Close

Rev3 - Getting closer to production

A project log for Open Source LiDAR - Unruly

We've been waiting a long time for someone to post the plans of a working LiDAR. Is it really going to happen?

Laser DeveloperLaser Developer 02/22/2019 at 07:324 Comments

DESIGN

Discussions

iliasam wrote 04/06/2019 at 12:35 point

Hello.
How is your project going?

  Are you sure? yes | no

Laser Developer wrote 04/07/2019 at 07:13 point

We're working through the financing of the tooling and production at the moment and hope to have a deal in the next few weeks. If everything works out as we hope then we can start moving forwards with a production version of the Unruly :).

  Are you sure? yes | no

Laser Developer wrote 02/22/2019 at 12:35 point

Thanks for the feedback iliasam and for spotting the typo on the MAX3658.

The main differences between the MAX3658 and the MAX40658 are:

The 3658 is a general purpose TIA with no particular features that suit the LiDAR market whereas the 40658 was designed with the LiDAR market in mind.

The 40658 can tolerate high overdrive signals without phase reversal. At least it's supposed to. My testing shows that this isn't actually true and under the wrong conditions it produces false return signals when there is a strong primary target. This is a major problem with the 3658 that the 40658 was supposed to eliminate.

The 3658 is far more sensitive to static during handling because there is no internal protection circuitry on the input pin. This has been fixed successfully in the 40658.

The 3658 has internal DC offset control while the 40658 requires external circuitry to correct for DC signals. These signals are caused by background light and degrade the performance of any TIA. The external method of the 40658 gives you greater dynamic control of background light than the internal system of the 3658 but..... The 40658 has a horrible problem in that the offset control feeds directly into the input of the TIA. This means that even the smallest amount of noise on the control line is amplified by the TIA and can easily swamp the return signal. I have tested a number of different circuit layouts to try and correct this problem but so far none have been very effective. Fortunately the Unruly doesn't need to have particularly good DC offset control so I have disabled the feature when the 40658 is used.

For comparison, the LTC6560 has the best signal overload handling capability, but even this fails when hit by a return signal from a reflective target and it has no DC offset control at all. In fact there are no TIAs on the market today that are perfectly suited to LiDARs.

  Are you sure? yes | no

iliasam wrote 02/22/2019 at 13:18 point

Thank you for the detailed answer!

  Are you sure? yes | no