OOAK Table Tennis Forum

I might be able to help with this. Can you explain further what the "clang" is? I played your video, but can't really tell what that sound is. If the side/vertical stepper motors get out of step, that would cause errors in the vertical or side-to side ball placement. This happens more frequently with the horizontal position than the vertical position. If the horizontal calibration is off, you can heaer a ringing sound when the stepper hits the side of the head housing. But that doesn't sound like the sound in your video - is it?

There isn't any calibration for ball speed though.

Can anyone who does the Falkenberg on their Amicus please share their numbers? I've actually never done the Falkenberg, and I'd like to add it to my practice regimen.

Stevieboy's excellent drills spreadsheet above does contain a "Falkenberg" drill, but he has 4 balls in the drill, whereas the Falkenberg traditionally only has one.

Also - searching online I found quite a few descriptions of the Falkenberg, which describes it as two balls to the backhand side (of which you return the second one with your forehand), followed by a ball wide to your forehand side. But all the youtube videos I looked at show people taking the "wide to the forehand side" ball more in the middle of the table or to the short forehand. Does anyone actually do it wide to the forehand? It seems like that would make it a lot more challenging - trying to get over to that forehand ball.

Hi,
Regarding the falkenberg

The original version is as you say 2 balls to the backhand and one to the forehand, which you play one backhand , pivot for a forehand from the backhand side then play the forehand from the forehand side and repeat.

The version I did inserted a ball to the middle before the forehand, if you program the drill like this and save it, you have the option of playing the original by simply calling up the program and holding down the ball 3 button till it goes out, this deletes that ball so the drill, dont resave the drill tho

There is also the Lazy falkenberg which requires a human feeder, where the feeder keeps playing all the balls to the backhand untill you are ready to pivot then they feed to forehand

Also you probably want to modify the speed and spin setting to match you and your robot

I figured I should update. I finally sent my amicus pro back to Butterfly for them to take a look. They ended up sending me a new one and thank goodness because I am not having any trouble with balls flying off the table on the pre-programmed drills now. Definitely a flaw in the robot I originally purchased. If you have this symptom too then you might want to have yours looked at as well. It was expensive to mail back, but all is well for now. Enjoying the amicus much more.

I'm trying to work on service returns. Has anyone had any success setting up good service return drills? I'm especially trying to work on how to return my friends nasty sidespin pendulum serve. He puts so much sidespin on it that it's hard to keep my return from missing off the side of the table. I'm having a hard time setting up short spinny serves.

I am a new owner of the Amicus Butterfly Pro and am hoping for some help from other owners if at all possible.

I have set up in the memory of the device a series of very basic routines just so that the same ball can be delivered to the same place. I am having a problem though - the routines which involve top spin are working fine. The ball can be delivered to backhand or to the forehand and the routine keeps on going as it should.

For some other routines with no spin the device will send out a ball correctly 11 times and then the E2 error message will appear. However nothing is actually stuck or broken and if you either hit the start button or turn the ball frequency down to 0 and then up again the device starts working again but only for 11 balls. Obviously I don't want to have to restart every 11 balls. Has anyone got any ideas as to what I could try to fix this?

Many thanks

PS - I meant to post this message in this thread but accidentally started a new one under the general robot category -happy for the other message to be deleted as this one is now here!

Yikes, that's bizarre! And it's always after exactly 11 balls? And only on certain routines? Do those routines have anything in common? How many balls are in the "bad" routines?

There is a way to set up drills where it pauses after a while, I'm wondering if that has something to do with it. I've never used that feature.

In any case, I'm happy to help, but I'm not sure I'm going to be able to figure this one out. You might want to just try contacting Butterfly support.

Did you solve this? I have had modest success with this.

_________________
Cobra Kai TT Exponent (Mercy effs up your Game)
One-Loop Man: One Loop... Again????
Lumberjack TT Exponent

"We don't rise to the level of our expectations, we fall to the level of our training" - Archilochus

No, haven't looked at it. I was hoping that someone had already done some drills that I could use.

Thanks for the reply Dave - yes it is after exactly 11 balls. All of the routines are simply placing the ball in the same place with the same spin. Just a one ball routine to be repeated hopefully more often than 11 times! I will check out next week to see if I have somehow got the pause feature included but I am not optimistic about that. I wouldn't have thought the E2 error message would come up if it was a planned pause but it is definitely worth trying to rule out. Looks like I might have to contact Butterfly support on this one!

Thanks Dave in any case - John

Dave, for short serves, make the trajectory as low as possible. The height of the feeder tube can vary depending on how low you want the serve to be. The speed I use is between 5 and seven. Spin is about -2 to -4 and the sidespin varies. Hope that helps. The serve is really short tho.

_________________
Cobra Kai TT Exponent (Mercy effs up your Game)
One-Loop Man: One Loop... Again????
Lumberjack TT Exponent

"We don't rise to the level of our expectations, we fall to the level of our training" - Archilochus

Hi everyone, I've just joined the forum! I have had two sessions with my new Amicus Pro.

Re the Sample button, the Engrish in one place in the manual says, "To press the sample button simulates the set of them over, by a single bale ejection" [all sic]. I too initially thought this meant it was only going to throw one ball, but I think what it means is that it throws only one type of ball, rather than starting to cycle through the various types of balls programmed for the exercise, as the Start/Stop button would do. And indeed, I see now that on the previous page, the one with a drawing of the control panel, the legend for the Sample button says "Simulates a ball delivery without saving it (press once to start, twice to stop)".

A few things I've noticed --

() As others have pointed out, it's too bad Butterfly chose not to use connectors with screws at each end to hold them on. At one point yesterday the connector to the control unit got partially pulled out, causing what seemed to be an unfixable ball jam. Once I noticed it and plugged it back in, the robot was fine.

() I set the robot up to give me a low backspin serve, and then tried to add sidespin to it. Even with the sidespin control set to 75 degrees, I didn't notice much sidespin on the ball. I tried side-topspin, and that worked fine. Has anyone else had trouble with side-backspin?

() At one point I tried turning on both the random-ball-type and random-placement functions. I had two balls programmed, one to the forehand and one to the backhand. Maybe one out of six times the robot selected the backhand ball, it came out at head height, nowhere near the table. This didn't happen on the forehand side (I'm right-handed). Has anyone else seen this?

That's interesting about the sidespin. It must be because the sidespin is controlled by the top wheels, which induce backspin. Did you try to specify more sidespin? You might need to decrease the backspin, in order to get the amount of sidespin you want.

The problem with random balls sounds very strange. Maybe you had the connector only partially connected again?

I haven't had any trouble with side backspin serves.

_________________
Cobra Kai TT Exponent (Mercy effs up your Game)
One-Loop Man: One Loop... Again????
Lumberjack TT Exponent

"We don't rise to the level of our expectations, we fall to the level of our training" - Archilochus

Okay, I've learned a couple of things since I last posted.

First, the "PLACE RND" function does something complicated that is not well documented at all. The manual that came with the robot says


Oddly, there's a different manual linked from the Megaspin product page: https://www.megaspin.net/download/butte ... ual-en.pdf

It says a little bit more:


When I first read this, I assumed that it would simply select the programmed balls in random order. That, however, turns out to be the function selected by "TYPE RND". "PLACE RND" is doing something more complicated; I don't know exactly what, but it seems to be something like randomly selecting individual parameters of the programmed balls and combining them. Thus, if you have programmed a high slow ball and a low fast ball, turning on "PLACE RND" will sometimes generate a high, fast ball or a low, slow one. The high, fast balls I was seeing at one point -- that came out at head level, nowhere close to the table -- appear to have been generated this way.

Anyway, I've stopped using "PLACE RND" and now use "TYPE RND" instead, and that's behaving as I expect: for each ball it just selects one of the balls I have programmed. (I think "PLACE RND" really deserves another name altogether, like "scramble mode" or something.) No more wild balls. If anyone has a better theory about what "PLACE RND" is doing, I'd love to hear it

Now for the side-backspin thing. I did an experiment: I set the robot to produce a fairly soft backspin ball. I then stood by the side of the table and volleyed the balls above the net, holding the bat vertically, and watching what direction the balls went after they bounced off the rubber. With pure backspin, of course, they went straight down. I then adjusted the sidespin. Even at full "90 degree" sidespin, I was still seeing them come off the rubber at only about 45 degrees from straight down. I tried the same experiment with topspin, and this time, with the sidespin setting at 90 degrees, they came off the rubber almost horizontally.

So it seems to me that there is a range of side-backspin, with more sidespin than backspin, that the robot either can't generate physically (maybe it would take a fourth throw wheel?) or is programmed not to for some reason.

If someone else wants to try the same experiment, I would be very interested to hear the results. I don't have the feeling that this is a mechanical problem with my unit; its behavior seems very consistent. It could be a recent firmware change, though, I suppose.