TDA4VH-Q1: TDA4VH 19.2MHZ crystal test results discussion

jingle.lan

Part Number: TDA4VH-Q1
Other Parts Discussed in Thread: TDA4VH

Tool/software:

Hi all

I would like to discuss an interesting phenomenon with TI hardware engineers.We tested the TDA4VH 19.2MHZ low-speed crystal waveform and found that the test results using active and passive probes were different, and the difference was very large. We are not sure which test result is closer to the actual result. Secondly, we asked TI to determine whether the waveform exceeds the TDA4VH crystal input voltage.

First, the circuit under test is shown in the figure. We connect the oscilloscope probe to both ends of C1005 to test the waveform of the crystal input to TDA4VH. The information of the two oscilloscope probes is as follows:

1. The active probe uses Tektronix's TAP2500, with a bandwidth of 2.5GHZ, input capacitance <0.8pF, probe input resistance of 40KΩ, and oscilloscope input impedance of 50Ω
2. The passive probe uses Tektronix's TPP1000, with an input impedance of 10MΩ, input capacitance <4pF, and oscilloscope impedance of 10MΩ.
The test results of the two probes are as follows.

This is the test result of the active probe.

This is the test result of a passive probe.

We can see that the negative voltage difference of channel 1 is very large, the active probe measures -600mV, and the passive probe measures -220mV.

We can see that the negative voltage difference of channel 1 is very large, the active probe tests -600mV, and the passive probe tests -220mV.
We suspect that the difference is due to the input capacitance and input resistance of the probe, but the input resistance of the active probe is 40KΩ, which is equivalent to a 40KΩ resistor in parallel on C1005. The capacitance of the passive probe is only 4pF larger, but it should not cause abnormal crystal output waveform.
Please ask TI colleagues to help us confirm which test is closer to the actual situation, and secondly, whether the -500mV voltage of the active probe can meet the requirements of the TDA4VH crystal. The manual does not specify the minimum voltage value of Vmin.

In addition, this crystal has passed the matching report of the crystal manufacturer, and its drive power is within the range.

over 1 year ago

0 jingle.lan over 1 year ago

Prodigy 120 points

Sorry forgot to add that we tested TI's EVM board and the test also showed the same waveform, so we are more sure that this is a test problem rather than a hardware design problem. If you are interested, you can also try it.

0 Kevin Lavery over 1 year ago in reply to jingle.lan

TI__Mastermind 19960 points

Jingle,

I agree that the result is a test artifact. The addition of even 1pF to a 12pF load capacitor is significant.

Kevin

0 jingle.lan over 1 year ago in reply to Kevin Lavery

Prodigy 120 points

Hi Kevin

Our test lead tested the left side of C1005, and the right side is GND, which is equivalent to the oscilloscope test lead being connected in parallel to this capacitor. For the low-impedance test lead, it can be considered that the test lead short-circuited this capacitor and connected a 40K resistor in parallel here. I think this will change the structure of the Pierce oscillation circuit. However, using an oscilloscope test lead with a large input capacitance may only affect the crystal output frequency, and should not have such a big impact on the change in amplitude. We tested the impact of the capacitor. If the capacitance is changed from 8pF to 10pF, there is only a small change in the amplitude, and the low voltage of the crystal will not change to negative hundreds of mV.