BrunoBronosky
New member
This servo came with my Adapt-a-Damper kit. I was able to fix it. Smoke on!
Well there's your problem! - No damper control
- Thread starter BrunoBronosky
- Start date
Bryan Mayland
TVWBB Hall of Fame
Oh wow, nice find there! What sort of symptom were you experiencing that lead you to investigating? I test all the servos with a servo tester when I get them and I will say that some of them sound different than others, and I have had a dead one too.
BrunoBronosky
New member
I never saw the servo move. I tested my assembly with an SG-50 servo I had and worked, but the servo wouldn't work for the Adapt-a-Damper. So I then deconstructed your SG-90 just out of curiosity. I tried to run the motor by connecting the 5v leads to the motor leads. I still got nothing. I noticed the bridge, but was not sure it was a mistake without a schematic. I couldn't find my solder sucker or solder wick. So I decided to heat the motor leads and remove the motor. It then worked with the 5v leads. I melted the bridge and slammed it on my table. Cleaned it up perfectly first try. Reassembled everything and it worked a treat.
BrunoBronosky
New member
I see you refunded me $5. Thanks! I wasn’t going to complain either way. I just wanted to share the solution in case anyone else ever faces a similar issue.
Bryan Mayland
TVWBB Hall of Fame
Yeah I gave you a partial refund because it is just supposed to work out of the box!
However, keen to understand how the servo worked for me before I boxed it up and then did not for you, I took one of the servos apart today to examine the layout and see what that bridge would affect. Unfortunately, that is not the source of whatever problem you're having. Those two bits are already connected by a trace on the PCB. That capacitor is connected across the two legs of the black SOT-23 component, which is a 3.3V regulator and feeds the controller IC as well as the position-reporting potentiometer. My examination didn't lead to any new information about why yours failed in transit, but at least I know now that I don't have to try to look through the case checking for that possible solder blob!
However, keen to understand how the servo worked for me before I boxed it up and then did not for you, I took one of the servos apart today to examine the layout and see what that bridge would affect. Unfortunately, that is not the source of whatever problem you're having. Those two bits are already connected by a trace on the PCB. That capacitor is connected across the two legs of the black SOT-23 component, which is a 3.3V regulator and feeds the controller IC as well as the position-reporting potentiometer. My examination didn't lead to any new information about why yours failed in transit, but at least I know now that I don't have to try to look through the case checking for that possible solder blob!
BrunoBronosky
New member
Good to know! It’s funny that just yesterday I had an issue with my daughter’s bubble machine where the fan motor stopped working and after disassembling it, turning the fan by hand, then powering on, it started working again. That’s 2 similar failures in as many weeks. Maybe it’s me.
I thought it was really weird that I couldn’t get the servo motor to turn by hand even after I had the gearing removed. It wasn’t until I had unsoldered it that I had the nerve to turn it with a pair of pliers. At that point I was convinced that it was destined for the rubbish bin.
Even though I never saw it move. I thought that I heard it try from the other room. I initially only used the micro-USB to power the RPi and do the initial setup. I donated my cache of wallwarts when I moved a few months ago (from Atlanta to Chicago). I eventually decided to use the one from my WiFi repeater and then switched to manual fan mode and let it run for a few hours. The servo never budged and I figured I had to enable something in the web interface that I would Google for later. Every now and then I thought I heard the servo but when I went in the room it was unmoved. I’m going to assume that I probably assembled it (or cycled it by hand) in a way that bound the gears and caused it to damage the motor when it tried to run at first.
I’m going to replace it anyway because once it closes it seems to need help moving to until 30-40%.
I did my first overnight smoke of a tiny boneless pork shoulder last night. It worked amazingly well. I’m so excited about the possibilities this opens up. Thank you for your years of investment and generosity (making this Open Source + Open Hardware). This is my favorite project ever.
I thought it was really weird that I couldn’t get the servo motor to turn by hand even after I had the gearing removed. It wasn’t until I had unsoldered it that I had the nerve to turn it with a pair of pliers. At that point I was convinced that it was destined for the rubbish bin.
Even though I never saw it move. I thought that I heard it try from the other room. I initially only used the micro-USB to power the RPi and do the initial setup. I donated my cache of wallwarts when I moved a few months ago (from Atlanta to Chicago). I eventually decided to use the one from my WiFi repeater and then switched to manual fan mode and let it run for a few hours. The servo never budged and I figured I had to enable something in the web interface that I would Google for later. Every now and then I thought I heard the servo but when I went in the room it was unmoved. I’m going to assume that I probably assembled it (or cycled it by hand) in a way that bound the gears and caused it to damage the motor when it tried to run at first.
I’m going to replace it anyway because once it closes it seems to need help moving to until 30-40%.
I did my first overnight smoke of a tiny boneless pork shoulder last night. It worked amazingly well. I’m so excited about the possibilities this opens up. Thank you for your years of investment and generosity (making this Open Source + Open Hardware). This is my favorite project ever.
Bryan Mayland
TVWBB Hall of Fame
I'm happy to hear it worked well! I appreciate the extra information about the servo operation. Because the servos are cheap I'm trying to come up with a better way of validating they work properly before I send them out. There's a lot of variation in the noise they make when they move and I am betting that there's a good noise and a not as good noise and if I can figure out which is which, I might be able to weed out troublemakers before they go out. I've only had two that are DoA out of probably over a hundred servos (I also use them in model planes and other small microcontroller projects) but there's been a couple now that failed on installation. Usually, if they work they stay working until they burn out so I was hopeful a simple DoA check would get more reliable results.
Thanks for the long explanation though, all information helps. Also make sure the mating faces of the dial top and the body are really clean. When they come off the 3D printer, there's some out of like a stickiness to the plastic that's not enough to really feel, but enough that when they rub against each other there's a lot of friction. Polishing the mating faces with a dry soft dishtowel for a minute or two usually removes whatever this is. There might be just enough stick on yours that it binds the servo until the error is enough to make it move.
Thanks for the long explanation though, all information helps. Also make sure the mating faces of the dial top and the body are really clean. When they come off the 3D printer, there's some out of like a stickiness to the plastic that's not enough to really feel, but enough that when they rub against each other there's a lot of friction. Polishing the mating faces with a dry soft dishtowel for a minute or two usually removes whatever this is. There might be just enough stick on yours that it binds the servo until the error is enough to make it move.
Larry Naylor
TVWBB Member
With the servos that are being used i don't think there is a reliable way to test them. Part of the problem is we are willing to put up with cheap crap from China.Back in the late 60s we had the same problem with cheap crap from Japan, back then we quit buying it and forced Japan to improve there quality control and quality of product. I fly large rc planes and if a product caused a few people to plant a plane that products reputation as being bad would spread like a wild fire through the rc hobby. A case in point was a radio made by Kraft great product till they put out a lemon and lied about not having problems with it but will fix it for free and the fix didn't work . 2 years out of business. I used to give my iffy servos to my sons for their rc cars. Sometime a $100 servo in a $50 car.When i bought mine for my RD3 i spent $15 for 4 servos, figured at least one should work, and with how we are using them they're not being too stressed.I see your point on trying to find reliable ones cause your supplying them
BrunoBronosky
New member
I replaced the servo with a TowerPro SG92R digital servo. I also polished the mating surfaces with toothpaste (one of my Dad’s old Navy tricks). It’s working fine now. Next task is to find a higher output blower that will fit in the case. Then I’m going to do a full write up.
Bryan Mayland
TVWBB Hall of Fame
Did that servo fit in the Damper body? I bought some to try out and they're slightly different dimensions. Everything is the same except the overall height (the SG92R is a few mm shorter) and the size of the large nub on top is just a little bigger diameter (~0.2mm) on the SG92R. I get around 11.66mm diameter on the SG92R and 11.48mm on the SG90. MG90 are larger than that at around 12.5mm if I remember correctly. The motors inside the SG90 and SG92R are identical in dimensions.
It does seem to have a little more torque, but the insides are even more janky. The one I opened had the stripped motor wire so close to the metal motor housing that I could see it making contact if there was enough of a bump. It looks more like the older SG90 before they started integrating the motor through the PCB. I don't think these are digital at all, despite what it says on the package. It is still a 50Hz bang-bang drive of the motor. I'm guessing the chip is just a slightly newer version of the same controller that has a little more efficient motor driver fets integrated so they can push out a little more power.
BrunoBronosky
New member
I did notice that the body is slightly shorter, but I didn’t any difference in the physical interface. I’ll take calipers to it and let you know.
This is the one I ordered.
TowerPro SG92R Micro Servo - 2 pack https://www.amazon.com/dp/B01CX63AOQ/?tag=TVWB-20
This is the one I ordered.
TowerPro SG92R Micro Servo - 2 pack https://www.amazon.com/dp/B01CX63AOQ/?tag=TVWB-20
Bob Walters
TVWBB Member
I made this point before, but it might be worth repeating. I'm an RC plane guy with a huge box of servos. Surely, I thought, one of my servos would fit the damper components I printed. As it turns out, the housing is designed to seal and operate properly only if the servo dimensions match the recommended servo. As Bryan points out, servos which appear very similar to the naked eye are not exactly the same shape. I discovered that none of the various servos I had on hand would fit correctly and that came as quite a surprise to me.
My solution was to re-design the damper housing to match one of my servos and print that component as a custom fit for MY servo. If you don't have that capability, it might be best to use the EXACT same recommended servo.
My solution was to re-design the damper housing to match one of my servos and print that component as a custom fit for MY servo. If you don't have that capability, it might be best to use the EXACT same recommended servo.
Bryan Mayland
TVWBB Hall of Fame
Oh those look like actual legit TowerPros. I got a pack of 5x knockoffs so mine might be slightly different. It fit into the Adapt-A-Damper slot, but only after some trimming and an unreasonable amount of force to insert it. Like Bob suggests, I am designing a new top to fit it exactly and put it out on the grill for some testing.
I'd like to find better servos but sourcing them in bulk reliably is a real pain in the butt. HobbyKing's HXT900 seems to be pretty good, but they are out of stock most of the time, along with many of their Turnigy models both analog and digital. TowerPros are good but finding real ones, in decent quantity, for reasonable prices, with the ability to get them in a week of shipping time and not a month is pretty tough. That's why I've been sticking with the SG90, because they're available everywhere around the world for cheap and quick. Their reliability may be questionable even clones can last years and replacement parts are easy to come by if they fail early. Still, it would be nice to not have them fail early.