original MG90D servo and strange problems

[elto...@gmail.com]

Hello everybody. After solving the buzzer problem (https://groups.google.com/g/vorpal-robotics-forum/c/oob5UXF_WIo), everything worked fine.

However, some servo motors (which were fake) started to break. So I bought other, they are marked as Original TowerPro MG90D and appear to be original (
https://a.aliexpress.com/_m0rfWFh).

After installing the 12 MG90D, strange problems started to happen, which can be seen in the videos. The problems happen in DEMO mode and in RC.

Another thing is that until 7 servos connected to the shield it has no problems. If I add 1 more servo, then the problems begin.

Any suggestions of what it can be and how to solve it? Thanks.

video 1

video 2

[Elton Torres]

some more information: if I put 7 MG90D and the other 5 with MG90S (fakes) everything works normally.

--
You received this message because you are subscribed to a topic in the Google Groups "Vorpal Robotics Forum" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/vorpal-robotics-forum/s_CyVMgQ_RE/unsubscribe.
To unsubscribe from this group and all its topics, send an email to vorpal-robotics-...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/vorpal-robotics-forum/b4ffb36b-3fd6-4d67-aa2c-439099371f63n%40googlegroups.com.

[vorpalrobotics]

This looks like a power issue. The MG90D might consume a bit more power than the fake ones. 

Also, note that if you're using digital servos you should change the robot code to use a higher frequency for the servo signals. Around line 130 you'll find this code:

int FreqMult = 1;   // PWM frequency multiplier, use 1 for analog servos and up to about 3 for digital.

                    // The recommended setting for digital is 2 (probably safe for all digital servos)

                    // A shunt between Nano D5 and D6 will set this to "1" in setup, this allows you

                    // to select digital servo mode (2) or analog servo mode (1) using a shunt 

                    // or short jumper wire.

If you set FreqMult to 2 you will likely have better luck with digital servos because they like running at higher frequency than analog. You can also put a shunt between D5 and D6 to enable this mode (shunt must be in place on boot). You could rig up some wires and route them to the cap area so you could plug and unplug this shunt at will and directly test the differnce without pulling the nano out. 

However, the symptoms shown in your video seem too severe for that to totally fix it. But it's worth a shot.

It's possible your batteries cannot output enough amps to handle the digital servos. Could you send me a picture of the battery label, being sure to include both sides of the battery?

[elto...@gmail.com]

Hello, i was aware of the need for a different code for the use of digital servos. However, on the page (https://vorpalrobotics.com/wiki/index.php/Tower_Pro_MG90S_Vs._Clones) it mentions "in our source files, it's the version of robot code with the word" DIGITAL "in the filename." So I looked on the project's github and dropbox and didn't find it.

Anyway, I changed the value of FreqMult to 2 and, unfortunately, the problem persists.

Regarding the batteries, I'm using these: https://www.aliexpress.com/item/4000465742173.html?spm=a2g0s.9042311.0.0.2742b90aW7Cwif

They seem to be of good quality.

Do you have any other suggestions? Thanks in advance.

[vorpalrobotics]

Oh, the code changed since that page was written, I will update the wiki.

As far as batteries from Aliexpress goes, that's probably your problem. Almost all of them are either counterfeit or very low quality. The amp hours listed on the nameplate is usually double or even triple what it really can output. I would never buy rechargeable batteries from Aliexpress or banggood or the like, you're just asking for trouble.

When I was researching batteries I tried several different ones from aliexpress and none of them worked. Testing revealed they were way, way lower in output amphours than advertised.  I even saw one 18650 that claimed to have a capacity of 10,000 milliamp hours. There's no LI-ON chemistry in existence that can come anywhere near that in the 18650 form factor. What they were claiming was physically impossible.

[elto...@gmail.com]

Thank you for the tips. I put 6 original MG90Ds together with 6 fake MG90S and everything worked fine.
I hope the originals make the fake ones last longer.

Thanks again.