| Products > Test Equipment |
| HDO4204 vs. SDS2104X HD |
| << < (4/6) > >> |
| Fungus:
--- Quote from: 2N3055 on February 26, 2023, 09:20:36 am ---Dear Fungus, You seem to misunderstood this or don't understand it well.... So let's learn together. --- End quote --- :popcorn: --- Quote from: 2N3055 on February 26, 2023, 09:20:36 am ---Rigol uses both ADC interleaved to achieve 4GS/s. From any single channel. Meaning analog routing to ADCs is running together from all channels to both ADCs in some way. --- End quote --- Oh, they're not separating the ADCs? You could have just said that and saved yourself some typing. In that case I retract what I said. There's probably some variation with combinations of channels but no guarantees. Measurements would have to be made to see how bad it is in practice. I guess it makes sense because they're using the same PCB for the HDO1000 which only has one ADC. |
| Kleinstein:
The channel to channel separation is usually dependent on the gain settings and the specified values should be an upper limit or the worst case combination. Chances are the a signal the is already attenuated at the input stage has less cross talk than one with a gain setting with less attenuation. The Rigol scope is still quite new and the specs may not be final - e.g. newer PCB versions could improve on things like channel separation or the UI. The channel separation would be one of the points to actually check on real hardware examples - possibly as port of the inital checks for a new scope to make sure it works well enough. It may even vary from unit to unit depending on how good a ground contact of a shield is. Using both ADC combined to get the full 4 Gs/s does not help with the channel saparation from channels 1/2 to 3/4 but there is still the seprate from chanel 1 to 2 that already share the same ADC chip for the most DSOs, including the SDS2104. I don't think the different ADC connections would explain that much difference in the isolation specs. For the isolation there can still be good and bad designs. The Rigol specs look pretty poor, even if looking only the the < 500 MHz range ( the SDS2104 does not support > 500 MHz at all) as it is rare to have a large signal at >500 MHz. The difference in the specs is 60 dB = factor 1000 for the sigilent and 100:1 for the rigol - so a fator of 10 for the comparable specs. Chances are the Rigol will also get better with lower frequency - there is just no separate spec values for 100 MHz or 350 MHz. Much of the channel sepration problem also comes with the probes: for the higher frequency signals on has to connect the ground on both probes and in most use cases one probes a single circuit with a linked ground. So a ground loop is a very common situation. Ideally one would have the channels isolated, at least for the higher frequency part. I wonder if it would make sense to have a clamp on ferrite with the probes. |
| Fungus:
It's also going to depend on whether your probes are in 10x mode or not ... and a dozen other things. --- Quote from: Kleinstein on February 26, 2023, 01:19:12 pm ---The Rigol scope is still quite new and the specs may not be final - e.g. newer PCB versions could improve on things like channel separation or the UI. The channel separation would be one of the points to actually check on real hardware examples - --- End quote --- ">100:1" seems very low to me. It's far lower than a Rigol DS1054Z or MSO5000 (both are 40dB). It's also a suspiciously round number and it ought to be in decibels. It's almost as if it's a made-up number for the first revision of the manual. That's just speculation though. Can anybody measure one? Just feed a big signal into channel 1 and see if it appears on channel 2. |
| switchabl:
I am not sure what you are trying to say? 100:1 is exactly the same as 40 dB, assuming that this refers to a voltage ratio. |
| Fungus:
--- Quote from: switchabl on February 26, 2023, 02:55:02 pm ---I am not sure what you are trying to say? 100:1 is exactly the same as 40 dB, assuming that this refers to a voltage ratio. --- End quote --- Oh, duh! I'm thinking of power gain, not voltage gain. One of those days... 40dB power gain = 10000:1 40dB voltage gain = 100:1 The point stands though. It seems suspiciously low considering all the effort they put into the rest of the front end. |
| Navigation |
| Message Index |
| Next page |
| Previous page |