HV to LV isolation barrier

Hi BP101,

I am assuming you are referring to ±200V digital inputs and not analog inputs.

I am not aware of any isolators accepting ±200V directly at their inputs without any voltages dividers. We do have ISO121x series devices that accept 60V inputs directly. Since the device supports input current regulation, there is no need of high wattage external resistors which makes the solution size really small and also makes the design thermally efficient. The device is also available in small SOIC/QSOP packages.

Please refer to the below Tech Note that describes how to use ISO121x with ±240V inputs.

Let us know if you have any questions, thank you.

Regards,
Koteshwar Rao

Hello Koteshwar,

Koteshwar Rao said:

I am assuming you are referring to ±200V digital inputs and not analog inputs.

Analog HVDC inputs buffer prior to low voltage ADC inputs. Recall seeing Rohm marketing of similar analog input isolator for HEV automotive but could not source it. I'll check out the ISO121x PDF, Thank you!

Something like in this TI link but for isolating analog signals into ADC channel inputs.

www.ti.com/.../ISOW7840

Rohm makes amp with internal Vref, can be unity fowler (LMR1801xxx-LB) and recall seeing analog amp isolator targeting HEV market.

Hi BP101,

Thanks for your inputs and clarification.
I will notify my colleagues who support isolators that accept analog inputs. You will hear back from them, thanks.

Regards,
Koteshwar Rao

Hi Alexander,

Very nice differential isolation buffer but is not 3v3 supply or analog output 3v3 ADC. Interested in a single ended analog isolate for 3v3 ADC embedded MCU as many are these days. The price of ISO224 is little less than the Rohm analog buffer I mention above $7.42 single. The TI Tiva C series, Newer Piccolo 200MHz dual core MCU class are 3v3 supply embedded ADC.

Thanks again.

HI Alex,

I have always used single end ADC channels to sample linear analog signals that steady state above 0v, <3v3. Otherwise use faster amps in the MHz band width range to acquisition faster slew wave forms seems prudent. It seems fast slew periodic signals mandate differential ADC inputs being more typical of late at TI. Using single ended channels to acquisition static amplitudes via low band width amps around fast periodic wave forms and the AD conversions are compressed producing inaccurate digital measures above a certain amplitude. That's a gotcha not much talked about.

Perhaps such warning should be more exaggerated in TI datasheets for beginners in analog acquisition. I have read numerous TI analog Wiki documents, none of them seem to specifically mention the benefits or quagmires of either channel acquisition method as it relate to amplifier band width mandates.

AMC 1301 ADC connection technical reference mentions VoutP can directly input single ended channels. Only if finer granularity is required recommended the final stage conversion 0.5v-4.5v. Yet the output centers around 1.44v VDD=3v3? As long as AMC1200 blocks down stream transients it might work too. The TLV6001 CMM 1.25v seems to infer AMC1301 output peaks well below VDD 3v3 rail.

Odd part it states MPS430 ADC internal reference 2.5v matches AMC1301 Out-P levels 0.415v-2.465v. Suffice to say a bit low for 3v3 full scale 12bit ADC. If the AMC1301/1200 VDD is 3v3 will it output just below VDD2 rail? Need any AMC datasheet to confirm!     

It seems the AMC1200 might be a good fit, always difficult to get from forum datasheets since searches only produce post links 

Hi Alexander,

Hope you had a wonderful holiday..

Alexander Smith said:

We have a lot of collateral to help beginners, such as circuit cookbooks, precision labs, application notes, etc

I have several of the analog engineer, cook books guides from TI. Don't recall differential ADC being any better for signal acquisition over single channels. Topic relates signal type (AC/DC) being the determining factor for selecting either one, nothing was mentioned about SNR being less for differential channel or with DC input signals.

Alexander Smith said:

however most high performance customers prefer differential as noise in the system is subtracted out at the input of the differential ADC. 

Seemingly 2 traces must be in my mind kept to similar impedance and capacitance making PCB design more challenging. Might consider differential for future designs though often PCB space is somewhat limited.

Thanks again for the great information.