Norton Amp with Augmentation (Trask)

[Colon]

Hi,

I have a Norton amp that works over the HF band with reasonable distortion and was thinking of trying Chris Trask's method of augmentation (see attached paper) to see what kind of improvement I could get.

I see I can use passive augmentation with a transformer with appropriate turns ratio or I can use active augmentation (at the expense of some noise). Is there anything I need to be aware of? Are there limits to how much performance you can get out of it? The reason I ask is that my figures are already better than the figures he quote in his paper, after augmentation, so I'm not sure if it will give me a big improvement

Has anyone tried it and had any success with it?

Thanks
James

 

[Bob60]

esquse Mr. Colon me. That is technology 20-30 years old. You can get a lot of fun for... with these. Now we have a lot of MIMIC s with various IP3, P1 saturation and accordingly, proportional current , from tens to hundreds milliamp. Top of them you can find out at following companies : Microcircuit, RFMD, AWAGO, Hittite, MaCom, Edinija and so on. For instance SGA6486 low current P1 up to 21dBm, IP3_35dBm and bandwidth up to 6GHz.
https://www.rfmd.com/CS/Documents/SGA-6486DS.pdf

 

[Colon]

esquse Mr. Colon me. That is technology 20-30 years old. You can get a lot of fun for... with these. Now we have a lot of MIMIC s with various IP3, P1 saturation and accordingly, proportional current , from tens to hundreds milliamp. Top of them you can find out at following companies : Microcircuit, RFMD, AWAGO, Hittite, MaCom, Edinija and so on. For instance SGA6486 low current P1 up to 11dBm and bandwidth up to 6GHz.
https://www.rfmd.com/CS/Documents/SGA-6486DS.pdf

Ok thanks, that's not quite what I'm looking for though. I am working with specs around P1dB >25dBm, with IIP2> 80 and IIP3 > 45dBm. Sometimes the older technology is what you need.

Thanks

James

 

[Bob60]

Ok thanks, that's not quite what I'm looking for though. I am working with specs around P1dB >25dBm, with IIP2> 80 and IIP3 > 45dBm. Sometimes the older technology is what you need.

Thanks

James

That is closer to you.
https://www.rfmd.com/CS/Documents/SXA-389BDS.pdf

 

[Colon]

That is closer to you.
https://www.rfmd.com/CS/Documents/SXA-389BDS.pdf

Even in the HF band?

 

[Bob60]

Yeah, it is up to 2.5-3GHz.MaCom & Hittite has for highest frequencies.
**broken link removed**

 

[Colon]

Yeah, it is up to 2.5-3GHz.MaCom & Hittite has for highest frequencies.
**broken link removed**

Thanks for that. They do look good but I don't think they are quite good enough performance for us. The key point I think is that they are quoting output referred IP2 and IP3. My figures above were input referred, with 7 dB of gain, and we are looking to improve on that so I don't think a MMIC will suit.(The best in the document you attach has OIP3 of 45 at low frequencies. We already achieve IIP3 of 45, which is OIP3 of 52.)

 

[Bob60]

I see but what frequency you talk about and is that cable solution, some CATV?

- - - Updated - - -

Should be warmer output Power (P3dB) =39.5dBm and OIP3 should be 53-55dBm. Gain as well 9-10dB.
https://www.rfmd.com/CS/Documents/RF3826DS.pdf

 

[Colon]

Thanks again. This does look quite good actually, although we would have to run some tests as not all the parameters are specified.

We're looking at HF comms.

- - - Updated - - -

One other thing, our existing amp has NF < 2dB. I can't find anything about NF on the device you suggest, but I will keep looking. I would think it would be higher than2dB though.

 

[Bob60]

Thanks again. This does look quite good actually, although we would have to run some tests as not all the parameters are specified.

We're looking at HF comms.

- - - Updated - - -

One other thing, our existing amp has NF < 2dB. I can't find anything about NF on the device you suggest, but I will keep looking. I would think it would be higher than2dB though.

For MIMIC yes, typically there are feedbacks for freq linearity and as result NF 3dB for SiGe technology and 5dB GaAs. But last one is nitride internal matched stage-transistor and here could be desired NF 2dB. Data sheet is not clear but company RFMD is placed somewhere at the England wide green fields. You can ask them?