TUSS4470: Cinp recalculation and effects on the Vout pedestal level

Paolo Lazzeri

Part Number: TUSS4470

Hello,

I'm using the TUSS4470 in the pre-driver mode with a transformer.

Since the peak voltage on the secondary circuit reaches 1400 V I recalculated the capacitance on the INP pin using the formula found on the forum:

INP_Impedance = R_INP + [ 1 / ( 6.28 * f_burst * C_INP ) ]

To have a current of 10 mA with a burst frequency of 235 KHz it is necessary to reduce the capacitance from 330pF (value indicated in the datasheet) to 47pF.

Therefore I measured the Vout level with LNA gain = 20 V/V, first and last logamp stages both enabled; by reducing Cinp from 330pF to 47pF the Vout pedestal level increases from 390 mV to 500 mV. Since I need to detect very small signals, losing 110 mV is quite annoying.

Is it possible to compensate for this effect in some way?

Am I using an excessively low Cinp value?

By the way, Cinn was set to 18.1 nF by applying the formula reported in the datasheet.

Thanks,

Paolo

9 months ago

0 Eddie LaCost 9 months ago

TI__Mastermind 39130 points

Hi Paolo,

Thanks for your inquiry. I am looking into this and expect to have feedback later today.

0 Paolo Lazzeri 9 months ago in reply to Eddie LaCost

Prodigy 140 points

Hi Eddie,

this afternoon I made further measurements which showed that even with 47pF the voltage limits on the INP input of the TUSS4470 were not respected:

according to the absolute maximum ratings the voltage on INP must be between 0.5V and 1.3V. From the measurements I made, the voltage on INP fluctuates between 0.27V and 1.52V.

At this point I measured the current flowing through Rinp (200 Ohm as indicated on the datasheet) and it is +/- 45 mA; looking on the forum there is talk of a max input current to INP of around 16 mA.

Based on these data I should further lower the Cinp value and I would not like this to lead to a further increase in the Vout pedestal level.

Thank you for your support.

Paolo

0 Eddie LaCost 9 months ago in reply to Paolo Lazzeri

TI__Mastermind 39130 points

Paolo,

Sorry, I don't have an update on this yet.

One question, is the secondary really at 1400V or is this a typo?

0 Paolo Lazzeri 9 months ago in reply to Eddie LaCost

Prodigy 140 points

Yes, the secondary is really at 1400V as show in the picture below:

Frequency is 235 KHz.

This is the screenshot of a 500 us pulse.

By the way, this morning I tried to put Cinp = 12 pF. Vout pedestal level rose further, reaching a value of 640 mV.

Let me know if you need further information.

Paolo

+1 Eddie LaCost 9 months ago in reply to Paolo Lazzeri

TI__Mastermind 39130 points

Paolo,

Sorry for the delay.

One thing you could do is use the intercept adjustment feature of the device. This is discussed on page 19 and 20 of the datasheet.

Also, you could consider disabling one of the log amps. In some cases, this can help clean up the signal.

Let me know if the intercept adjustment feature is able to reduce your pedestal.

0 Paolo Lazzeri 9 months ago in reply to Eddie LaCost

Prodigy 140 points

Eddie,

the intercept adjustment is able to reduce the pedestal level of Vout.

At this point I have to calculate the correct value of Cinp.

As mentioned on the forum I found this formula:

Vp/I_INP = INP_Impedance = R_INP + [ 1 / ( 6.28 * f_burst * C_INP ) ]

where Vp = peak voltage, I_INP = INP pin sink current.

The maximum value of the sink current of the INP pin is not reported in the datasheet. Can you confirm that the value is 15 mA?

0 Eddie LaCost 9 months ago in reply to Paolo Lazzeri

TI__Mastermind 39130 points

Hi Paolo,

Yes, the max sink current is 15mA. I assume you were looking at this thread with Akeem, which also shows the input impedance as a function of frequency.

https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/927077/tuss4470-wrong-absolute-maximum-ratings-on-vinp-pin/3431638

0 Paolo Lazzeri 9 months ago in reply to Eddie LaCost

Prodigy 140 points

Hi Eddie,

Thanks for the info on maximum sink current.

I've seen the input impedance graph, but I'm not clear how to use it to calculate the Cinp value or how it relates to the INP_Impedance formula. Can you give me some more details?

0 Eddie LaCost 9 months ago in reply to Paolo Lazzeri

TI__Mastermind 39130 points

Paolo,

You can use the input impedance with the formula below.

Vp/I_INP

Vp=700V

I_INP = 15mA max (you could use a lower value to have some margin)

700 / 15mA = 46666.66 ohms

You can then use the formula below to determine the correct capacitor value. I am calculating about 14pF, so you were pretty close with 12pF.

INP_Impedance = R_INP + [ 1 / ( 6.28 * f_burst * C_INP )

Sensors

Sensors forum

TUSS4470: Cinp recalculation and effects on the Vout pedestal level