EEVblog Electronics Community Forum
Electronics => Repair => Topic started by: Traceless on January 30, 2022, 10:58:57 pm
-
Hey everyone,
I just pulled the part shown on the image below out of 24V, 0.5A power supply. By the looks of it I'd say it is probably a 20 Ohm wire-wound resistor either 2W or 3W that served as fuse. It was the first component after mains input. I found these parts (https://de.aliexpress.com/item/1005001508405319.html?spm=a2g0o.productlist.0.0.392c53b7uqgCJO&algo_pvid=657441bf-d09f-4a43-bf51-3435b8d66035&algo_exp_id=657441bf-d09f-4a43-bf51-3435b8d66035-25&pdp_ext_f=%7B%22sku_id%22%3A%2212000016401186337%22%7D&pdp_pi=-1%3B1.09%3B-1%3B182%40salePrice%3BEUR%3Bsearch-mainSearch) that look quite similar, however the specs rate the 2 and 3W resistors for a max. voltage of 200V. The smallest one rated for 300V is the 7W version which according to the sellers chart (https://ae01.alicdn.com/kf/Hbf62badd4aea4e4793185cb40d57937cL.png)is a lot bigger that the component I have here. Am I wrong, and my broken part is something different?
-
https://www.aliexpress.com/af/fuse-resistor.html?d=y&origin=n&SearchText=fuse+resistor&catId=0&initiative_id=AS_20220130224826 (https://www.aliexpress.com/af/fuse-resistor.html?d=y&origin=n&SearchText=fuse+resistor&catId=0&initiative_id=AS_20220130224826)
-
@eurgenca: I had already looked at Aliexpress (see link in my original post). The big problem is that most of these resistors seem to be not rated for the purpose or vendors list no specs at all.
I settled for the 2W version of these now (https://de.aliexpress.com/item/1005002531753445.html?spm=a2g0o.productlist.0.0.3bc9147bst1nzF&algo_pvid=aba819f9-4cf4-4d34-a857-db2e0d40dce6&algo_exp_id=aba819f9-4cf4-4d34-a857-db2e0d40dce6-0&pdp_ext_f=%7B%22sku_id%22%3A%2212000021024198304%22%7D&pdp_pi=-1%3B0.85%3B-1%3B-1%40salePrice%3BEUR%3Bsearch-mainSearch), which are physically larger than the part I am going to replace. The 1W version would have had the right dimensions but is only rated for an "upper working voltage" of 150V. Also I don't get how the "upper limit load voltage" (400V) can be rated 50V higher than the the "upper insulation voltage" (350 V), this seems sketchy.
-
No wonder they have been hiding it in shrink tubing
-
I would not bother finding a replacement.
Just use a regular resistor followed by a proper 250V fuse. This is a protection for overload only so you do not need a HRC fuse.
If this is 12VA out then it most likely does not have any PFC and draws high rms current. Much higher than 60mA. I'd start from a T/100mA/L/250V fuse and 33R 1W. This decreases rms current. Just make sure that fuse blows before resistor gets overheated.
-
Increasing the resistor power rating is a bad idea, as it is equivalent to putting a higher current fuse in a circuit that was designed for a lower current one.
Working voltage ratings don't really apply to low value resistors in a particular product range, as they are dissipation limited. e.g. for your example of 1W resistors rated at 150V, as P=V2/R, any resistor of under 22.5 kiloohms will exceed its power limit before it can reach its voltage limit. However you do need to consider pulsed power and peak voltage, so assuming a brief 5x power overload is permissible, that reduces the 'corner' resistance between power and voltage limiting to 4.5 K, and if you are dealing with sinusoidal AC, the working voltage limit should be reduced by a factor of sqrt(2), which halves the 'corner' resistance again, to 2.25 K.
The load voltage limit indicates how much peak voltage the fusible resistor can stand-off once blown. 400V is suitable for 220V-240V AC applications. The insulation voltage limit of 250V indicates that you *MUST* *NOT* allow anything conductive to contact the resistor body, as the peak voltage (relative to ground) at either end of the resistor is likely to exceed that. As the insulation voltage is less than the peak supply voltage the whole resistor body must be treated as a live part from a safety viewpoint, e.g. when doing creepage and clearance distance checks.
It is unclear whether the 'dielectric with standing' (sic) voltage of your original link is the outer insulation breakdown voltage or the core insulation breakdown voltage. It seems likely its the former, but as its unclear, I'd treat that range as 'live body'.
Its obvious that your low value 'fuse' resistor will fall into the power limited region, so as they don't have a 20 ohm resistor in their 1W 150V WV fusible resistor series (https://www.aliexpress.com/item/1005002434216098.html), I'd use the 22 ohm one, which will fuse at approx. 5% lower current.
You could also use 2x 10 ohm 1/2W fusible resistors in series, *IF* you can mount them safely.
I would not bother finding a replacement.
Just use a regular resistor followed by a proper 250V fuse. This is a protection for overload only so you do not need a HRC fuse.
If this is 12VA out then it most likely does not have any PFC and draws high rms current. Much higher than 60mA. I'd start from a T/100mA/L/250V fuse and 47R 1W. This decreases rms current. Just make sure that fuse blows before resistor gets overheated.
Alti is 'blowing smoke'. Changing passive protection components at the power input cant do anything to reduce RMS current, and the original resistor could theoretically withstand 220mA indefinitely (assuming it was actually 1W rated, operated at an ambient temperature that doesn't require derating) so if it blew due to long term use at its limits, his suggestion will fail far sooner. Also you have the problem of safely mounting the fuse . . .
My repair trade experience of fuses in SMPSU input circuits is that if the chopper transistor or bridge rectifier fails, non-HRC fuses frequently rupture and the resulting damage can be difficult to clean up. You'll usually see a lot less mess and circuit damage from an fusible resistor covered with fiberglass sleeving
-
For a 20R 1W fusible resistor to blow it would have to draw over 220mA, that is over 50VA at 230V continuous. And that does not add up for a 12W supply. Maybe this 20R fusible resistor was designed to survive just 1h under 1W conditions (and lets say 1000h at 0.5W).
No wonder they have been hiding it in shrink tubing
Perhaps this heatshrink tubing was added to alter tripping characteristics, this works like an insulation. AFAIK proper fusible resistors do not require any sleeving as these contain a proper flameproofing in their design. I remember I did see a PSU with low rupture fuse some time ago, packed in shrink tube but I thought this is just for clearance reasons.
-
I agree its unlikely to have been pushing the original resistor's continuous power rating. Even if we assume 80% efficiency, that's only 15W input power. The current crest factor for a bridge rectifier directly feeding a large reservoir capacitor will push up the RMS current and thus the VA, but I don't see any way it could go over 25VA. I suspect the original design was for a factor of two margin on the 20 ohm resistor, which they assumed gave them enough margin for ambient temperature + covering it in heatshrink, and was probably adequate as designed for the continuous dissipation.
*IF* there is no other fault present, I would suspect that the inrush current repeatedly slightly exceeded the original resistor's pulse power rating, slowly degrading it till it blew. Its an easy trap to fall into, and if purchasing screws you over by substituting an ordinary fusible resistor when you had specified a pulse-rated one, can happen to any of us.
However the O.P. hasn't confirmed yet that this was the only fault . . .
-
Thanks Alti and Ian.M for the very detailed feedback. The blown fuse was indeed only a secondary fault. The PSU belongs to the docking station of a vacuum robot. And worked pretty well for 1,5-2 years. One day the robot would dock and start charging (indicated by a yellow status led and a beep), but powercycle and restart the charging process every few seconds. I first suspected a dead battery, however with a brand new replacement battery the issue persisted. Apparently the power supply was no longer able to sustain the load required to charge the robots battery. I replaced the PSU with a new 24V 1A supply which resolved the issue.
Still I am curious what went wrong and decided to take a look at the old PSU. Once I opened it I measured the capacitors and checked some diodes all were fine. Then I connected it to mains and checked voltages which were fine as well (with no load). Next I hooked it up to my electronic load set to 24V, 0.5A to see if I can trigger the fault that way. Voltage immediately dropped to 0, and the primary side was dead because the resistor died silently (no poof or smoke). Right now I'm mainly interested to get the PSU running again just to improve my PSU troubleshooting skills and educate myself a bit. If I manage to fix it along the way it will make a nice addition to my bench power supply drawer.
-
https://www.ebay.com/itm/224812332678?hash=item3457dbf686:g:Um8AAOSwba1h9HiG (https://www.ebay.com/itm/224812332678?hash=item3457dbf686:g:Um8AAOSwba1h9HiG)