You can now help support WorldwideDX when you shop on Amazon at no additional cost to you! Simply follow this Shop on Amazon link first and a portion of any purchase is sent to WorldwideDX to help with site costs.
The Feb 2025 Radioddity Giveaway Results are In! Click Here to see who won!

Reply to thread

Message

Hi LC
Yes, one of the things to do is alter the coupling cap. However, this does reduce the signal level so there is a tradeoff here. To counter this the gain of the first stage can be increased by altering the bias point. However, this needs care to avoid instability and there's other mods required to prevent instability.

I'll have a think about this but I used to do it with just my ears and a signal generator because you can hear that L7 has a noisy side to its adjustment.

It might be possible to do it with an old analogue AC volmeter even if it doesn't read true rms. I'll see if this is possible.

BTW my 2SC2999 transistors arrived today. They are marked as D versions although this really isn't significant.

I have just done a comparison test between this and a standard radio.

I did these tests on a mk2 SS360FM PB010 which is the same as the mk2 148GTL-DX.

Basically, the amplifier gain as measured on the network analyser and probe was identical (within 0.1dB)

This was measured after optimising to try and match the gain response of the original 2SC1674L. The difference was so tiny between them it wasn't worth measuring.

I then did some sensitivity tests.

Now it's very difficult to get the 'same' results across all 120 channels even with no changes if L7 gets tweaked but I tried very hard with the 2SC2999 to get the lowest noise setting for L7.

Here are the results for SSB sensitivity on the high tone setting

2SC1674L

26.515MHz -119.7dBm for 10dB (S+N)/N
27.185MHz -122dBm for 10dB (S+N)/N
27.855MHz -119dBm for 10dB (S+N)/N

2SC2999D
26.515MHz -119.7dBm for 10dB (S+N)/N
27.185MHz -122.5dBm for 10dB (S+N)/N
27.855MHz -119dBm for 10dB (S+N)/N

You can see than the 2SC2999 was 0.5dB better in the middle of the band.

This is a very small improvement and I could maybe get the same result on the 2SC1674L if I tried as hard to find the sweet spot for NF at 27.185MHz.

However, testing it for NF in the radio isn't really fair on the 2SC2999 and if there is an improvement in transistor noise figure then the best way to quantify this is to test it in isolation with a test circuit and use my other test gear. Not sure when I can do this. The reason I think it is worth doing is because it may show that there is more than a 0.5dB improvement in the amplifier itself.

If so then this transistor may support my other mods slightly better.

I should point out that a 0.5dB change in S/N ratio is so tiny it really isn't going to be perceptible to human ears.

There's probably a few people that can spot it in controlled tests when flicking back and forth but I don't think I can do it...

<blockquote data-quote="G0HZU" data-source="post: 328851" data-attributes="member: 19109">Hi LCYes, one of the things to do is alter the coupling cap. However, this does reduce the signal level so there is a tradeoff here. To counter this the gain of the first stage can be increased by altering the bias point. However, this needs care to avoid instability and there&#039;s other mods required to prevent instability.I&#039;ll have a think about this but I used to do it with just my ears and a signal generator because you can hear that L7 has a noisy side to its adjustment.It might be possible to do it with an old analogue AC volmeter even if it doesn&#039;t read true rms. I&#039;ll see if this is possible.BTW my 2SC2999 transistors arrived today. They are marked as D versions although this really isn&#039;t significant.I have just done a comparison test between this and a standard radio.I did these tests on a mk2 SS360FM PB010 which is the same as the mk2 148GTL-DX.Basically, the amplifier gain as measured on the network analyser and probe was identical (within 0.1dB)This was measured after optimising to try and match the gain response of the original 2SC1674L. The difference was so tiny between them it wasn&#039;t worth measuring.I then did some sensitivity tests.Now it&#039;s very difficult to get the &#039;same&#039; results across all 120 channels even with no changes if L7 gets tweaked but I tried very hard with the 2SC2999 to get the lowest noise setting for L7.Here are the results for SSB sensitivity on the high tone setting2SC1674L26.515MHz&nbsp; &nbsp; &nbsp; -119.7dBm for 10dB (S+N)/N 27.185MHz&nbsp; &nbsp; &nbsp; -122dBm for 10dB (S+N)/N 27.855MHz&nbsp; &nbsp; &nbsp; -119dBm for 10dB (S+N)/N 2SC2999D26.515MHz&nbsp; &nbsp; &nbsp; -119.7dBm for 10dB (S+N)/N 27.185MHz&nbsp; &nbsp; &nbsp; -122.5dBm for 10dB (S+N)/N 27.855MHz&nbsp; &nbsp; &nbsp; -119dBm for 10dB (S+N)/N You can see than the 2SC2999 was 0.5dB better in the middle of the band.This is a very small improvement and I could maybe get the same result on the 2SC1674L if I tried as hard to find the sweet spot for NF at 27.185MHz.However, testing it for NF in the radio isn&#039;t really fair on the 2SC2999 and if there is an improvement in transistor noise figure then the best way to quantify this is to test it in isolation with a test circuit and use my other test gear. Not sure when I can do this. The reason I think it is worth doing is because it may show that there is more than a 0.5dB improvement in the amplifier itself.If so then this transistor may support my other mods slightly better.I should point out that a 0.5dB change in S/N ratio is so tiny it really isn&#039;t going to be perceptible to human ears. There&#039;s probably a few people that can spot it in controlled tests when flicking back and forth but I don&#039;t think I can do it...</blockquote>

[QUOTE="G0HZU, post: 328851, member: 19109"] Hi LC Yes, one of the things to do is alter the coupling cap. However, this does reduce the signal level so there is a tradeoff here. To counter this the gain of the first stage can be increased by altering the bias point. However, this needs care to avoid instability and there's other mods required to prevent instability. I'll have a think about this but I used to do it with just my ears and a signal generator because you can hear that L7 has a noisy side to its adjustment. It might be possible to do it with an old analogue AC volmeter even if it doesn't read true rms. I'll see if this is possible. BTW my 2SC2999 transistors arrived today. They are marked as D versions although this really isn't significant. I have just done a comparison test between this and a standard radio. I did these tests on a mk2 SS360FM PB010 which is the same as the mk2 148GTL-DX. Basically, the amplifier gain as measured on the network analyser and probe was identical (within 0.1dB) This was measured after optimising to try and match the gain response of the original 2SC1674L. The difference was so tiny between them it wasn't worth measuring. I then did some sensitivity tests. Now it's very difficult to get the 'same' results across all 120 channels even with no changes if L7 gets tweaked but I tried very hard with the 2SC2999 to get the lowest noise setting for L7. Here are the results for SSB sensitivity on the high tone setting 2SC1674L 26.515MHz -119.7dBm for 10dB (S+N)/N 27.185MHz -122dBm for 10dB (S+N)/N 27.855MHz -119dBm for 10dB (S+N)/N 2SC2999D 26.515MHz -119.7dBm for 10dB (S+N)/N 27.185MHz -122.5dBm for 10dB (S+N)/N 27.855MHz -119dBm for 10dB (S+N)/N You can see than the 2SC2999 was 0.5dB better in the middle of the band. This is a very small improvement and I could maybe get the same result on the 2SC1674L if I tried as hard to find the sweet spot for NF at 27.185MHz. However, testing it for NF in the radio isn't really fair on the 2SC2999 and if there is an improvement in transistor noise figure then the best way to quantify this is to test it in isolation with a test circuit and use my other test gear. Not sure when I can do this. The reason I think it is worth doing is because it may show that there is more than a 0.5dB improvement in the amplifier itself. If so then this transistor may support my other mods slightly better. I should point out that a 0.5dB change in S/N ratio is so tiny it really isn't going to be perceptible to human ears. There's probably a few people that can spot it in controlled tests when flicking back and forth but I don't think I can do it... [/QUOTE]

Top Bottom