General receiver questions

[brandon7861]

I have been looking at a galaxy 2547 schematic and letting my mind wonder. I have 2 questions.

#1) What kind of difference does having a TR relay make for receive quality?

The DX2547 has a great receive, but could it be better? For instance, the 2547 schematic shows an 18pF capacitor taking the RX from the filter network and coupling it into a transformer that further lowers the impedance for the common-base RX preamp. Keeping that capacitor small prevents it and whats after it from affecting the matching network much on TX and it also limits the amount of RF that makes it to the (powered off) preamp during transmit.
However, at 27MHz, 18pF is 327Ω in line with the RX path. At first I thought, hey, thats nuts, but then I saw the transformer. So, that leads me to the next question....

#2) If I were tasked with making a transformer to replace that first receive transformer, am I correct in assuming the inductance of the primary is set to cancel the reactance of that little 18pF capacitor and that the impedance ratio is then set to transform whatever it is at the RX pickup point on the output network (something under 50Ω) down to that of the common-base input?

Thanks!!!

 

[brandon7861]

 

[Riverman]

Brandon, I see you posted this at 2:00 in the morning.
At that hour, most people’s brains are in energy-saving mode not overdrive.

I believe there are four members who can answer your question. Am anxious to see which will be the first to reply. Shouldn’t be long now.

 

[Riverman]

Well, it’s been 15 hours.
It appears you may have stumped them.

 

[brandon7861]

I hope not...

I do have a parts 2547 that I could take L5 out of and investigate it, but I was more interested in the theory than the actual values they chose. Parts are often chosen based on price, availability, standard values etc, so taking it out and measuring it might not give me the answer I am looking for.

 

[Riverman]

Wish I could help you, but it‘s WAY over my head.

 

[nomadradio]

The 2547 receiver's design is pretty good the way the factory sends it out. Receiver performance is a product of how selective it is, to reject off-frequency signals, how sensitive it is to the weakest signals, and how much internal noise it creates even without an antenna.

The ideal compromise between those qualities depends in part on the environment where it's used. If the noise level coming down the antenna is never below a S-2 on the receiver's meter, making it more sensitive only makes the noise louder.

Most mods I have seen or tried raise the receiver's internal noise level, cancelling out any improvement in sensitivity to a weak signal.

I would start at the other end of the proposition, and first decide which characteristic you seek to improve, and how much.

Just changing parts to see what will change is a bit like playing pin-the-tail-on-the-donkey. Then again, if you can measure the before-and-after difference, that's probably worth finding out, just to see what makes a favorable difference, and what works the opposite way.

One accurate measurement is worth a thousand expert opinions.

73

 

[loosecannon]

Brandon you seem like a pretty sharp cookie.

If you are looking to make improvements to the receiver design, might i suggest you put some effort into this area:

ive always wanted to see what others come up with here.

as for the receiver front end, in the past people have experimented with using a MOSFET or even a GAaSFET for the RF amp.

i never got a chance to see a unit that had the mod done, but from what i understand the results were at least noticeable.

best of luck in your endeavor!
LC

 

[brandon7861]

Been gone, just got home.

I don't have a working 2547 anymore, so I wasn't looking to change that transformer, just curious about how one goes about doing it. Curiosity is all. I have been tempted to do some experimenting with receivers, and probably will soon, just not sure if that will be on an existing radio or if it will be the beginnings of a homebrew rig. I have a couple SMT low noise J310's and a few BF998 dual gate mosfets, both would probably make great receiver front ends. I originally wanted to make an entire radio out of altoid cans connected together with SMAs so each stage was shielded, maybe some day.

That single-balanced mixer does look like it needs one of these in its place

https://www.mouser.com/datasheet/2/1030/ADE-1H_2b-1700018.pdf

 

[nomadradio]

Gotta remember the difference between a passive mixer like the Mini Circuits ADE and an active mixer like the push-pull J310s.

The LO drive level for the ADE is shown as +17dbm, or around 60 (I think) milliwatts. Lots more than the Galaxy's local oscillator circuit can pump out. And since it's passive it will lose roughly 5 or 6 db of your signal level passing through it.

The factory P-P FET mixer has some signal, or as they call it "conversion" gain, so changing to a passive mixer would involve making up the gain deficit. Might reduce the receiver's internal noise level, though. And the ADE mixer would stand up to larger off-channel signals before it overloads and creates its own bleedover.

The best tool for zeroing in on the best compromise between noise, sensitivity, overload resistance and selectivity is called a noise-figure meter. Never have properly learned to use one, but who knows, maybe some day?

73

 

[brandon7861]

Gotta remember the difference between a passive mixer like the Mini Circuits ADE and an active mixer like the push-pull J310s.

The LO drive level for the ADE is shown as +17dbm, or around 60 (I think) milliwatts. Lots more than the Galaxy's local oscillator circuit can pump out. And since it's passive it will lose roughly 5 or 6 db of your signal level passing through it.

The factory P-P FET mixer has some signal, or as they call it "conversion" gain, so changing to a passive mixer would involve making up the gain deficit. Might reduce the receiver's internal noise level, though. And the ADE mixer would stand up to larger off-channel signals before it overloads and creates its own bleedover.

The best tool for zeroing in on the best compromise between noise, sensitivity, overload resistance and selectivity is called a noise-figure meter. Never have properly learned to use one, but who knows, maybe some day?

73

I figured it would need a bit of external gain. I may have been too excited about that mixer though, it's first time I saw one of those. I don't have much experience with integrated circuits and I had no idea I could get a passive mixer in a chip..

 

[brandon7861]

I'd like to make a preamp for 11m. Not sure where to start though. I think I would try using a geared variable capacitor in parallel with the best inductor I can build and have it resonate on 11m. The antenna jack would be tapped at a low impedance point on the coil and use ether one of these jfets or mosfets to sample off the top of the tank. Not sure how the antenna will load it down, but the fets shouldn't load it much. If I can keep the BW to a couple dozen kHz or so in that first amp and use one of those low noise parts there, I think i'd have it made. The noise figure on these low noise jfets and mosfets are both 1dB at 100MHz, so I'd think they would do better at 11m. I don't know...

 

[Riverman]

Do you ever get the feeling the world is a tuxedo and you are a pair of brown shoes? That’s the way I feel reading this thread. Thought I knew a good bit about radios, but clearly I don’t.

I can’t take credit for the shoe quote. That was George Gobel on Johnny Carson’s Tonight Show many years ago.

 

[brandon7861]

Thought I knew a good bit about radios, but clearly I don’t.

Thought I did too, turned out to not be the case lol. Don't assume everything I say is accurate, especially in this thread, im just putting up my best guess as I ask. My last working amplifier was for audio, so yea, Im not there yet. I have lots of theory crammed upstairs. but very little practice.