2025-02-06 12:07 AM
Hi All,
we tried to migrate from VL53L1CX to VL53L4CD for better precision in short range.
both use ULD driver, we expect they work in similar way however it seems not.
we set both to TB20 and compare power consumption in single shot case.
It seems VL53L4CD takes much long time and power than VL53L1CX even TB is same.
Is the observation true and why ?
2025-02-06 8:13 AM
The evolution goes like this. We invented the VL53L1 and were very proud of it. But a cell-phone customer wanted it cheaper, so we removed the lens on the top of the L1 Receive side and the extremal filters, and called it the VL53L3. (I have no idea why it wasn't an L2).
But then we thought we could improve the L3, by moving to a single-stage, more-narrow Field of View VCSEL (laser). And the L4 was born.
Notice we didn't change much in the way of operation. It's only the optical elements.
What I see in your graphs is that it's not the max power that's changing, but it's the duration of the range that is different. And you can control that by calling the TimingBudget function.
This setting adjusts the amount of time dedicated to each range. Longer means better accuracy, better detection, but as you can see, more power consumption. Reducing this setting will save power. You can match that of your L1 and I expect you will get the same accuracy and detection you were getting with the L1. The new VCSEL should give you an advantage, but it's hard to beat the L1's optics.
I expect your results will be nearly identical, but at a lower cost.
- john
2025-02-06 11:07 PM
However from my testing, under same TB (time budge) setting VL53L4CD takes much longer time than VL53L1CX so it consume much more power.
Usually VL53L1CX ranging time follows TB setting with small overhead however VL53L4CD takes 1.5 times of TB for total ranging time.
If VL53L4CD hardware is similar to VL53L1CX, is it possible the time difference happens on ULD driver ?
Due to the power consumption difference it's difficult for us to use VL53L4CD to replace VL53L1CX.