|
Homebrew Equipment A place to show, design and discuss the weird and wonderful electronic creations from the hands of individual members. |
|
Thread Tools |
6th Sep 2010, 6:02 pm | #1 |
Tetrode
Join Date: Jan 2007
Location: Lincoln, Lincs. UK.
Posts: 94
|
Hybrid receiver
Sat here recovering after minor surgery with not much else to do but think and read, I was musing on the following. If one was to design a very good HF receiver, with some sections using valves and some solid-state, which would be the best areas for each. In other words, what would be a good receiver which uses the strengths and tries to avoid the weaknesses of each of these technologies?
|
6th Sep 2010, 8:40 pm | #2 |
Nonode
Join Date: Jan 2004
Location: 1966-1976 Coverack in Cornwall and Helston Cornwall. 1976-present Bristol/Bath area.
Posts: 2,965
|
Re: Hybrid receiver
For the front end a valve or FET device. For the IF's as you need dual IF, to eliminate image interference, and the ability to vary the bandwidth I think semiconductors here but for the sound output a nice either single ended or push pull valved output.
__________________
Simon BVWS member |
6th Sep 2010, 8:50 pm | #3 |
Rest in Peace
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
|
Re: Hybrid receiver
I think the honest answer has to be that with modern techniques solid-state is the best technology for every stage of an HF receiver. On the other hand, if you want to modify your criterion to a very good but fairly simple HF receiver then valves can be used in many stages. Exceptions might be local oscillator and frequency display/synthesizer.
Valves give higher stage gain than transistors, although this can be a mixed blessing as it can lead to poor gain distribution. However, by transistor standards, even a low-signal valve is a power device (and fairly linear) so can handle bigger signals without having to use advanced techniques like local feedback. Use valves for RF and IF gain, Schottky diodes for switching mixers, and silicon chips for a synthesizer? Last edited by G8HQP Dave; 6th Sep 2010 at 8:52 pm. Reason: spelling |
6th Sep 2010, 8:56 pm | #4 |
Retired Dormant Member
Join Date: Nov 2009
Location: Carmarthen, Wales, UK.
Posts: 79
|
Re: Hybrid receiver
Good one this, I'm in the middle of putting together a homebrew receiver, I would try to stick to one or the other. But if I had to mix technologies I'd stick to solid state for the power supply, and valves for signal processing & audio.
At the moment just because I like to make life awkward for myself I'm trying to do the receiver all hollow state, no semiconductors anywhere, although I might make a concession to using copper oxide rectifiers. I'm also trying to implement it with as few different valves as possible, preferably all double triodes and diodes. I rather think that mixing technologies it would be easy to get the worst of both with none of the benefits. |
6th Sep 2010, 10:20 pm | #5 |
Rest in Peace
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
|
Re: Hybrid receiver
If you restrict yourself to double triodes then you may have trouble with dynamic range and AGC. The only remote cutoff triodes have too much gain, as they were intended for VHF front-ends rather than HF and IF.
|
7th Sep 2010, 1:42 am | #6 | |
Retired Dormant Member
Join Date: Nov 2009
Location: Carmarthen, Wales, UK.
Posts: 79
|
Re: Hybrid receiver
Quote:
Dynamic range & sensitivity seems to be OK for now I've been using a cascode RF amp ahead of a pullen style mixer. Still its one of those long term projects at the breadboard stage, keeps me busy! |
|
7th Sep 2010, 10:08 am | #7 |
Tetrode
Join Date: Jan 2007
Location: Lincoln, Lincs. UK.
Posts: 94
|
Re: Hybrid receiver
Thanks for all the interesting comments on this. Part of what contributed to the thought processes was that in the past I have owned a number of solid state receivers and transceivers, but none of them compare in terms of signal and noise handling to the AR88. This does not appear to be affected at all by being surrounded by various RF 'dirty' devices such as computers and computer switches, CRT screen etc. Also, it does not seem to pick up the static crashes as much during high static periods. Is this down to the greater signal handling capacity in the front end, and/or better front end filtering?
|
7th Sep 2010, 11:07 am | #8 |
Retired Dormant Member
Join Date: Nov 2009
Location: Carmarthen, Wales, UK.
Posts: 79
|
Re: Hybrid receiver
I suppose this shows how good the original almost no compromise design and build of the AR88 was. Out of interest what where you comparing it too?
|
7th Sep 2010, 12:49 pm | #9 |
Dekatron
Join Date: Jul 2008
Location: Charmouth, Dorset, UK.
Posts: 3,601
|
Re: Hybrid receiver
In the early days manufacturers would do anything to get the word 'Transistor' on their sets. Pye brought a range of TVs with one transistor - in the sync circuit, but it enabled them to display the word 'Transistor' on the front.
The first really succesfull combination I remember was in the Pye TRC 1000 car radio (if I remember correctly) it used a special range of valves that ran with 12v HT and an OC16 transistor in the output stage, although only single ended it gave surprising good quality. The real breakthrough though was the fact that an HT supply was no longer requred, thus considerably reducing the size, weight and cost. Peter |
7th Sep 2010, 2:22 pm | #10 |
Retired Dormant Member
Join Date: Jun 2010
Location: Co. Limerick, Ireland.
Posts: 1,183
|
Re: Hybrid receiver
I can't think of any place I'd use a valve for preference in an HF radio or Transmitter.
I have an FT101ZD with 12BY7 driver and 2 x 6146 in parallel rather than push pull. I'd not change any of it. But I have built "better" P.A. for 36V using £1 SMPSU FETs with same o/p power. I've built valve gear since early 1970s. But I have also designed and built "state of the art" 900MHz LNA, mixer, Synth etc. There are ICs now with amazing dynamic range (more important than noise figure on HF RF stage and IF). NF is not an issue. The really low noise IC cascode amps (<0.6dB) are only good above 100MHz. But HF noise is so much higher, that simply picking an IC based on Dynamic range will do. Such as GAV84 or GALI74 DC to GHz MMICs as they have heat sink and can handle High IP3, +38 dBm at 0.1 GHz, < 3dB NF, typically 26dB gain and 20dBm pout 1dB compression. If you want a wide band 1st IF with massive overload ability use HELA10. It's designed to handle 110 x 8MHz analogue or Digital TV channels at once! (Minicircuits) For synth use LMX series, but only 1MHz steps. (Use 1MHz to 10MHz wide 220MHz 450MHz st IF with LC and SAW filters!) Use an AD DDS for 1Hz tuning on final IF. Use I & Q downmixers from DDS to Feed final IF into an FPGA via two 2MHz or higher sample rate 16 bit ADC as IF Filter and demodulation. 10kHz to 1.3GHz, AM, Synchronous AM (lower, upper, both), SSB, FM, NBFM, RTTY, WeFax, SSTV, PSK etc... No PC needed. Use a 3.5" to 5" colour touch screen plus PIC18F or ARM Cortex, plus keypad, loads of real knobs (all read by cpu) etc... |
7th Sep 2010, 2:52 pm | #11 |
Rest in Peace
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
|
Re: Hybrid receiver
An interesting Thread; a few thoughts . . .
It is worth bearing in mind that valves in a common-cathode configuration, compared to transistors in a common-emitter configuration, will present a much higher shunting load across any L/C circuit in the anode cct. compared to the equivalent in the transistor case. In the latter, this usually necessitates a tapped coil in the I.F. stages and lowered Q in tunable-RF amplifiers - where it is usually difficult to get adequate Q in the first place. Consequently, it is usual to find you can achieve more gain per stage in such circuits with valves than transistors. Although valves are a lot noisier than modern HF transistors, the incoming noise level in the 3 ~ 30 MHz range is usually such that it is this 'feature' that nearly always determines the ultimate 'effective' sensitivity of the overall receiver. In the VHF / UHF ranges, valves like the G.E.C. A2521 can produce excellent S/N ratios, but these freq. ranges are now almost exclusively the domain of very low noise solid-state active devices, since the incoming noise at these freqs. is not the determining factor of the overall S/N figure: usually, the noise performance of the first stage of RF amplification determines this - the gain of which needs to be adequate to overcome the subsequent noisy mixer. For oscillators, it has to be solid-state devices, mainly on account of their relative freedom from freq. drift due to (lack of) heat. For mixers & freq. conversions, I'd plump for F.E.Ts - choosing those that have the nearest square-law transfer characteristic. Valves, as mixers, are usually incredibly noisy (the 6BE6 springs to mind ) Audio stages: I'd choose solid-state integrated circuit packages, mainly on the grounds of maximising the density of component count per cubic inch. Power-supply: a bit tricky this one, since there will be a requirement for something like 250v for the valves (and, of course, LT for the transistors) - but I'd still choose solid-state, mainly on the grounds of less heat, greater density of component count & that suitable HV devices are now readily available at moderate cost. A 6v.a.c. winding for the valve heaters should be easily obtainable from a suitable transformer that provides all the required voltages: HV and LT. A stabilized LT for the various oscillators should not be difficult to arrange. Finally, serious consideration needs to be given to all the mechanical arrangements. Many have been the otherwise-excellent designs that have proven to be marginal on test on account of poor screening, jumpy oscillators (chassis too thin), drift due to poor heat dissipation & poor layout of operator controls. Overall, of course, what must be considered how much £ is available for a given receiver project. In the event of a well-filled 'bits box', there could be plenty of room for choice! So there we have it - my thoughts on this Thread - which are 'subject to amendment', following other members' experiences. Al. / Skywave. |
7th Sep 2010, 4:06 pm | #12 | ||
Rest in Peace
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
|
Re: Hybrid receiver
Quote:
Quote:
Triode and pentode mixers are quieter, although FETs and diodes are better in this respect. |
||
7th Sep 2010, 8:11 pm | #13 |
Tetrode
Join Date: Jan 2007
Location: Lincoln, Lincs. UK.
Posts: 94
|
Re: Hybrid receiver
Thanks one and all for all the interesting posts on this. As I said at the start, the musings were due in part to my incapacity to do anything other than read and think (and listen to the radio) for the last 2 weeks or so and was not a serious question with a design in mind. But it is certainly interesting to read of the advances in some of the semiconductor components (thanks neon) that are available to the designer these days. Having only dealt with, in homebrewing, with the likes of NE602/612 mixers, or simple fet mixers which can overload easily, it is interesting to read of other components that are 'out there'. What I think it possibly does show is that the designer of receivers such as the Collins range i.e. R390, must have got something very right with their designs (and fairly unlimited resources!) to produce the quality of these receivers.
My comparisons with the AR88 (which I don't put in the same league as the Collins) were such transceivers as FT707, FT101 and Trio R600. Plus of course homebrew receivers, both DC and Superhet but these can be discounted anyway due to comments above re NE602. |
7th Sep 2010, 8:16 pm | #14 | ||
Rest in Peace
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
|
Re: Hybrid receiver
Quote:
Al. |
||
10th Sep 2010, 11:01 pm | #15 |
Rest in Peace
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
|
Re: Hybrid receiver
I think the trick that Collins used was good gain distribution (i.e. less before the mixers than other people used) and good mixers (e.g. 6C4/EC90 in the R390A). They really knew how to engineer a receiver.
Other people were sticking with high gain 6BZ6 amps and noisy 6BE6 mixers. |
10th Sep 2010, 11:26 pm | #16 |
Dekatron
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,081
|
Re: Hybrid receiver
One of the hardest things in this is comparing like with like. If you use 100 transistors to replace a single valve, you will probably do better, anywhere - andmany IC's have far more than 100 transistors and yet cost less than a single valve.
G8HQP Dave has, unfortunately, summed it up in post no. 3. However, using state-of-the-art solid state, and making a 'better' receiver, you still won't have the electrical robustness of valves - immunity from static discharges, nuclear detonation, etc. And you won't have a receiver that glows in the dark... |
11th Sep 2010, 12:19 am | #17 | |
Rest in Peace
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
|
Re: Hybrid receiver
Quote:
Al. |
|
12th Sep 2010, 8:53 am | #18 |
Dekatron
Join Date: Jul 2008
Location: Charmouth, Dorset, UK.
Posts: 3,601
|
Re: Hybrid receiver
Or one that doesn't self distruct if the test prod slips!
|