JANUS OpAmp Bakeoff

From HPSDRwiki
Jump to: navigation, search

While Lyle's (KK7P) Janus board was here for some checkout with the PowerSDR software I did some A/B comparisons of Lyle's board using TI NE5534ADR op amps in place of the LT1128 op amps called for in the original design. The op amps Lyle used are considerably cheaper (at $1.14 in singles from Digikey -- the design uses four) than the LT parts (at $8.63 in singles -- also requires four parts), so if they work close to as well as the LT's we'd like to use them in the final design.

The first test I tried was MDS using an SDR 1000 and a signal generator. I did not notice any difference between the two - both came in with an MDS of less than -127 dbm on 40 meters (my sig gen does not go below -127 dbm and I've misplaced my attenuators - argh)

The second test I tried was to look at the spectrum display from PowerSDR when the Janus was fed from an audio signal generator with the Janus attenuators jumpered into the circuit. I set my audio generator to its lowest level (about 0.5 V) and then switched in 60 db of attenuation. This resulted in a pretty strong signal, so I quickly threw together a 30 db pi network attenuator and put that in line (it appears my 30 db attenuator is more like a 20 db attenuator) The screen shots are below - LT1128 op amp on the left and NE5534 on the right.

Note: I've updated this to indicate the attenuators on the Janus were in line for the screen shots -- Phil Harman (VK6APH) pointed out I'd misinterpreted the sense of the attenuator jumpers -- ugh.

60 db attenuation of 0.5 v 12.8 khz input signal 60db.PNG

60 + 30 db attenuation of 0.5 v 12.8 khz input signal 90db.PNG

60 + 30 db attenuation of 0.5 v 12.8 khz input signal - peak mode selected on spectrum display 90db-peak.PNG

I don't see much difference in the spectrum displays - suspect we could get away with the NE5534 op amps on Janus

Bill Tracey (kd5tfd) 20 Oct 2006 (slight edits Lyle KK7P 20 Oct 2006)