EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: randomOracle on September 01, 2019, 08:46:10 pm
-
Hi everyone,
One of the features that is still very difficult to get at a low price is the 10 MHz REF_IN, i.e., using an external signal generator and its reference/synchronization output to phase-lock the scope to the output of the signal generator. I recently bought a Siglent SDS2000X and was wondering if this scope or any other scope in the same price range can be hacked such that the 10 MHz REF_IN can be provided? Or extracted?
According to the teardown: https://www.youtube.com/watch?v=E3B4OTV8f1o (https://www.youtube.com/watch?v=E3B4OTV8f1o)
The relevant (?) clock gen is the ADF4360: https://www.analog.com/media/en/technical-documentation/data-sheets/ADF4360-7.pdf (https://www.analog.com/media/en/technical-documentation/data-sheets/ADF4360-7.pdf)
This one at least has a REF_IN made for 10 MHz, so it must be generated somewhere. Therefore, I think we should be able to extract the 10 MHz from the scope or to provide it externally from the signal generator.
Did someone ever look into this? Any help appreciated before destroying my scope ;-)
-
If you don't mind me asking, what do you expect to accomplish with this? What is use case for you so built in timebase has to be synchronized to outside world, or that internal timebase is not accurate enough?
The way I see it, your decision should be driven by clear benefits, or otherwise you might damage perfectly good scope for no real gain...
My 2 cents..
-
If you can find the reference oscillator, then it should not be a problem.
-
Yes to both, there should be a TCXO in a little can somewhere near the clock generation for the ADCs, it shouldn't be too tricky to find, but it may be worth probing the oscillator output to be sure it's the right one.... but I agree that it probably won't be helpful.
If you really want to try to synchronize signals, you can always trigger off of a different input signal (like channels 1 and 2 for measurement and 3 for your trigger signal), and aside from internal frequency counter accuracy, I don't see a lot of benefits with lining up your sample clock with your input signal (or vice versa). If you're synchronizing multiple units, maybe, or maybe if you're trying to null out drift in the long term? The triggering method should fix the second, at least.