Doubts regarding the selection of fuses for 4-20mA signals
Thread Content
Dear colleagues in the oil and gas industry, I’m sure that those of you working with process control often come across situations where 0-20mA/4-20mA signals need to be connected to controllers. Those who are focused on cost savings simply connect the signals directly to the controller, while those who prioritize safety install a 5x20mm fuse in series with the signal. Of course, I have used both approaches; my practical experience tells me that it’s possible to use the system without a fuse as well, and by \"normal operation\" I mean no problems for several years. Fuses with fusing elements present do come with quite a few issues, especially during the debugging phase. There are those with 1A (1000mA) fuses, those with 0.5A (500mA) fuses, and those with 0.2A (200mA) fuses. I once asked my supplier to use fuses of 0.1A (100mA), and that worked as well; there were no frequent cases of the fuses blowing. This is because our signal current is at most 20mA, so a 100mA fuse is more than sufficient. As for 1A or 0.5A fuses, they are even too large, and in situations where a fuse actually needs to blow, they hardly serve any purpose. Therefore, when selecting fuses for current signals, I always choose ones with a rating of 0.1A. I wonder if anyone has any better approaches? ?
Most systems don’t require AI circuit fuses, right? Maybe I’ve worked on few projects; in any case, I’ve never seen an AI circuit with a fuse.
This post was last edited by jlshnlhj on 2017-7-25 at 10:44. The site should be a intrinsically safe area. “Those equipped with fuses do have quite a few problems,..., and there hasn’t been a frequent occurrence of the fuses blowing. What are these problems? You say, “...there have also been no frequent cases of the fuse blowing.”"
This post was last edited by xzzgjxmf on 2017-7-25 at 12:29. There is no need to install fuses in 4-20mA circuits; it seems like an unnecessary addition.
4-20MA signals mostly originate from two-wire transmitters. We can consider the transmitter as a constant current source, and such a source does not fear short circuits at its terminals; what it fears is an open circuit. Therefore, accessing insurance does more harm than good.
For 0-20mA/4-20mA signals at the site, signal isolators are generally used; intrinsically safe instruments also require safety barriers along with surge and lightning protection measures. Therefore, installing fuses in this case is unnecessary, and it only adds more failure points to the system, which is not worth it!
I’m very grateful for everyone’s participation and their individual opinions! I also checked the old oil field system diagrams; they used fuse-equipped terminals for the AI, AO, and DI signals, while only the DO output used a relay, so no fuse was added there. There are also measures such as AI, AO, surge protection, and isolation; this is still a CNPC project, as shown in the figure below
I really have never seen AI or AO systems equipped with insurance; usually, either no such protection is provided or safety guards and surge protection are used instead. I’ve been in contact with Yokogawa, Control Union, and DelataV; none of them provide insurance.
Fusing is added to the circuit mainly to prevent the equipment from being damaged when too much current flows through it. In the case of AI and AO circuits, large currents generally do not occur in these circuits; even if a short circuit occurs, no large current will be generated. Considering the impact of thunderstorms, a lightning protection grid can be added. I think DI and DO should be equipped with fuses; when the positive signal voltage is at a low level, a large current is generated, and without fuses, this can damage the channels or related components. If some system terminal boards already come equipped with fuses, there is no need to add additional ones.