Can I use a 4N25 optocoupler as RF insulator?

Is it a good idea to use an optocoupler such as 4N25 as an RF isolator? Is there a better one than this, something that consumes very little power and uses USB serial to draw its power from? I am trying to isolate RF as well as any type of noise such as a switching power supply noise or computer noise between a PC and a short wave transceiver.

I sort of understand that you want an isolated USB link between PC and shortwave radio. Maybe because of groundloops? Laptops can be really bad there. AFAIK optos are too slow for USB. There are special USB(1) isolators available. Some even with isolated power transfer. Google "USB isolator"/images. Leo..

Yes, thank you, I am using a laptop and they are as bad as you say. That would be one of the things I want to isolate, USB connections, didn't know there were such things and I see there are also isolated USB hubs but quite pricey, some of them as much as a brand new laptop!!! I suppose they isolate RF as well as computer generated noise.

The other is an RS232 TTL device with its own very noisy switching power supply. I will only be using RS232 signaling CTS/RTS or DSR/DTR pairs so Tx/Rx speed is not important at all, only low power consumption and RF/noise isolation are the critical things here. I was thinking of using 4N25 or something equivalent or better than that.

You really should explain exactly what sort of connection between the PC and the transceiver you require?

Generally, any interface supplied by the transceiver manufacturer will already have implemented whatever isolation is practical such as opto-isolation and RFCs on the cables.

Realistically, expecting to out-engineer the transceiver manufacturer is a little optimistic for a beginner. :astonished:

I am not sure what kind of info you are asking me of.

It is a MAX3232 to TTL converter, exactly like this one http://www.ebay.com/itm/290712442058?_trksid=p2060353.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT

I use the converted RS232 to TTL CTS/RTS signals to control an antenna device but the problem is that the antenna is very sensitive and picks up some noise from the switching power supply of the MAX3232. I replaced the switching power with a simple transformer and it got much better but still the antenna picks up noise. When I disconnect the CTS/RTS lines, the noise is gone. Noise is also gone when I use batteries. This is one part of the problem during receive.

Second part is transmission mode. MAX3232 picks up RF from the antenna this time so good isolation is badly needed both ways.

OK, let's go back to basics: you have a device at each end of a long cable. The cable is picking up unwanted interference and putting it into both devices.

Imagine you add an opto-isolator at one end. How does that stop interference getting into the device at the other end? How does it stop bad data already on the cable from going through the opto because the opto doesn't know that data is bad?

The best solution is to use a technology which doesn't pick up interference. Put a RS485 converter on each end. Probably not what you want or need.

Second, apply some filtering at each end of the cable, to reduce the amplitude of high-frequency pulses without excessively attenuating the frequency of the signals you want. A resistor and a capacitor on each input will probably work. Inductors are also useful.

Third, wrap some ferrite cores around the cable. Cut them off old equipment or buy them. They are only a few cents to a few dollars depending on size.

Fourth, easiest and cheapest, route the wire away from the source of interference. Put it in a metal conduit if you have to.

What frequencies are we talking about. 0-30Mhz, 144Mhz, higher. They all have different ways of solving the problem. Basic fault-finding. Try everything. e.g. is the laptop silent when it runs on battery power (supply unplugged!). Don't start fixing things untill you know what's wrong. As MorganS is sugesting, don't let digital hash enter a wire. Almost every device is a transmitter. No problem untill you connect an aerial (unshielded wire) to it. Use USB leads with ferrite cores on both ends. Does your antenna tuner device live in a metal box. Is that box grounded to your radio with shielded wire. etc. We might be able to help you here, but I think it's better to ask these questions on a Ham radio forum. Leo..

This isolated USB device allows you to provide isolated power as well and for this sort of thing is not too expensive.

Prospekt-ISOUSB-Cable-B-en.pdf (224 KB)

Grumpy_Mike: This isolated USB device allows you to provide isolated power as well and for this sort of thing is not too expensive.

Except!

Coupling capacity About 1.5nF

Dubious benefit for RF isolation.

Wawa:
What frequencies are we talking about.

Antenna picks up noise at low frequencies, usually lower than 7MHz during Rx. However, RF gets through during Tx even at 10MHz. I am already using quite a few ferrite cores in both ends of the cable. They helped a bit with RF but did not solve the noise problem nor the RF completly. The device has to be mounted very close to the antenna as it is an antenna tuner so moving it away is not an option. Using a symmetrical antenna would help a lot more than anything with the RF but it is not an option either. A shielded UTP type of cable to the tuner only made things worst!!! The box is not metal but this is probably the last thing I want to try. Increased band noise is also there when I operate anything (laptop and tranceiver) on batteries. Only way to get rid of it completely is to disconnect CTS/RTS. When switching power supplies to on there is more noise and spikes every 30-40kHz are added.

I was kind of hopping an optocoupler would isolate anything not wanted both ways. The circuit in use is only an on/off type of manual switch so it is working at DC levels, something like 2-3 times/sec max in extreme conditions. It simply starts the tuning process of the antenna when I send a manual CTS high and waits for the RTS signal when the tuning is over, that simple. Is 4N25 the correct choice though for this job? It really doesn’t have to be fast at all, probably in this case the slower, the better.

Thank you for your suggestions and ideas. I think the next think I will try is RC type of filtering for the incoming RF combined with an optocoupler as this helps to protect my equipment in cases of high voltage leaks but I am not sure which one to use.

So to sum it up, there are three noise sources:

1) PC generated 2) Power supply of MAX3232 generated 3) RF from the antenna

Not a very easy problem to solve with all these noise sources around.

I read there are also digital isolators that can replace optoisolators but their functionality is RF based. Could they work any better? Any hints or ideas at all of what isolator should I try for all this?

A PC is about the worst RF noise source there is out there, breaking a serial link with slow optoisolator will work pretty well I would think - don't go for the more expensive fast logic isolators. USB - no, that's a high clock speed and fast edges, it will always carry lots of digital noise.

Basically logic circuitry is all noise to a sensitive RF system, you need to bandwidth limit its connections coming out, in particular no fast logic edges should be escaping - think > 1us per logic transition and there will be a lot less RF energy (typical CMOS signals switch in a few ns, a thousand times faster).

If you've looked inside a decent transceiver that has a digital logic section you'll see its in its own shielded box with the signals coming out usually via feed-through caps or other filtering circuits.

Yes, I agree, I think this is a case of the slower the better. Something like the HCPL-3760 with max speed of 4 kbits/sec or similar to that with some kind of RLC type of filtering for the RF.