Best buy SA for conducted EMC pre-compliance tests for SMPS

[bodnandi]

Hi,

I read around a bit on the topic, and although many threads touch upon this, I haven't came up with a straight answer.

I m looking to make a proposal to buy a SA for our company. We are a small company and are developing products that chop the AC mains directly, with a bunch of LC and EMC filters before and after the chopping. As some of the products are coming to a late stage in development, it would be about time to get a feel of the EMC performance. As going to a test lab costs a fortune, the idea is to do the testing to some extent in house. Options are: New Rohde and Schwarz ESPR3 with various addons in excess of £50k (out of budget); Refurbed HP E7402 over £6k; or a new Siglent SSA3021x with EMI kit at £2200.

As to a newbee the following comparison makes the argument for the Siglent a no brainer:

Agilent E7405A has:
DANL: pre amp off @ 2GHz: -135dBm
Min Res B/W: 10Hz
Phase noise: -90dB @ 10KHz off.
Total amplitude accuracy: ±0.54 dB+ absolute frequency response

Siglent SSA3021x has:
DANL: pre amp off @ 2GHz: -135dBm
Min Res B/W: 1Hz
Phase noise: -98dB @ 10KHz off.
Total amplitude accuracy: ±0.7

I have been warned about saturating the input on the Siglent and thus not having correct results (due to large amount of noise from the chopping), but I guess there is a way to check if that happened or not. What else can go wrong?

So my question really is: What is the line of thoughts that I should take at this situation?

[tautech]

I have been warned about saturating the input on the Siglent and thus not having correct results (due to large amount of noise from the chopping), but I guess there is a way to check if that happened or not. What else can go wrong?

Any spectrum analyser can suffer from an overloaded input and the operator must take all possible steps to prevent it or at least minimize the chance of it happening. The SSA3kX's have up to 51dB of internal attenuation and a max of 50V input into the inbuilt DC block. However an external DC block and additional attenuators are wise purchases to prevent internal damage of these very sensitive instruments.

[AndyC_772]

Last time I had this problem to solve, the difficulty wasn't finding an SA, it was finding a reasonably priced, working, calibrated LISN. I suggest making sure you have a good solution for this first.

You don't need high bandwidth for PSU noise measurements, and for qualitative measurements, any working SA at all is probably OK. If your design is a bit close to the limit, then you might want to find one with a quasi-peak detector, and you might want to look at getting your complete set-up calibrated.

Something like a used HP 8594E could be a good buy.

[bodnandi]

Thanks both,

I am going for it then.

I got a quote for a LISN at £1200 - HM5060. Should be calibrated and all set up.

The quasi-peak detector is available with the EMI kit of the SSA3021.

Cheers