Our customer has a trouble about AD7657-1. Only CH 1 is noisy as compared to other CHs. Do you have any idea about causes ans/or countermeasures?
Thank you for your answer.
I think so, too. But there are some questions.
The customer have used AD7657 for their some projects in past time. They have not seen like that noise at those time. This time, they modified the board fusing AD7657 and use AD7657-1. Major modified point is reduction of capacitor of power supply lines. Overall layout and routing of the wiring are not changed. However, CH1 noise is worce than the board using AD7657. Is AD7657-1 sensitive than AD7657 about noise?
Thanks & Regartds,
Would you be able to provide some more information about the system? I also agree with Harry in that it may be a layout problem. Are there any traces running close to the analog input of channel one that may be coupling noise onto the input? Can you provide any more details of the layout and the decoupling capacitors being used in the system?
How much worse is the noise being seen on the AD7657-1 compared to the original system using the AD7657? The typical SNR seen for the AD7657-1 is in the same region as that specified in the AD7657 datasheet.
Looking from your graph, it looks like both Ch1 of AD7657 and AD7657-1 are noisy, but AD7657-1 is much noisier. If there are issues with the Vcc lines, this should affect other channel, but for this one it affects the Channel 1. Can you share the schematic that you are using? I am attaching the link of the evaluation board for AD76567-1 http://www.analog.com/media/en/technical-documentation/evaluation-documentation/EVAL-AD7656_7_8-1-SCH.pdf. You may use this as your reference for your board.
Yes, I also think it is layput proble. But we confirmed it was different noise level between AD7657 and AD7657-1. Could you please look at the attached file? They are output wave form of AD7657 and AD7657-1. We can see CH1 of AD7657-1 is noisier than AD7657. They measured same board. Different points are only decoupling capacitor for Vcc lines. 1uF + 10uF are for AD7657. 1uF is for AD7657-1. 1uF is recommended capacitor value. Howeve, we got different result about the noise level. Therefore, our customer concerned that AD7657-1 is sensitive than AD7657 about layout. So, we would like to know that AD7657-1 is sensitive than AD7657 about layout or not. If it is correct, I'll recommend to the customer to use AD7657 instead of AD7657-1. The customer has no time for improve the layout of the board.
Can you send me a gerber data of evaluation board of AD7657 and AD7657-1? The customer would like to use the gerber data as a reference.
And, when we use two or three ADC on the board, how ahold we connet analog GND and digital GND? Should we connect analog GND and digital GND of ADC only one point on the board or connect analog GND and digital GND of ADC by each ADC around the ADC?
Hi Harry-san and Jonathan-san,
Yes, AD7657-1 is much noisier than AD7657. In case of AD7657, our customer can accept the noise level, but can not accept the noise level of AD7657-1. Our customer would like to know the reason of the difference. Is it individual difference (like a dispersion) of the product? Or is it difference between AD7657 and AD7657-1? Of course, I understand the root couse of CH1 noise is layout. But I have seen AD7657-1 is noiser tha AD7657 on the same board.
Thank you for advising the link of Eva board schematics. But I can not attached customer's schmatics on this site.
Can you provide the layout of the Eva bopard or gerber data for reference?
And, do you have any guide line or disuments about grounding for multiple using ADC/DAC on a board?