| Electronics > Projects, Designs, and Technical Stuff |
| Common mode choke galvanic isolation |
| (1/1) |
| Jr.Maxwell:
Hello, I'm trying to get to understand grounding in circuits better, so I'm building a small circuit on a PCB and wanted to galvanically isolate the input power source - a 24V battery which is also earthed- from the rest of the circuit. (See Attachment: Gal_1). I googled and researched what my options where. 1 i could use a optocoupler 2 i could use a dc/dc power converter 3 i could use a tranformer. My question consists of 2 parts: 1- Are those 3 options the only option one can Galvanically isolate the input power from the rest of the circuit? I'm trying to get a feeling for this stuff so my brains are telling me i'd use an optocoupler for signals and low power stuff. - Descrete DC/DC converters and transformers for the power galvanic isolation, but when do I choose the which one over the other? 2- My main question: What if i use a common mode choke? would that work as a galvanic isolator too in this situation? technically it's a coil transformer.? So I'm thinking I could use a CMC instead of a regualr coil? That would give me the pros of CMC and also galvanically isolate the planes. or am I in the wrong here? Thank you all for your help, I'm trying to get a better understanding of electronics :) Thank you for your time |
| T3sl4co1l:
The power supply? The circuit? What are these, why give this hint of information, but stop short of actually describing anything..? Like, are they connected somehow -- but then why isolate? Are you designing a power supply (that needs galvanic isolation from mains)? We have no idea. Anyway: CMCs are transformers, yes. Typically with quite high leakage inductance, so that Zo is relatively high, and Fc relatively low. Zo typically 300 ohms or thereabouts, which makes getting much power through them a challenge (typically a watt or two is about it). Fc limits bandwidth for signal purposes, maybe a few MHz. Tim |
| Jr.Maxwell:
Hello, I see That I didn't do a good job presenting a detailed question, I've upated it though, maybe it's now more clear! |
| Ian.M:
1. Opto-coupled power transfer has sh!tty efficiency. 10% end to end is doing very well indeed. Its used for applications like proving a few watts of power over optical fibre when you need hundreds of KV isolation, typically with a laser diode into one end and the other end illuminating a large area photodiode. You could use a LED array closely coupled to a large area solar panel, with the LED colour selected to best match the panel spectral response, but its only worth doing if you need extremely clean DC out, with absolutely no ripple, don't give a rat's @ss about efficiency, and absolutely cannot swap out rechargeable battery packs. 3. You cant use a transformer on DC without a chopper or power oscillator to drive it with AC and a rectifier on the secondary side to convert back to DC, as a transformer has limited bandwidth and doesn't couple DC. Implementing that gives you a DIY version of (2) an isolated DC-DC converter. 4. Its a bit 'agricultural' or 'prehistoric', but if you need a lot of power transferred with a high (but not extreme) isolation voltage, more isolation voltage than can easily be handled by a transformer, consider a DC motor-generator set with an insulated shaft or belt drive between them Its also one of the better ways of getting lots of power into a Faraday cage, if you split the insulated shaft and insert a conductive coupler well grounded by ring brushes to the cage. |
| T3sl4co1l:
Ah okay, power isolation then. One way, or bidirectional (i.e. charge the battery too)? A DC-DC converter module is just a transformer internally (with some odd exceptions), so that really reduces your question to light or transformer. Even more generally, we need to use some kind of electromagnetism, because if we stick with just one (electric or magnetic), we have no power flow, and if we stick with none, we have a direct connection (DC and galvanic path). Or, well, nothing at all (no connection). :P Or if we add transducers to other media, we can do whatever. For example, generators coupled via insulating shaft, or acoustic waves ("piezo transformer"). But within E&M, we can do low frequencies in transformers; high frequencies in coupled inductors; optical frequencies in photovoltaics; um, er, heat engines I guess in THz-IR bands, since we don't really have rectennas suitable for those frequencies yet; or.. nuclear something or other for x-rays+. That covers all of E&M. But yeah, clearly the most efficient and affordable among those options will be the first three... So, a better question might be, what drives the choice of frequency and transformer type? Cost, simplicity, size, efficiency, isolation capacitance and voltage, noise, etc.? Tim |
| Navigation |
| Message Index |