Hitec 7954 Servo Failures
#26
My Feedback: (1)
Join Date: Feb 2002
Location: private, UNITED KINGDOM
Posts: 3,672
Likes: 0
Received 25 Likes
on
15 Posts
To move a control surface with a given amount of air pressure on it, it takes a certain amount of watts to move the surface. If the voltage is lower, due to whatever reason, more current is required. (Watts = Voltage x Current.) To hold that surface in place on a flap, it takes that amount of current until the surface is returned to neutral position.
That is not how things behave. A servo can not maintain a constant wattage, the wattage is whatever voltage is present and the current that can be driven by that voltage. The voltage drives the current, so as the voltage falls the current driven by it will also fall, and the total wattage at the servo will fall dramatically. A voltage sag means a current sag.
#27
My Feedback: (11)
Join Date: Sep 2010
Location: marina del rey, CA
Posts: 199
Likes: 0
Received 0 Likes
on
0 Posts
My understanding is that if the voltage is lower or drops the power of the servo suffers, if the servo was set to give a constant output torque then i'd get why the current would increase but since its not working to attain a goal like that and rather to use the available power to produce a response I think harry is correct, I think what you are seeing is the other way around, constant load on a servo that then produces excess heat and then the system voltage drops because it takes watts to generating that heat, maybe?!
#28
My Feedback: (2)
[QUOTE=Len Todd;11795013]To move a control surface with a given amount of air pressure on it, it takes a certain amount of watts to move the surface. If the voltage is lower, due to whatever reason, more current is required. (Watts = Voltage x Current.) To hold that surface in place on a flap, it takes that amount of current until the surface is returned to neutral position.
With analog servos, they each operate a different times. With digital servos, they can all operate at the same time. When they all operate at the same time (potentially), all the individual currents add up. When you constrain all that current on a typical Rxer buss, the buss is essentially a resistance and voltage on the output side of the Rxer will go down (voltage sag) resulting in an increased current draw to get the power needed to move all the surfaces. That is why there are power distribution systems out there. When Resistance goes up the Voltage goes down (Power = Current x Resistance.) You may not ever see a drop on the bench. But when the plane is moving, it is there. The faster it moves, the more prevalent it is. At some point, the Battery's capability may even begin to factor in. Long servo leads and maybe a little resistance in the connectors and now what is the voltage at the servo going to be?
When you throw these conditions together with several digital servos all servos operating off that tiny Rxer buss, in a high speed plane, you have a situation looming (i.e. Many or all the servos operating at the same time with high loads on the surfaces.) Now, you throw in a flap servo, at maybe high speed, and there is now a constant high current load placed on that Rxer buss and the battery, plus all the other control surface loads. The flap's constant load can be up around 5 Amps+. As the flap servo begins to heat, the resistance goes up and even more current is needed resulting in an even higher current draw. That's why the Ultra Premium servos have obvious heat sinks. They can handle the higher currents and resulting heat.
Bottom line: Slow down before dropping the flap. Make sure you have a servo that can handle the load and resulting heat for a long period of time. Set the servo's failsafes. Keep the battery near fully charged (as the Voltage goes down on it, the above negative effects are amplified.) When using several digital servos, consider a power distribution system. Remember, the flap servo, when down, is a constant load versus just momentary like most of the other servos. It has to be a pretty healthy servo capable of handling the load and the resulting heat for a long period of time.
----------------------------------------------------------------
Len,
I have a different understanding of some of the statements made above. The start time of a servo is not determined by whether it is analog or digital but by the receiver design and the software design for 2.4GIG radios. I know of one radio where all the servos start at the same time whether they are analog or digital. I don't like this because it puts a very high current drain for a few milliseconds when more than one servo starts at the same time.
I am surprised there are not more servo failures when I see servos being run above the rated voltage. I don't recall seeing any mention of the voltage used on the servos that are mentioned in the start of this thread. Servo electronics failure is caused by combination of too much voltage and current (total power) so if we only overvoltage but keep the current low we may get away with it. Flaps tend to keep the load(current) higher for longer periods of time so the possibility of failure goes up.
Sid
With analog servos, they each operate a different times. With digital servos, they can all operate at the same time. When they all operate at the same time (potentially), all the individual currents add up. When you constrain all that current on a typical Rxer buss, the buss is essentially a resistance and voltage on the output side of the Rxer will go down (voltage sag) resulting in an increased current draw to get the power needed to move all the surfaces. That is why there are power distribution systems out there. When Resistance goes up the Voltage goes down (Power = Current x Resistance.) You may not ever see a drop on the bench. But when the plane is moving, it is there. The faster it moves, the more prevalent it is. At some point, the Battery's capability may even begin to factor in. Long servo leads and maybe a little resistance in the connectors and now what is the voltage at the servo going to be?
When you throw these conditions together with several digital servos all servos operating off that tiny Rxer buss, in a high speed plane, you have a situation looming (i.e. Many or all the servos operating at the same time with high loads on the surfaces.) Now, you throw in a flap servo, at maybe high speed, and there is now a constant high current load placed on that Rxer buss and the battery, plus all the other control surface loads. The flap's constant load can be up around 5 Amps+. As the flap servo begins to heat, the resistance goes up and even more current is needed resulting in an even higher current draw. That's why the Ultra Premium servos have obvious heat sinks. They can handle the higher currents and resulting heat.
Bottom line: Slow down before dropping the flap. Make sure you have a servo that can handle the load and resulting heat for a long period of time. Set the servo's failsafes. Keep the battery near fully charged (as the Voltage goes down on it, the above negative effects are amplified.) When using several digital servos, consider a power distribution system. Remember, the flap servo, when down, is a constant load versus just momentary like most of the other servos. It has to be a pretty healthy servo capable of handling the load and the resulting heat for a long period of time.
----------------------------------------------------------------
Len,
I have a different understanding of some of the statements made above. The start time of a servo is not determined by whether it is analog or digital but by the receiver design and the software design for 2.4GIG radios. I know of one radio where all the servos start at the same time whether they are analog or digital. I don't like this because it puts a very high current drain for a few milliseconds when more than one servo starts at the same time.
I am surprised there are not more servo failures when I see servos being run above the rated voltage. I don't recall seeing any mention of the voltage used on the servos that are mentioned in the start of this thread. Servo electronics failure is caused by combination of too much voltage and current (total power) so if we only overvoltage but keep the current low we may get away with it. Flaps tend to keep the load(current) higher for longer periods of time so the possibility of failure goes up.
Sid
Last edited by sidgates; 05-01-2014 at 10:35 AM.
#29
My Feedback: (10)
Hi Steve,
Back in the day I avoided using early digitals on flaps. The flaps are usually big, and with a lot of bouncing on the ground and G forces in the air it puts a lot of load on the servo and the early digitals were not always up to the task. Lately I do use digital on everything. FWIW
Back in the day I avoided using early digitals on flaps. The flaps are usually big, and with a lot of bouncing on the ground and G forces in the air it puts a lot of load on the servo and the early digitals were not always up to the task. Lately I do use digital on everything. FWIW
#30
You seem to be saying that a servo is a constant wattage device, so that if the voltage falls the current will rise. In that case as voltage falls to near zero, the current will rise to near infinite. On the other hand according to you, if the voltage rises the current will fall!
That is not how things behave. A servo can not maintain a constant wattage, the wattage is whatever voltage is present and the current that can be driven by that voltage. The voltage drives the current, so as the voltage falls the current driven by it will also fall, and the total wattage at the servo will fall dramatically. A voltage sag means a current sag.
That is not how things behave. A servo can not maintain a constant wattage, the wattage is whatever voltage is present and the current that can be driven by that voltage. The voltage drives the current, so as the voltage falls the current driven by it will also fall, and the total wattage at the servo will fall dramatically. A voltage sag means a current sag.
A flap servo has a constant demand for power (load) when the flap is deployed. To keep that flap down, at the same speed, takes a constant amount of power. If you need 30 watts to keep the flap down at any given speed and then reduce the voltage, the current needs to increase to give you the correct power level needed to match the servo's Pot to the signal resolution. Thus when you reduce voltage, current demand goes up. P=IR.
For the other servos, they are momentary loads. But the same concepts apply, only momentarily. However, as digital servos move, the loads can be cumulative.The more servos moving simultaneously the higher the demand on the battery through the Rxer buss. At times, (e.g. when they all are moving) the load is higher than the power system can supply thus the voltage sags through the resistance caused by the Rxer buss, servo leads and connector, and the limits of the battery, etc.. Thus with reduced voltage it takes more current to accomplish the same work. Increase the current too much for too long and the unit's electronics will heat up and fail.
Shutting off the voltage is not a "sag." and yes when you shut off the voltage, there is no force to move the current. But on servos you are not shutting off the voltage. The voltage is always there. When the servo sees a variance between the resolution setting on the signal line it is receiving and the resistance across the pot in the servo, it uses current to move the servo's motor to match the pot to the resolution setting. When this takes place, the more voltage that is present, (withing the limits of the servos design) the less current used for the same amount of movement. A High Voltage movement is also faster as there is more force behind the current. High voltage servos are designed to take advantage of this variable.
I am not talking about shutting off the voltage and then the current cuts off. That never happens, until you shut off the power switch disconnect the battery or the servo's power circuit fails open, etc. When the system is on, the voltage is always present at the servo. Some folks even bypass the Rxer with the servo's Ground and VDC line Only the signal line comes from the Rer. That is how Smartfly eliminates the limiting effects of the Rxer buss.. Point being, the voltage is always present on the servo. If you don't want the servo's available voltage to be affected by Rxer resistance, you bypass the Rxer. Also, when you bypass the Rxer, you can increase the voltage to the servos so that the resistance of long leads does not impact the voltage at the servo. And, there are no voltage sags/fluctuations on the Rxer as you can provide a dedicated regulated power circuit to the Rxer that is totally isolated from the servo's power circuits.
Back to the point: If you need, say, 30 watts of power to hold a flap in place and decrease the working voltage, the current will go up. P= IR. No way around it. If the current is increased above the servo specs, then eventually it may heat up and will fail. The flap servo's load and the load's effects on the power circuits make the flap servo more susceptible to failure as it's load is constant, once deployed. So, all things being equal, I would expect to see a voltage or undersized servo problem first show up on the flap.
#31
Power equals Voltage times current.
Voltage equals current times resistance.
Therefore power equals current squared times resistance or voltage squared divided by resistance.
The servo amplifier varies the current to the motor by means of pulse width modulation (PWM). That is it switches the current on and off when it is near the commanded position (small error signal). If it is at zero error signal it turns off. If it has a large error signal it will turn full on for the duration of the PWM period. Analog servos have a PWM period of about 20 msec and digitals have a PWM period of about 3 msec. Therefore digitals send more but shorter pulses for a given error signal.
If the servo is stalled, that is it dosn't have enough power to zero the error signal, the amplifier is sending pulses to a motor that isn't moving and so the load is the resistance of the motor.
Let's say we have stalled the flap servo and the amp has the PWM at 100%. The power into the servo is the voltage squared, say 56.25, divided by the motor resistance say one Ohm so the power is 56.25 Watts. If we drop the voltage the power goes down.
To be fair there is a small range where the PWM is not 100% where dropping the voltage might increase the PWM duty cycle to increase the current but I doubt if that is what is burning out the servos.
Hope this helps.
Jim O
Voltage equals current times resistance.
Therefore power equals current squared times resistance or voltage squared divided by resistance.
The servo amplifier varies the current to the motor by means of pulse width modulation (PWM). That is it switches the current on and off when it is near the commanded position (small error signal). If it is at zero error signal it turns off. If it has a large error signal it will turn full on for the duration of the PWM period. Analog servos have a PWM period of about 20 msec and digitals have a PWM period of about 3 msec. Therefore digitals send more but shorter pulses for a given error signal.
If the servo is stalled, that is it dosn't have enough power to zero the error signal, the amplifier is sending pulses to a motor that isn't moving and so the load is the resistance of the motor.
Let's say we have stalled the flap servo and the amp has the PWM at 100%. The power into the servo is the voltage squared, say 56.25, divided by the motor resistance say one Ohm so the power is 56.25 Watts. If we drop the voltage the power goes down.
To be fair there is a small range where the PWM is not 100% where dropping the voltage might increase the PWM duty cycle to increase the current but I doubt if that is what is burning out the servos.
Hope this helps.
Jim O
#33
Thread Starter
My Feedback: (1)
Join Date: Dec 2012
Location: Studio City, CA
Posts: 51
Likes: 0
Received 0 Likes
on
0 Posts
By way of clarification I was running 8 Hitec 7954's directly off a spectrum 12 channel rx bus powered directly by 2 each 2 cell lipos. The on board spectrum data logger registered 8.2 volts and after 5 flights I still had about 60 percent left in each lipo. Both batteries drained the same. I dismantled and inspected all wiring and found no evidence of any shorts.
Raul with jets Models Factory runs his Ultima-tun with a power distribution box so maybe that is what I will do. I have no experience with these so will have to do some research. I am real reluctant to put my plane back in the air anytime soon until I gain some understanding of what went wrong and the best course of action to prevent the same issue in the future. I appreciate all the responses so far guys.
Steve
Raul with jets Models Factory runs his Ultima-tun with a power distribution box so maybe that is what I will do. I have no experience with these so will have to do some research. I am real reluctant to put my plane back in the air anytime soon until I gain some understanding of what went wrong and the best course of action to prevent the same issue in the future. I appreciate all the responses so far guys.
Steve
#34
My Feedback: (11)
Join Date: Sep 2010
Location: marina del rey, CA
Posts: 199
Likes: 0
Received 0 Likes
on
0 Posts
There is no way that environment needs anything other than you have in terms of power supply and distribution, I have flown plenty of jets similarly set up with no issues and on my uf and viper I wasn't really concerned about my speed when I deployed the flaps. Just fit 8911's on the flaps and be done with it. At least until you solve the mystery.
as a side note I was interested in power requirements before and on my uf I taped all the surfaces in place, fitted a watt meter to the battery and worked everything at the same time. The flash had 4 8711's and 4 8611's in it powered directly through a 12121 with life's and I couldn't get the total amperage to go past 0.8, flight loads are probably higher but how much!?
as a side note I was interested in power requirements before and on my uf I taped all the surfaces in place, fitted a watt meter to the battery and worked everything at the same time. The flash had 4 8711's and 4 8611's in it powered directly through a 12121 with life's and I couldn't get the total amperage to go past 0.8, flight loads are probably higher but how much!?
#35
My Feedback: (15)
Join Date: Aug 2002
Location: Tamuning, , GUAM (USA)
Posts: 189
Likes: 0
Received 0 Likes
on
0 Posts
I'm Running 7 7954's In my H9 100cc Beast at strait lipo to receiver with a Fut 6208 Rec. Lost count on how many flights with no problems. Love these servos. I do a lot of hard 3D and still work flawlessly.
#36
My Feedback: (1)
Join Date: Feb 2002
Location: private, UNITED KINGDOM
Posts: 3,672
Likes: 0
Received 25 Likes
on
15 Posts
I have equipment that records genuine peak currents. On the ground, I have recorded a pair of JR8511s waggling the tailplanes of an AD F-100 peaking at 10 amps for the two, so 5 amps each, these peaks were for just a few thousandths of a second and would have been the point at which the motor was trying to reverse direction, so it is stalled current plus the back-emf. In flight, on a small Grumania Eurofighter with 2 analogue elevon servos, it always records a peak around 4.5amps during each flight, but I can get close to that on the ground as well so the peak aerodynamic load can't be much, once again it must be reversing current occurring for a tiny amount of time.
#37
My Feedback: (1)
Join Date: Feb 2002
Location: private, UNITED KINGDOM
Posts: 3,672
Likes: 0
Received 25 Likes
on
15 Posts
A flap servo has a constant demand for power (load) when the flap is deployed. To keep that flap down, at the same speed, takes a constant amount of power. If you need 30 watts to keep the flap down at any given speed and then reduce the voltage, the current needs to increase to give you the correct power level needed to match the servo's Pot to the signal resolution. Thus when you reduce voltage, current demand goes up. P=IR.
#40
My Feedback: (172)
Join Date: Apr 2008
Location: Waynesboro,
GA
Posts: 39
Likes: 0
Received 0 Likes
on
0 Posts
Subd. I am just about to finish a Bates 115" Sea fury and have bought 9 of 7945's for the flight systems including the quad flaps. I run the 7955s all the time on 40% aerobatic planes with excellent results. I would like to know where this ends up. Please inform us as to Hitec's response to this issue.
They have always steered me right and taken excellent care of me with any questions.
Best,
D
They have always steered me right and taken excellent care of me with any questions.
Best,
D
#41
No one has really discussed leverage.. I try to spend some time setting up the linkages where the servo has some leverage, and the horn is rotated more in line with the linkage.. this will put a lot less electrical load on the servo when it gets into position..
My best example is probably BVM's flap setup on his jets.. they emphasize that in the manual.. using the shortest servo arm you can, and the longest flap horn you can.. lowered demand..lower constant current
My best example is probably BVM's flap setup on his jets.. they emphasize that in the manual.. using the shortest servo arm you can, and the longest flap horn you can.. lowered demand..lower constant current
#42
Senior Member
My Feedback: (2)
Join Date: Nov 2002
Location: Jackson, MI
Posts: 2,102
Likes: 0
Received 0 Likes
on
0 Posts
Is people on your club slowing down the plane before flaps down?
Full-scale have flap and gear limit speeds for just this reason. If one lowers the flaps at excess speed or get too fast with them lowered, the flaps can 'blow up' (retract or fail unexpectedly).
on my uf and viper I wasn't really concerned about my speed when I deployed the flaps.
#43
My Feedback: (61)
I heard awhile back that BVM doesn't even recommend digital servos on flaps . maybe for this reason. i smoked a 5945 " i think" right on the bench just setting a plane up . it didn't even have a load on it. sure got a lot of smoke off that little critter. thought my jet was going to burn down on my bench .
#45
Have used the old 5955's for flaps forever with success, as well as lately using the spektrum AR6030's.. so far so good.
I think its so very important to monitor the overall or servo current during setup so that you are not loading the servo at flaps up and it heats up, then when its needed its already hot.. Servos on direct lipos that have a higher voltage can produce more wattage load in general, and can heat faster if loaded..
When I help set up guys planes, most often, if there is any binding its usually the gear or flap servos.. One time, we found a gear servo loaded at 2 amps..it chewed thru an A123 in short order.. If we had not put an ammeter on it, it would have run the battery out on second flight.
also in General: Most a/c should never have more than 1 amp of load on them sitting on the bench.. I get usually somewhere between .25 and 1 amp depending on how electronic crap is in the plane..
I think its so very important to monitor the overall or servo current during setup so that you are not loading the servo at flaps up and it heats up, then when its needed its already hot.. Servos on direct lipos that have a higher voltage can produce more wattage load in general, and can heat faster if loaded..
When I help set up guys planes, most often, if there is any binding its usually the gear or flap servos.. One time, we found a gear servo loaded at 2 amps..it chewed thru an A123 in short order.. If we had not put an ammeter on it, it would have run the battery out on second flight.
also in General: Most a/c should never have more than 1 amp of load on them sitting on the bench.. I get usually somewhere between .25 and 1 amp depending on how electronic crap is in the plane..
#46
My Feedback: (44)
I agree Goose, I have been using the HS 5955 and now the HS 7955TG with no problems.
At the moment servo city has a great deal on the HS 7955TG
http://www.servocity.com/html/hs-795...l#.U2eLPVcrhe8
At the moment servo city has a great deal on the HS 7955TG
http://www.servocity.com/html/hs-795...l#.U2eLPVcrhe8
#47
My Feedback: (44)
Here is a tip for your guys in case you didn't know....
By posting my post above with the link to servo city, they gave me 10% off my order of servos I just placed...
So the great price of $93.99 for the HS-7955TG just got better at $84.59 Ea, I don't know where you can find a better deal!
All you need to do is post something in a form or other social media and provide the link, a bit later they will e-mail you a discount code.
Here is the servo city link to sign up for the discount
http://www.servocity.com/html/promotions.html
By posting my post above with the link to servo city, they gave me 10% off my order of servos I just placed...
So the great price of $93.99 for the HS-7955TG just got better at $84.59 Ea, I don't know where you can find a better deal!
All you need to do is post something in a form or other social media and provide the link, a bit later they will e-mail you a discount code.
Here is the servo city link to sign up for the discount
http://www.servocity.com/html/promotions.html
#48
Join Date: Jul 2007
Location: Frederick, MD
Posts: 615
Likes: 0
Received 0 Likes
on
0 Posts
Really!!! So reposting this link will automatically guarantee me 10% off $93.99? Definitely interested in getting 7 of these servos for my big Hawk project.
Last edited by RonTins; 05-05-2014 at 01:15 PM.
#49
Join Date: Jul 2007
Location: Frederick, MD
Posts: 615
Likes: 0
Received 0 Likes
on
0 Posts
I agree Goose, I have been using the HS 5955 and now the HS 7955TG with no problems.
At the moment servo city has a great deal on the HS 7955TG
http://www.servocity.com/html/hs-795...l#.U2eLPVcrhe8
At the moment servo city has a great deal on the HS 7955TG
http://www.servocity.com/html/hs-795...l#.U2eLPVcrhe8
#50
My Feedback: (44)
There is no way that environment needs anything other than you have in terms of power supply and distribution, I have flown plenty of jets similarly set up with no issues and on my uf and viper I wasn't really concerned about my speed when I deployed the flaps. Just fit 8911's on the flaps and be done with it. At least until you solve the mystery.
as a side note I was interested in power requirements before and on my uf I taped all the surfaces in place, fitted a watt meter to the battery and worked everything at the same time. The flash had 4 8711's and 4 8611's in it powered directly through a 12121 with life's and I couldn't get the total amperage to go past 0.8, flight loads are probably higher but how much!?
as a side note I was interested in power requirements before and on my uf I taped all the surfaces in place, fitted a watt meter to the battery and worked everything at the same time. The flash had 4 8711's and 4 8611's in it powered directly through a 12121 with life's and I couldn't get the total amperage to go past 0.8, flight loads are probably higher but how much!?
http://www.rcuniverse.com/forum/rc-j...-sizing-4.html