Manhattan island cutter and HF pre amp

[mole42uk]

When I see the title “Manhattan island cutter” I keep thinking it’s about an American sailing ship……

[G0HZU_JMR]

Originally Posted by Terry_VK5TM

Just adding my thanks to Jeremy for the info he has added - I played with this circuit some time ago and didn't have much luck with it.
I do note seeing that circuit some time ago that split the R5 feedback resistor with a cap to ground in the middle to tailor the frequency response.
Might give it another go now with the extra info above

Hi Terry, thanks. This circuit can give problems with gain flatness and a lot depends on the BJT used and the gain the amp is designed for. Sometimes the response will show a fairly pronounced peak at the top end before it rolls off and other designs will be quite droopy. In my experience it is asking a lot of a typical BJT Spice model to capture the way this amplifier behaves towards the upper end of its passband. Spice models tend to be a bit optimistic here in my experience and they often fail to model the high frequency behaviour of this amplifier correctly.

Sometimes L and C components are included with the feedback resistor R5 as you say and this can give a flatter response at the upper end of the amplifier range.

Hopefully the spreadsheet I've uploaded will be OK across various design choices in terms of supply voltage and gain and TR2 collector current. I did put it together quite quickly and it really is meant as a starting point. If a split power supply is used then I think it should be possible to do away with R6 and this would reduce the noise figure. However, this would require a different spreadsheet to be created.

I've just had a go at making the original circuit from Malcolm in post #1. It does seem to work as expected. I've turned on some test gear and when it warms up I'll have a go at measuring the noise figure.

The noise figure is usually quite high with this type of amplifier and the other niggle is the way the linearity collapses once above a certain drive level. That's why this amplifier often gets used in a controlled environment and it might be used as a VFO buffer or maybe as a post mixer amplifier in a receiver that has AGC in the system somewhere.

I keep meaning to box up the BFR91 based version I have here and this thread is a timely reminder to finally get this done. This amp is a useful general purpose test amplifier to have on the workbench because it offers excellent reverse isolation and the linearity is quite good as long as it doesn't get driven above the knee point where the linearity suddenly collapses.

[G0HZU_JMR]

Here's the noise figure plot for the original preamp circuit where R1 and R2 are both 47R, R3 is 680R, R5 is 820R, R6 is 100R and R4 is not fitted.

I built this using the ugly method using a pair of BC547B transistors from Farnell. These are Multicomp BC547B parts so I don't know the real manufacturer.

The first plot show the gain at 6.5MHz is just under 18dB and the noise figure is just over 12dB. The second plot shows the impact of adding a 100uH choke in series with R6. The noise figure improves to be about 8dB. This still isn't that great although the noise figure can sometimes be a couple of dB lower than this if better transistors are used.

[G0HZU_JMR]

I dug out the BFR91 version and this still has the 100uH choke fitted. See below for the gain and noise figure. This amplifier was designed for a lower gain of about 13.5dB and you can see how flat the gain response is. The noise figure is just over 5dB with a 100uH choke in series with R6. This is a lot better than the BC547B version. However, there is a real risk of instability when using BFR91 BJTs in this circuit so the layout and the choice of components is fairly critical.

It's tempting to build it using a classic NEC 2SC3355 BJT as I would expect the noise figure to be even lower but stability and the gain response might be difficult to manage with this device.