Hi, I choose this medical grade SMPS to try out Mark Johnsons AmyAlice filter. Results were OK but not comparable with my linear PSUs. The added necessary filter makes the SMPS quite usable (when deciding for SMPS) but it makes costs of course higher. Still a nice experiment just to keep in touch with recent cheap electronics.
However it keeps blowing mains fuses. It can not be taken into service as it is unreliable. I even upped the mains fuse to an unrealistic 500 mA and it just blows without any apparent reason. So today I got fed up and disassembled the PSU. While disassembling I heard a rattling noise. This appeared to be IN the plastic casing of the resin potted SMPS. So it is not fully potted either. When I measure it at the primary side and shake the module gently I can see a short circuit happening on my DMM.
Since this is a medical grade PSU I think this is worrying. Now I am not fond of low power SMPS anyway and possibly I maybe attract problems just like people afraid of dogs always seem to get attention of dogs but my question is if anyone has similar experiences with these modules.
However it keeps blowing mains fuses. It can not be taken into service as it is unreliable. I even upped the mains fuse to an unrealistic 500 mA and it just blows without any apparent reason. So today I got fed up and disassembled the PSU. While disassembling I heard a rattling noise. This appeared to be IN the plastic casing of the resin potted SMPS. So it is not fully potted either. When I measure it at the primary side and shake the module gently I can see a short circuit happening on my DMM.
Since this is a medical grade PSU I think this is worrying. Now I am not fond of low power SMPS anyway and possibly I maybe attract problems just like people afraid of dogs always seem to get attention of dogs but my question is if anyone has similar experiences with these modules.
Last edited:
If it is a few years old, replace it. If it is failry new, get it replaced under warranty.
From your description, it sounds like there may be a stray screw floating about in there, possibly fallen in when fitting the psu to a project. Have a look as if it is, that is not covered under any warranty.
Very unlikely to have come like that from the factory.
From your description, it sounds like there may be a stray screw floating about in there, possibly fallen in when fitting the psu to a project. Have a look as if it is, that is not covered under any warranty.
Very unlikely to have come like that from the factory.
It is new from Digikey but ordered months ago. It is a sealed model but apparently not fully potted with resin although the datasheet mentions that.
https://www.meanwell.com/webapp/product/search.aspx?prod=MPM-20
Yes it came like this from the factory. It can AFAIK not be manipulated as it is potted.
https://www.meanwell.com/webapp/product/search.aspx?prod=MPM-20
Yes it came like this from the factory. It can AFAIK not be manipulated as it is potted.
Last edited:
Inrush current for SMPS is typically large (tens of amps), try 3A slow blow, or larger, 500mA is comically small for this application. Remember the fuse protects the wiring from catching fire, its not to protect the supply, just make sure the mains wiring can handle the current at which the fuse blows.
The input of a SMPS is a bridge rectifier straight into electrolytic capacitors (ironically larger power supplies are more likely to have a slow-start circuit and have lower inrush).
The input of a SMPS is a bridge rectifier straight into electrolytic capacitors (ironically larger power supplies are more likely to have a slow-start circuit and have lower inrush).
The internal fuse is 2A T. That is the handbrake if things really go wrong like the trademark of switchers seems to be. It would not make sense to have an external 3A T fuse.
I also don't think 500 mA is comically small for a 20W PSU. The loose part in it making a short probably is the cause. For some reason the IRM-20-5 could be bought for just 3,99 Euro today including taxes (the MPM version was 14 Euro without taxes) so I ordered a few of those for testing.
I also don't think 500 mA is comically small for a 20W PSU. The loose part in it making a short probably is the cause. For some reason the IRM-20-5 could be bought for just 3,99 Euro today including taxes (the MPM version was 14 Euro without taxes) so I ordered a few of those for testing.
Last edited:
Well its inrush rating is 20A for 115V and 45A for 230V (just checked the datasheet), so yes 0.5A fuse is just going to explode (maybe even if it is slow-blow).
It just does NOT do so at power on and it also does not explode. It blows when it has a very light load just out of nothing probably because of the sometimes occurring short circuit in it. For world peace let's not concentrate on the fuse (I promise to try 1A T too), the weather or class D being superior to class A but on the defect in the darn switcher.
Last edited:
I repeat 500mA is completely wrong for this sort of power supply. Perhaps the fuse is being damaged at switch on? Is it an HRC (sand-filled)?
If I choose higher than 1A T the internal fuse likely will blow rendering the already suspect SMPS useless. I can short circuit the fuse and I am pretty sure the internal 2A T fuse or the 16A RCBO breaker will come into action as the SMPS has an intermittent short circuit in it.
The short circuit in a new medical grade SMPS is the subject.
The short circuit in a new medical grade SMPS is the subject.
Last edited:
But its a point of uncertainty that is easy to remove. Anyway the point about medical grade is the strength of the isolation. From the description if its damage it might be from mechanical damage (a drop or poor packaging in transit). And the other question I'd ask is could it be counterfeit? Meanwell would be an obvious target for the counterfeiters.
Counterfeit it could be, I thought the same as it apparently is not fully potted. But... it was ordered at Digikey.
I've used the MPM and IRM series for some time without issues. I found the MPM to be less noisy than the cheaper IRM, but nothing that couldn't be filtered out. Perhaps just a quality escape on this particular unit.
Meanwell usually do a 36 month warranty, so I'd definitely contact them or Digikey and insist on a refund / replacement unit.
Meanwell usually do a 36 month warranty, so I'd definitely contact them or Digikey and insist on a refund / replacement unit.
Already did. I think it is a good habit to start with light fusing when assembling any DIY device. Also one better forgets a too light fuse than a 6.3A one. It is a mental fuse too in a way 🙂
I repeat fuses protect the wiring from burning, not the device electronics (semiconductors typically fry much faster than fuses).
For commisioning a new amplifier I use an adjustable current-limit power supply, as that prevents over current, rather than reacting to it. This is much more capable than using a range of fuses.
There are e-fuse chips for over-current protection of electronics, but those are basically doing the same job as a current limited power supply, and they can fail in a dangerous state, so if wiring/cabling is involved you need a fuse too.
The fuse should match the wiring, basically. So for a mains fuse internal to a device the size of wire for internal mains wires is what determines the safe values for the fuse. This extends to transformer primaries as these are effectively internal mains wiring. The 2A inside your PSU is likely as example of this.
Most electronic devices fail without starting a fire and don't usually require fuse protection as they act as their own fuses - there are exceptions of course.
An aside:
Back in the day in the UK you could get 1A, 3A, 5A and 13A mains fuses, 1A was for very small devices (wall clocks), 3A for various low drain devices, 5A for (tungsten) lighting, 13A for power. These days you rarely see 1A or 5A as most devices are either low current or heaters/kettles, so 3A and 13A are commonly available, and all mains flex sold is rated at least 3A in the UK, so the 1A rating isn't needed at all.
In the UK mains plugs by law all have integral fuse-holders so the fuse is explicitly matched to the cable by the manufacturer.
For commisioning a new amplifier I use an adjustable current-limit power supply, as that prevents over current, rather than reacting to it. This is much more capable than using a range of fuses.
There are e-fuse chips for over-current protection of electronics, but those are basically doing the same job as a current limited power supply, and they can fail in a dangerous state, so if wiring/cabling is involved you need a fuse too.
The fuse should match the wiring, basically. So for a mains fuse internal to a device the size of wire for internal mains wires is what determines the safe values for the fuse. This extends to transformer primaries as these are effectively internal mains wiring. The 2A inside your PSU is likely as example of this.
Most electronic devices fail without starting a fire and don't usually require fuse protection as they act as their own fuses - there are exceptions of course.
An aside:
Back in the day in the UK you could get 1A, 3A, 5A and 13A mains fuses, 1A was for very small devices (wall clocks), 3A for various low drain devices, 5A for (tungsten) lighting, 13A for power. These days you rarely see 1A or 5A as most devices are either low current or heaters/kettles, so 3A and 13A are commonly available, and all mains flex sold is rated at least 3A in the UK, so the 1A rating isn't needed at all.
In the UK mains plugs by law all have integral fuse-holders so the fuse is explicitly matched to the cable by the manufacturer.
No need to repeat, I know a thing or 2 about fuses 🙂 Amateurs think the fuse protects the device but that is illogical as the device already has failed when the fuse melts. In other words there is no device to protect, only the wiring. One of the standard misunderstandings really.
In many devices in the EU 100 mA, 160 mA, 200 mA, 250 mA, 500 mA, 630 mA, 1A, 1.25A, 1.6A, 2A fuses in either 5 x 20 mm or TR5 format can be found. It would be ridiculous to use 3A T fuses in sensitive devices meaning certain death or total destruction when a fault occurs while the cabling does not blink an eye even at 10A. Fuses may seem simple but they have the task to melt both at overcurrent over longer time and directly at sudden severe overcurrent/short circuits. Choosing them with care is good, better safe than sorry.
In many devices in the EU 100 mA, 160 mA, 200 mA, 250 mA, 500 mA, 630 mA, 1A, 1.25A, 1.6A, 2A fuses in either 5 x 20 mm or TR5 format can be found. It would be ridiculous to use 3A T fuses in sensitive devices meaning certain death or total destruction when a fault occurs while the cabling does not blink an eye even at 10A. Fuses may seem simple but they have the task to melt both at overcurrent over longer time and directly at sudden severe overcurrent/short circuits. Choosing them with care is good, better safe than sorry.
This contradicts heavily with protecting wiring. And EU regulations.
Last edited:
I said electronic devices, ie capacitors, transistors, etc - didn't mean a electrical appliance!
As I said there are exceptions, and it does depend on the power available. Many electronic parts can blow holes in themselves rather than get red hot. Many resistors are unable to overheat given the supply voltage of the circuit, they self-limit - of course others may need to be specified as fusible. PCB traces are an issue, but again they can be used as fuses on suitable substrates. The point about wiring is the flammable insulation which either ignites or melts away exposing a red hot copper wire to some other flammable materials around - many electronic components are explicitly made from fire-resistant materials which char without igniting. A transistor encapsulated in PVC rather than glass-filled epoxy might well be a fire-risk!
As I said there are exceptions, and it does depend on the power available. Many electronic parts can blow holes in themselves rather than get red hot. Many resistors are unable to overheat given the supply voltage of the circuit, they self-limit - of course others may need to be specified as fusible. PCB traces are an issue, but again they can be used as fuses on suitable substrates. The point about wiring is the flammable insulation which either ignites or melts away exposing a red hot copper wire to some other flammable materials around - many electronic components are explicitly made from fire-resistant materials which char without igniting. A transistor encapsulated in PVC rather than glass-filled epoxy might well be a fire-risk!