# Bad orientation after blend

**URL:** <https://forum.universal-robots.com/t/bad-orientation-after-blend/5176>\
**Category:** Technical Questions\
**Tags:** bugs\
**Created:** [September 24, 2019, 7:57am UTC](https://forum.universal-robots.com/t/bad-orientation-after-blend/5176 "2019-09-24T07:57:41Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![damian.cieslak](https://avatars.discourse-cdn.com/v4/letter/d/7ea924/32.png) [@damian.cieslak](https://forum.universal-robots.com/u/damian.cieslak)\
**Post date:** [September 24, 2019, 7:57am UTC](https://forum.universal-robots.com/t/bad-orientation-after-blend/5176/1 "2019-09-24T07:57:41Z")

</div>

**Software version** : 3.7 - 3.11  
**Robot generation** : CB 3.1

Single access to the point

 ![Test_image_single_endpoint](https://us1.discourse-cdn.com/flex020/uploads/universal_robots/original/2X/a/a169537b1ae4faeb54c8d4cb591ea31b3881be4a.jpeg)

Double access to the point

 ![Test_image](https://us1.discourse-cdn.com/flex020/uploads/universal_robots/original/2X/5/5b474ca51089bfe2e9efd3d97a25cee95a58ead0.jpeg)

**Steps to reproduce** :

First waypoint - stop at this point  
Second waypoint - blend  
Third waypoint - stop at this point - non exact rotation  
Fourth waypoint - stop at this point - exact rotation

Third and fourth waypoint are the same.

**Expected behavior** :

If We have “stop at this point” command we should reach exact point with exact rotation

**Actual behavior** :

At third waypoint we reach exact position but we have a little bit different and only after reaching the same point again we reach exact orientation.

---

<div class="post-metadata">

**Author:** ![mmi](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.universal-robots.com/mmi/32/691_2.png) [@mmi](https://forum.universal-robots.com/u/mmi)\
**Post date:** [September 24, 2019, 1:14pm UTC](https://forum.universal-robots.com/t/bad-orientation-after-blend/5176/2 "2019-09-24T13:14:10Z")

</div>

Hi Damian,  
This bug was fixed together with blending algorithm changes for eSeries. Unfortunately this was breaking change, and could not be ported back to G3.  
Workaround is to have the same waypoint repeated twice at the end of move sequence. This compensates for rotation over or undershoot.

The same problem is mentioned in this post:  
[https://forum.universal-robots.com/t/movep-not-reaching-target-tcp-orientation/1217](https://forum.universal-robots.com/t/movep-not-reaching-target-tcp-orientation/1217)