Re: LF: Loop preamp with the BF862 / optical link

[SAM JEWELL <jewell@btinternet.com>]

Folks,
I think it is worth pointing out that most cable TV systems use fibre links, at least as far as the cabinet. The light (laser) is amplitude modulated with the RF carrier(s) at the cable head end or regional Centre. These carriers are usually locally generated but can also be from directly off-air if the signal to noise ratio is adequate. The modulation index can be held quite high for a single carrier, progressively reducing for multiple carriers in order to minimize intermodulation between carriers. Ultimately, with many carriers, the modulation spectrum will be noise-like. Those of you who have  worked with the old telecom carrier based systems will recognise the same characteristics.

The earliest cable (HFC) systems carried AM VSB signals in the VHF and UHF frequency range in their native transmission format with minimal distorion and added noise. Digital modulation (using QAM modulation, but still amplitude modulation of the light) came later.

From the laser (or LED) input to photodiode/PIN/APD output there will be a certain link loss and system noise figure. If the system parameters are  known, this can be modelled using e.g. AppCAD, Noisecalc. In some systems signal to noise ratio is critical. For LF systems, where there is a fairly high noise level, it may be far less critical (famous last words).
Not all fibre systems use digital modulation of the light, although in most cases it is still amplitude modulation.

I should also point out that fibre is often used to carry the RF and IF signals from the SHF rooms (at the back of the big dishes) at satellite earth stations across the site to the demodulation and signal processing buildings, in preference to using waveguide, as was done many years ago. Whilst working at BT Research Labs my team produced one of the first radio over fibre test systems for BT's Goonhilly and Madley Earth stations. These early fibre links used multimode 50/125u fibre and multimode lasers.

As I mentioned in an earlier posting, microwave and LF are just the adjacent ends of the spectrum. Not opposite ends!


73 de Sam, G4DDK

---

From: James Moritz <james.moritz@btopenworld.com>
To: rsgb_lf_group@blacksheep.org
Sent: Tuesday, 2 February, 2010 11:57:13
Subject: LF: Loop preamp with the BF862 / optical link

Dear Ken, Stefan, LF Group,

If I remember correctly, the components Stefan is discussing use a polymer fibre and are intended for short-distance optical links of a few metres, rather than telecomms-type applications with miles of fibre, so fibre loss is not really an issue. The ones I encountered some years ago used visible red LEDs, although I think there were infra-red ones too. The plastic fibre could be simply cut with a sharp knife, and clamped into the emitter/detector components with reasonable results, rather than requiring precision optical connectors. There is an overview at http://www.avagotech.com/docs/AV00-0143EN

Of course, since the optical link is effectively a signal-frequency gain component in the receiver front end, its noise figure will contribute towards the receiver noise figure. Since the overall efficiency of the electrical/optical/electrical conversion is rather low, with the loss of the fibre in addition, there will be much less signal power coming out of the detector than going in to the emitter. Also the detector and emitter will contribute their own noise. So the noise figure of the optical link by itself is probably rather high. But the FET voltage-to-current input stage driving the emitter will have a high power gain due to the very high input impedance, which would greatly reduce the impact of noise in the optical link - clearly the overall noise level in Stefan's system is reasonably low.

Most optical links are used with digital signals, so this analogue optical link is quite unusual - as well as noise, distortion products will also be produced by the optical link. But DK7FC's antenna obviously works, so the principle is viable.

Cheers, Jim Moritz
73 de M0BMU


----- Original Message ----- From: "Ken" <ken.h.wright@btinternet.com>
To: <rsgb_lf_group@blacksheep.org>
Sent: Monday, February 01, 2010 5:56 PM
Subject: Re: LF: 2nd try of sending the pic of my loop preamp with the BF862


> Hi Stefan.
> 150dB/km, is a very high loss for modern fibre, even thhe early 850nm multimode fibre was only 1 or 2dB /km...