway, because beyond certain high voltages and energies no amount of insulation or electromagnetic/electronic dynamic headroom is going to hold it back better. Even most battery / continuous inverter systems have MOVs in them somewhere. Different, sometimes necessary jobs.
There must be dozens of papers out there that can help you choose an appropriate rating for the MTBF you want.
The OP asked for a line conditioner design, and I just wanted him to be sure he needed a line conditioner, because if he wants very high energy surge protection most line conditioners fall back on transient absorbers of one kind or another anyway, because beyond certain high voltages and energies no amount of insulation or electromagnetic/electronic dynamic headroom is going to hold it back better. Even most battery / continuous inverter systems have MOVs in them somewhere. Different, sometimes necessary jobs.
Last edited:
idealistic tech guy
I'm not interested in flaming, but I'm not sure how much experience tech guy has with real life lightning events. I've found an exploded cap across the power switch of the PAS2, an exploded neon bulb in the power switch of the ST120, and seen plasma jump out of the front adjustments door of my television during a strike. I had a PC take a hit probably through the phone line, and despite Tech Guy's mumbling, the power supply and CPU did not go out, only the modem and PCI slot. Southern indiana is proof against earthquakes and wildfires, has few tornadoes and limited sensitivity to hurricanes, but we are a proving groud for lightning strikes. I have well used factory equipment (years of life), variable frequency motor drives, that have S14 500v MOS suppressors at the input of the 3 phase bridge. I have a dead 50V dc power supply, after 5 years life, from an ishida scale, that has MOS suppressors. All these suppressors read 9999 meg ohms, and some them I have reused. Ishida and Allen Bradley and TB Woods have major reliability reputations to protect, and all use S14 MOS suppressor on their inputs.
Direct strikes on home electronics, not bridged by the refrigerator compressor, AC compressor or blower, are rare.Blowing the fuse is my idea, instead of blowing the expensive or impossible to replace power transformer. As most transformers in the tube age were wound with 600 V rated wire, I feel justified in putting 500V MOS suppressors (salvaged from the motor drives) from case to neutral and from neutral to hot on the imputs of the dynaco equipment. These won't be reacting to nuisance surges from the A/C or refrigerator. I might point out the motor drive suppressors withstood years of on-off surges from parallel non-speed controlled motors, and lightning strikes which took out building power numerous times, without shorting across.
If you're designing a system to stand in the middle of a field with no motors around to take most of the strike energy, I supposed it would pay to be as worried about lightning as tech guy is. In a home environment, a couple of free salvaged MOS suppressor shouldn't hurt anything and may save trashing a historic piece of audio gear and replacing it with something new that might last less than a year, like all these dead PC switching power supplies I have piled up for suppressor and current limiter salvage. In no case are these PC power supplies showing damage near the fuse/input area, all the burned spots seem to be tiny components near the switching controls or lands leading to the switching transistors..
I'm not interested in flaming, but I'm not sure how much experience tech guy has with real life lightning events. I've found an exploded cap across the power switch of the PAS2, an exploded neon bulb in the power switch of the ST120, and seen plasma jump out of the front adjustments door of my television during a strike. I had a PC take a hit probably through the phone line, and despite Tech Guy's mumbling, the power supply and CPU did not go out, only the modem and PCI slot. Southern indiana is proof against earthquakes and wildfires, has few tornadoes and limited sensitivity to hurricanes, but we are a proving groud for lightning strikes. I have well used factory equipment (years of life), variable frequency motor drives, that have S14 500v MOS suppressors at the input of the 3 phase bridge. I have a dead 50V dc power supply, after 5 years life, from an ishida scale, that has MOS suppressors. All these suppressors read 9999 meg ohms, and some them I have reused. Ishida and Allen Bradley and TB Woods have major reliability reputations to protect, and all use S14 MOS suppressor on their inputs.
Direct strikes on home electronics, not bridged by the refrigerator compressor, AC compressor or blower, are rare.Blowing the fuse is my idea, instead of blowing the expensive or impossible to replace power transformer. As most transformers in the tube age were wound with 600 V rated wire, I feel justified in putting 500V MOS suppressors (salvaged from the motor drives) from case to neutral and from neutral to hot on the imputs of the dynaco equipment. These won't be reacting to nuisance surges from the A/C or refrigerator. I might point out the motor drive suppressors withstood years of on-off surges from parallel non-speed controlled motors, and lightning strikes which took out building power numerous times, without shorting across.
If you're designing a system to stand in the middle of a field with no motors around to take most of the strike energy, I supposed it would pay to be as worried about lightning as tech guy is. In a home environment, a couple of free salvaged MOS suppressor shouldn't hurt anything and may save trashing a historic piece of audio gear and replacing it with something new that might last less than a year, like all these dead PC switching power supplies I have piled up for suppressor and current limiter salvage. In no case are these PC power supplies showing damage near the fuse/input area, all the burned spots seem to be tiny components near the switching controls or lands leading to the switching transistors..
Last edited:
hi, this is my home built power conditioner i posted in this forum....
WiredState Audio Community • View topic - DIY Line Conditioner, rig your own
WiredState Audio Community • View topic - DIY Line Conditioner, rig your own
I don't like the look of this. Doesn't a common mode choke need capacitance to ground ?
This project looks like someone just threw together a random collection of chokes they had lying around. Not a well thought out design.
And the disadvantage is any noise common to Live and Neutral isn't filtered.
---
I'm throwing together an RF filter too.
All the RF EMI filters I've seen in PC PSU (switch mode) design and general RF suppression have at least one set of Y caps. Y1 and Y2 compliant capacitor's have an absolute max leakage specified by electricity authority regulations (various compliance logos are awarded per country, obviously, as regulations differ between them) such that they will not turn the device into a personal hazard, will be efficient and 'green' (more or less) and will not trip residual current safety devices.
That whole project is massively over-engineered (me LIKES massively over-engineered!!!) - dual (or triple) filter stages are for switch mode supplies that require a huge amount of decoupling in the RF band to stop the house's power lines being used as giant broadcast antenna: Something that never happens with transformer based PSUs.
It's only barely a 'power conditioner' in terms I'm familiar with. It's a line noise suppressor. For that full rich creamy conditioner feeling, a complex [AC-rectified-to-DC] -> [filter] -> [buffer] (capacitor or battery) -> [sine-wave-inverter] (with over-current and short monitoring) is ideal. And it's going to be much much harder and more expensive for a DIY fan to build than buying a suitabe VA rated model of computer uninterrupted power supply made in China. Depressing, really.
Be careful with 1uF or larger X capacitors: X caps will always alter the power factor of the power supply. Usually for the better but some measurements and sums are needed to be sure large values are not making things worse (that is, even worse).
Many electricity authorities specify approved filter designs must have a drain resistor across L & N. A good idea if there's a chance the amp PSU will be disconnected while the filter is still connected. Schematics I've seen vary between 250K and 1M. Use cement or fire/flame-proof resistors when using them in the mains AC area.
Sorry, that was all a bit lofty and proselytizing in tone. Hate that.
yes...indeed.....earthing is a joke in my country....
anyways, the design objectives were:
line isolation,
surge protection,
emi/rfi filtering, those inductors capacitor filters surely present a ver very low impedance for higher frequescies...
my contraption works.....and i use it for my pc....
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.