The little Marconi 2022 generators can be prone to microphony effects that modulate the synthesiser through vibration. This adds spurious sidebands that appear close to the carrier as in your analyser plot. The internal cooling fan in the 2022 can be a major offender if it is old or loose and causing vibration. Temporarily try forcing the fan to stop to see if the sidebands improve on the 2022E.
Thanks very much for the information. I no longer have that version. I've now got the D version which is white. I'm not sure if I've checked it for noise sidebands. I will test it sometime. Its in a stack of equipment so difficult to remove for doing the fan test. But I will at some point. Regards Nick
Hey there. The noise on the 2022e looks a lot more like AM noise than phase noise, and I don't think it is due to design, but rather an issue that it is having. I've never had a 2022, so I could be wrong. But I think you should dig in and diagnose that one---it could be a misbehaving amplifier somewhere in the chain, due to a bad capacitor letting some noise in the power line, or a bad op-amp. It appears to me by the trace that you have one or two primary AM modulation carriers that are mixing several times and giving the multiple peaks, sortof like intermodulation distortion.
I just realized they weren't all on the same reference---first thing I would do is put both the 2022s on a single reference (if the C model has a reference output that you could connect to the E, that would be ideal) and see if there is any change. If there is, and the two look closer together in terms of the noise, then you can be sure the issue is in the reference oscillator section of your E model. If not, and they still look as different as they did in this video, then it's probably an amplifier somewhere in the 2022E's signal chain.
You could well be right, especially the two small peaks either side of the carrier. But I'm doing no more work on it, I've spent many hours fixing it in the past. In normal use you wouldn't even notice a problem. It's only when you go deep in to spectrum analysis. Thanks for watching
@@ptronix Hehe I totally understand. Especially when you already have multiple instruments that do the same thing, sometimes it's too much effort for little gains to put time into fixing small issues. So much easier just to put a physical or mental sticker on the unit that says "works great, but don't trust for critical stuff". 😀
Something not right with 2022E. Not just phase noise, but you can see the equal spaced harmonic dirt on each side of the fundamental. On the flip side, the IFR is impressive. Thanks as always.
Yes Danish there may be issues with it, but the spurious signals are way down on the carrier. Its only when you go deep down into spectrum analysis that you see these things. I checked it on a radio today with a plain carrier, absolutely no noise or modulation. So it's good enough. The ifr is great! I was very lucky to find it fairly cheap. Thanks for watching
Could it be that the C has all the original gain blocks, but the E has had them all replaced with substitutes thats causing the noise? Also the brass cover on the RF board has been known to cause problems if it's not tightened down evenly, of if there's dirt/grease on the mating surfaces. I have an E version & the same spectrum analyser as you. My generator starts to produce noise around 56dB down, compared to yours at 40dB down. Anyway, I assume your one has long been sold on Ebay by now. :)
The original E version was the first one I owned and one by one all the om345 gain blocks failed. I replaced them with modern mmic's. I then had the C version with all original om345's all working still. Now have the D version which is the last version they made, again using om345's and again no failures in those modules. So maybe they had a bad batch at one time?
@@ptronix Those OM345 modules certainly have a high failure rate. It's been suggested that it may be down to thermal stress, breaking the track on the R3 printed resistor. Don't know if it's batch related, but only the other week I noticed that the 1dB steps don't work at 500MHz & up on one of my 2022E generators. It has 7,420 hours on the clock. What's the betting that it's down to an OM345!
@@Manticore1960 well my D version has a lot more hours on it than that and as far as I know it's got the original om345's . Your problem may be related to the gain block, but to do the 1 db steps it uses a pin diode attenuator. I wonder if the rf levels are close to normal?
@@ptronix As my fault was only in the range of 500MHz & up, I looked at the components in the frequency doubler circuit first. Sure enough IC3 (OM345) had failed. Reading 0v on pin 1 & 12v on pin 5. Used the 4k7 resistor trick to get it going again. That's the 4th OM345 to fail in this unit now.
Once again your drunken purchase has been a good one!
The little Marconi 2022 generators can be prone to microphony effects that modulate the synthesiser through vibration. This adds spurious sidebands that appear close to the carrier as in your analyser plot. The internal cooling fan in the 2022 can be a major offender if it is old or loose and causing vibration. Temporarily try forcing the fan to stop to see if the sidebands improve on the 2022E.
Thanks very much for the information.
I no longer have that version. I've now got the D version which is white. I'm not sure if I've checked it for noise sidebands. I will test it sometime. Its in a stack of equipment so difficult to remove for doing the fan test. But I will at some point. Regards Nick
Hey there. The noise on the 2022e looks a lot more like AM noise than phase noise, and I don't think it is due to design, but rather an issue that it is having. I've never had a 2022, so I could be wrong. But I think you should dig in and diagnose that one---it could be a misbehaving amplifier somewhere in the chain, due to a bad capacitor letting some noise in the power line, or a bad op-amp. It appears to me by the trace that you have one or two primary AM modulation carriers that are mixing several times and giving the multiple peaks, sortof like intermodulation distortion.
I just realized they weren't all on the same reference---first thing I would do is put both the 2022s on a single reference (if the C model has a reference output that you could connect to the E, that would be ideal) and see if there is any change. If there is, and the two look closer together in terms of the noise, then you can be sure the issue is in the reference oscillator section of your E model. If not, and they still look as different as they did in this video, then it's probably an amplifier somewhere in the 2022E's signal chain.
You could well be right, especially the two small peaks either side of the carrier. But I'm doing no more work on it, I've spent many hours fixing it in the past. In normal use you wouldn't even notice a problem. It's only when you go deep in to spectrum analysis. Thanks for watching
@@ptronix Hehe I totally understand. Especially when you already have multiple instruments that do the same thing, sometimes it's too much effort for little gains to put time into fixing small issues. So much easier just to put a physical or mental sticker on the unit that says "works great, but don't trust for critical stuff". 😀
@@ruhnet yes indeed, its going on ebay when they do a special offer on cheap selling fees. Hopefully I will get back what I paid for the 2022c!
Something not right with 2022E. Not just phase noise, but you can see the equal spaced harmonic dirt on each side of the fundamental. On the flip side, the IFR is impressive. Thanks as always.
Yes Danish there may be issues with it, but the spurious signals are way down on the carrier. Its only when you go deep down into spectrum analysis that you see these things. I checked it on a radio today with a plain carrier, absolutely no noise or modulation. So it's good enough. The ifr is great! I was very lucky to find it fairly cheap. Thanks for watching
Hi Nick, what about 2nd and 3rd Harmonics?
Hi Simon I checked all that in previous videos
Could it be that the C has all the original gain blocks, but the E has had them all replaced with substitutes thats causing the noise? Also the brass cover on the RF board has been known to cause problems if it's not tightened down evenly, of if there's dirt/grease on the mating surfaces. I have an E version & the same spectrum analyser as you. My generator starts to produce noise around 56dB down, compared to yours at 40dB down.
Anyway, I assume your one has long been sold on Ebay by now. :)
The original E version was the first one I owned and one by one all the om345 gain blocks failed. I replaced them with modern mmic's.
I then had the C version with all original om345's all working still.
Now have the D version which is the last version they made, again using om345's and again no failures in those modules. So maybe they had a bad batch at one time?
@@ptronix Those OM345 modules certainly have a high failure rate. It's been suggested that it may be down to thermal stress, breaking the track on the R3 printed resistor. Don't know if it's batch related, but only the other week I noticed that the 1dB steps don't work at 500MHz & up on one of my 2022E generators. It has 7,420 hours on the clock. What's the betting that it's down to an OM345!
@@Manticore1960 well my D version has a lot more hours on it than that and as far as I know it's got the original om345's . Your problem may be related to the gain block, but to do the 1 db steps it uses a pin diode attenuator. I wonder if the rf levels are close to normal?
@@ptronix As my fault was only in the range of 500MHz & up, I looked at the components in the frequency doubler circuit first. Sure enough IC3 (OM345) had failed. Reading 0v on pin 1 & 12v on pin 5. Used the 4k7 resistor trick to get it going again. That's the 4th OM345 to fail in this unit now.