pulsin then servo
- Thread starter alband
- Start date
Rickharris
Senior Member
JPN MA 550 APPEARS TO BE THE GYROTell me about it!
Right. You took the cover off though, yes?
There sees to be from those pictures two main chips. Can you "dicttype" the codes on them?
EZ4S546 MAY BE AN AMPLIFIER
EM78P458 AM-G LOOKS LIKE A MICROPROCESSOR
3 CONNECTIONS BETWEEN THE BOARDS ALL SOLDERED.
Darn sorry for the Caps wasn't watching
BeanieBots
Moderator
As you may have noticed, when it comes to "airborne", safety is paramount.
Have a browse around you-tube. there are videos.
I don't have my "B" liscense so can't fly at night.
VERY spectacular to watch though. You should pop down to your local club and see if they do it. Don't mention the 35Mhz/tank thing though. Baseball bats are hard to remove when inserted sideways
Have a browse around you-tube. there are videos.
I don't have my "B" liscense so can't fly at night.
VERY spectacular to watch though. You should pop down to your local club and see if they do it. Don't mention the 35Mhz/tank thing though. Baseball bats are hard to remove when inserted sideways
Last edited:
BB baseball bats; that from personal experience?
don't think they do.
I can't find anything from those codes.
I think I can see the chip in the 2nd picture - next to the screw.
Sorry to keep asking for stuff but can you get a picture of it with the numbers - might recognize the brand.
Thanks.
don't think they do.
I can't find anything from those codes.
I think I can see the chip in the 2nd picture - next to the screw.
Sorry to keep asking for stuff but can you get a picture of it with the numbers - might recognize the brand.
Thanks.
Rickharris
Senior Member
Can do may not post until tomorrow
Rickharris
Senior Member
Me too but its got too late - tomorrow will do - Any way taking challanging photographs is more interesting than School work.
Photos here http://picasaweb.google.co.uk/rickharriss/Gyro2#
I think you can see as much as there is in them.
Got it!
Picture 1 of the old lot. That is from the bottom view isn't it.
The gyro is meant to be mounted as in the attached picture shows.
In Picture 2 of the new lot. That chip is the sensor, I'm sure.
That means that it is mounted vertically.
Can you give me the length of the yellow line. The yellow line is meant to indicate the length of that board.
Picture 1 of the old lot. That is from the bottom view isn't it.
The gyro is meant to be mounted as in the attached picture shows.
In Picture 2 of the new lot. That chip is the sensor, I'm sure.
That means that it is mounted vertically.
Can you give me the length of the yellow line. The yellow line is meant to indicate the length of that board.
Attachments
Hmmm. Not to flog a dead horse, but:There are inexpensive receivers more flat than that in 75 and 72MHz as well as 900MHz (the Plantraco stuff), but I am not sure of the offereings in 35 or 40MHz.
By the way, do you have space under the turret for a cage toi drop into the hull to hold stuff and recapture you a bit of the space in the hull? If it is filled with batteries, can these be separated and rearrainged?
The explanations on gyros has been good, by the way. My son uses them in helis, and they are cool to see operate. Some of the ones built in to the micro heli bricks (all in one RX/Gyro/ESC, etc. boards) are incredibly tiny. Too bad they have only unidirectional ESC's, or I would have lots of uses.
By the way, for efficient motors with low draw (NOT for550s or other honking power hogs), RobotPower makes a nice, compact, ESC (the scorpion Mini) for thow\se with single motors. For a tank, of course, you e\would need two, or one of the dual ESCs. The transmitter mixing should have worked fine if set up right. That is basically what Vex Ro9biotocs uses for its tank-steer mode. I remember seeing an article on how to set it up properly on a Futaba su\omewhere when googleing the subjewct (I believe it was something like programming a Futaba 6 something for robots or tanks or something).
Cheers,
Wreno
By the way, do you have space under the turret for a cage toi drop into the hull to hold stuff and recapture you a bit of the space in the hull? If it is filled with batteries, can these be separated and rearrainged?
The explanations on gyros has been good, by the way. My son uses them in helis, and they are cool to see operate. Some of the ones built in to the micro heli bricks (all in one RX/Gyro/ESC, etc. boards) are incredibly tiny. Too bad they have only unidirectional ESC's, or I would have lots of uses.
By the way, for efficient motors with low draw (NOT for550s or other honking power hogs), RobotPower makes a nice, compact, ESC (the scorpion Mini) for thow\se with single motors. For a tank, of course, you e\would need two, or one of the dual ESCs. The transmitter mixing should have worked fine if set up right. That is basically what Vex Ro9biotocs uses for its tank-steer mode. I remember seeing an article on how to set it up properly on a Futaba su\omewhere when googleing the subjewct (I believe it was something like programming a Futaba 6 something for robots or tanks or something).
Cheers,
Wreno
Rickharris
Senior Member
Your yellow line would be 15 mm long
You mean stuff like this: http://microflierradio.com/Receivers.html
Lets see...
Take the 1-Ch. Mini Receiver:
$55.00 for the item
$35.00 P&P
= $90.
Current exchange = 0.659369643
= £59.34
£20.00 customs charge
£10.00 because Parcel fore demand this just for doing there job (kind of mandatory tip)
= £89.34
= £90 (If there was a smiley that depicts fainting here.)
Once you ship to UK it gets very expensive.
See attached picture.
I've got some nice bi-directionals from homefly.
You've brought up a good point though. I've got an ESC controlling the turret which it looks like I'm going to put a gyro on. BB has since convinced me it will work, but I forgot to mention that the ESC is bi-directional. Will it still work?
15mm. Think I'll be able to fit that in. Expect a PM soon(ish).
Lets see...
Take the 1-Ch. Mini Receiver:
$55.00 for the item
$35.00 P&P
= $90.
Current exchange = 0.659369643
= £59.34
£20.00 customs charge
£10.00 because Parcel fore demand this just for doing there job (kind of mandatory tip)
= £89.34
= £90 (If there was a smiley that depicts fainting here.)
Once you ship to UK it gets very expensive.
See attached picture.
I've got some nice bi-directionals from homefly.
You've brought up a good point though. I've got an ESC controlling the turret which it looks like I'm going to put a gyro on. BB has since convinced me it will work, but I forgot to mention that the ESC is bi-directional. Will it still work?
15mm. Think I'll be able to fit that in. Expect a PM soon(ish).
Attachments
Actually, I was thinking about some of the various ones from http://www.micronradiocontrol.co.uk/ which are only 7-9mm thick )depending on model) and are in the £10.00 to £16.00 range. Surely, Micron being in Malton, North Yorkshire YO17 6N, there is no special shipping or customs? Though you will have VAT, I guess.
Remember, you can use 2 4 channel radio receivers on the same transmitter - just hook up channels 1 and 2 for tracks on your maneuvering RX - leaving 3 and 4 bare, hook up to channels 3 and 4 on the turret control RX - leaving 1 and 2 bare. Oops - 1,2 and a third channel, since you need turret rotation, on the main shell maneuvering RX, so, a 5 channel radio minimum there. This allows you to get rid of the center post thingamajiggy and free up a bit of space under the turret.
The picture helps with visualization, thanks.
Cheers,
Wreno
Though you will have VAT, I guess.Remember, you can use 2 4 channel radio receivers on the same transmitter - just hook up channels 1 and 2 for tracks on your maneuvering RX - leaving 3 and 4 bare, hook up to channels 3 and 4 on the turret control RX - leaving 1 and 2 bare. Oops - 1,2 and a third channel, since you need turret rotation, on the main shell maneuvering RX, so, a 5 channel radio minimum there. This allows you to get rid of the center post thingamajiggy and free up a bit of space under the turret.
The picture helps with visualization, thanks.
Cheers,
Wreno
BeanieBots
Moderator
Why would you want to do that? I thought the servo which required "gyro" adjustment was in the turret?
Bi-directional ESC + gyro....
Think about it for a while..... What do you think a servo is?
In a servo, the position pot adjusts the bi-directional ESC signal to control position. With a gyro, the gyro also adjusts the bi-directional ESC signal to control position. With the pot, it controls the position of the pot. With a gyro, it controls the position (angle) of the gyro.
The ESC for the turret rotation is in the hull. I'd have the gyro for the turret rotation in the turret and then I'm going to mount an accelerometer in the gun for thegun elevation. The gun elevation will be controlled by the accerlerometer throught a servo in the turret and the turret rotation is being powered by a motor and ESC in the hull. I would therefore need the gyro in the turret and send the line down to the ESC.
Could I do all this communication in one pin?
Could I do all this communication in one pin?
BeanieBots
Moderator
Don't think you can do it all on one pin.
Rather than second guess your hardware, here's how it works.
Rx sends out servo position demand. That goes to the gyro. Gyro sends out "corrected" position demand. That goes to the sero. (or ESC in your case).
It is a 'continuous' signal just like any other servo signal. However, the amount of 'correction' varies as the turret moves. If you tried a system like the one using serial data discussed earlier, I think you would have problems with timing.
The tank turn a little bit. The gyro detects it and "corrects' the servo pulses. A little bit later, the "communication" method sends the corrected servo command. Meanwhile, the gyro has not seen an immediate position correction so it adjusts further. By now, the turret has moved to the correct position. then the over-corrected pulses start to arrive. I'm sure you get the picture.
As mentioned earlier, the gyro to servo pulse algorythm is a finely tuned PID loop. By adding a 'comms' delay you may well upset the terms enough to cause oscillation. Turning down the gyro gain will help to prevent that, but if too low, the respsonse will be very 'sloppy'. It would be quicker and easier to try it than it would be to calculate with any level of certainty.
Rather than second guess your hardware, here's how it works.
Rx sends out servo position demand. That goes to the gyro. Gyro sends out "corrected" position demand. That goes to the sero. (or ESC in your case).
It is a 'continuous' signal just like any other servo signal. However, the amount of 'correction' varies as the turret moves. If you tried a system like the one using serial data discussed earlier, I think you would have problems with timing.
The tank turn a little bit. The gyro detects it and "corrects' the servo pulses. A little bit later, the "communication" method sends the corrected servo command. Meanwhile, the gyro has not seen an immediate position correction so it adjusts further. By now, the turret has moved to the correct position. then the over-corrected pulses start to arrive. I'm sure you get the picture.
As mentioned earlier, the gyro to servo pulse algorythm is a finely tuned PID loop. By adding a 'comms' delay you may well upset the terms enough to cause oscillation. Turning down the gyro gain will help to prevent that, but if too low, the respsonse will be very 'sloppy'. It would be quicker and easier to try it than it would be to calculate with any level of certainty.
westaust55
Moderator
Would you not be better using an incinometer to measure thegun elevation?
Inclinometers are intended to measure angles/slopes/elevation.
Do not know of any hobby versions but a google search should put you onto something.
BeanieBots
Moderator
I'd agree with westaust except for the minor detail of actually doing it.
The gyro method is tried, tested, works, packaged and designed specifically for use with servos (& ESCs). Not only will it be a lot smaller than any other method, by the time you've bought all the other bits required, quite seriously, I would also expect it to be cheaper. Might be fun to have a go though. Really depends where you want your project to go. ie. fit a solution to your tank, or make the solution the major part of the project with the tank being an excuse for it.
The gyro method is tried, tested, works, packaged and designed specifically for use with servos (& ESCs). Not only will it be a lot smaller than any other method, by the time you've bought all the other bits required, quite seriously, I would also expect it to be cheaper. Might be fun to have a go though. Really depends where you want your project to go. ie. fit a solution to your tank, or make the solution the major part of the project with the tank being an excuse for it.
I'm sorry, I thought you were looking for 35 or 40MHz??? In fact, I understood that you were currently using 35MHz, correct? That is why I listed that site. If you are looking for something else, there are different Micron also carries the Plantraco stuff, though I don't, offhand, know of more than 4 channels units available in Plantraco.
What frequency band are you wanting?
Another, "outside the box" solution: One of the tiny remote control offerings (as in key-fob thingies) that give you 2-4 on-off buttons. Instead of a servo, just a geared motor for the elevation, then use 3 of the channels for up, down, and laser. Sparkfun carries these, but you can also find them on E-bay for cheap.
Cheers,
Wreno
Need the servo. Think of having a laser in the gun. The gun travels about 60 degrees. That is split up into about 100 steps (due to EPA). That give 0.6 degree accuracy. The wall is about 3m away (you're not just going to aim at the closes wall, you'd have some fun). Therefore 0.6 degrees accuracy equates to about 3cm on the wall. That in itself isn't great.
Now imagine trying to time that accuracy with just on/off, up/down controls compared to "go to point" servo's. You'd need the fingers of a robot on steroids.
I really don't know about the frequency. The reciever I got is marked as 35Mhz and I'd like to keep it. Would it work on something like 2x35Mhz; 72Mhz?
I'll need to try some combinations at some point.
BB: I'm surprised enough you managed to sway me about the gyro for turret rotation. You can keep trying for elevation too but don;t hold your breath (although that was my sentiment the first time).
West: What is the difference between a Inclinometer and an accelerometer? I found a suitable accelerometer for about £11. (http://www.active-robots.com/products/sensors/sparkfun/8658.shtml). And I'm not even sure that is small enough (see post 132 (page 14) I have to fit it in that slot just to the left of the end of the red wire).
Now imagine trying to time that accuracy with just on/off, up/down controls compared to "go to point" servo's.
You'd need the fingers of a robot on steroids.I really don't know about the frequency. The reciever I got is marked as 35Mhz and I'd like to keep it. Would it work on something like 2x35Mhz; 72Mhz?
I'll need to try some combinations at some point.
BB: I'm surprised enough you managed to sway me about the gyro for turret rotation. You can keep trying for elevation too but don;t hold your breath (although that was my sentiment the first time).
West: What is the difference between a Inclinometer and an accelerometer? I found a suitable accelerometer for about £11. (http://www.active-robots.com/products/sensors/sparkfun/8658.shtml). And I'm not even sure that is small enough (see post 132 (page 14) I have to fit it in that slot just to the left of the end of the red wire).
BeanieBots
Moderator
alband, you WON'T be able to run a 35Mhz Rx at 72Mhz.
You can JUST ABOUT get away with running it at 40Mhz. The range will be reduced. By how much depends on the make.
Inclinometer measures angle. Accelerometer measures accelleration (rate of change of velocity).
Gyro measure rate of change of angle.
A "hobby gyro" also calculates the amount of corrective servo pulse required to add to the servo pulse to bring it back to the correct position. ie saves the requirement of a very fast processor and writing (in assembler) the required control algorythm.
You can JUST ABOUT get away with running it at 40Mhz. The range will be reduced. By how much depends on the make.
Inclinometer measures angle. Accelerometer measures accelleration (rate of change of velocity).
Gyro measure rate of change of angle.
A "hobby gyro" also calculates the amount of corrective servo pulse required to add to the servo pulse to bring it back to the correct position. ie saves the requirement of a very fast processor and writing (in assembler) the required control algorythm.
Last edited:
BeanieBots
Moderator
You can make your own inclinometer using a POT and a suspended weight.
As you rotate the body of the pot, the weight (fixed to the shaft), turn the pot and hence changes it resistance. Instant inclinometer.
Trouble is, if you accelerate forwards, the weight stays behind, oh dear, now it's an accelerometer! Then again, you could always use a gyro
As you rotate the body of the pot, the weight (fixed to the shaft), turn the pot and hence changes it resistance. Instant inclinometer.
Trouble is, if you accelerate forwards, the weight stays behind, oh dear, now it's an accelerometer! Then again, you could always use a gyro
BeanieBots
Moderator
Whatever gave you that idea?
I love anything that does the job. When I want something to roll, I use wheels. Never felt the urge to re-invent that system either.
Why not just get a few of those sensors and have a play?
I love anything that does the job. When I want something to roll, I use wheels. Never felt the urge to re-invent that system either.
Why not just get a few of those sensors and have a play?
BeanieBots
Moderator
Well, that's a good and healthy attitude. But first, you need to have a working system done the "conventional" way before you can improve on it.
Most things have many different solutions and with just a pen and paper, it is often difficult to know if any method is better. It's not until you have the hardware in your hand that many of the "realities" come to light. It's the bits that don't quite fit with that nice theory that often determine the "best" method.
Most things have many different solutions and with just a pen and paper, it is often difficult to know if any method is better. It's not until you have the hardware in your hand that many of the "realities" come to light. It's the bits that don't quite fit with that nice theory that often determine the "best" method.
Here's a few.
http://www.microstrain.com/inclinometers.aspx?gclid=CJH58462lpcCFYwH3god6QamDA
I bet they're expensive though.
e
http://www.microstrain.com/inclinometers.aspx?gclid=CJH58462lpcCFYwH3god6QamDA
I bet they're expensive though.
e
BeanieBots
Moderator
You can't have tried very hard.
The second hit with Google gives a
wait for it....
GYRO ENHANCED inclinometer
Anyway, I think you have already spotted that this thread is going round in circles. It's time to try something. Anything except another theory. Get some kit, play with it. Come back and explain your findings.
I've been there and done that so there's no point me going on and on at you about it. It's not until you have tried it for yourself that you will understand or appreciate the pro's and con's of each method.
Don't let getting it into the tank be an obstical. Just try some ideas out on broom hande if you have to.
"GYRO ENHANCED inclinometer" too big and too expensive - now where have I heard that several times before?
Don't want to buy unnecessaries that's all.
You'll love the fact that I am currently trying to buy that accelerometer of Rick.
I've been getting a tiny servo set up in the front right section of the turret so that I don't need a linkage from the back. Unfortunatly, I need to carve about 2mm out of that inner gear, and it is made of as stubbener metal as you can think of. I'm going to have to WAITeek until the weekend to rip it out (R.I.P. perfect positioning) and then grind the 2mm off. Then (hopefully (I use brackets way too much)) I will be able to glue it in in exactly the same place using the glue left behind.
Don't want to buy unnecessaries that's all.
You'll love the fact that I am currently trying to buy that accelerometer of Rick.
I've been getting a tiny servo set up in the front right section of the turret so that I don't need a linkage from the back. Unfortunatly, I need to carve about 2mm out of that inner gear, and it is made of as stubbener metal as you can think of. I'm going to have to WAIT
eek until the weekend to rip it out (R.I.P. perfect positioning) and then grind the 2mm off. Then (hopefully (I use brackets way too much)) I will be able to glue it in in exactly the same place using the glue left behind.Gotcha. You are wanting absolute positioning. A couple of things you might want to consider, however. Your weapons platform is moving and irregular (not as irregular as our model warships, but not stable, nonetheless). With a geared motor that fives you a slow elevation, you can fine-tune the elevation visually on the fly. If you want to walk it to the target. Many of our guys use a simple motor for turret rotation (my ship happens to be proportional, but I am in the minority) on our ships and are deucedly effective.Need the servo. Think of having a laser in the gun. The gun travels about 60 degrees. That is split up into about 100 steps (due to EPA). That give 0.6 degree accuracy. The wall is about 3m away (you're not just going to aim at the closes wall, you'd have some fun). Therefore 0.6 degrees accuracy equates to about 3cm on the wall. That in itself isn't great.
Now imagine trying to time that accuracy with just on/off, up/down controls compared to "go to point" servo's.
As pointed out by others, no, you cannot mix frequency bands (or for that matter, frequency crystals in those bands). Once a TX is on a frequency, the RX frequency must match it.
However, several of the RX's carried by Micron match your TX.
Cheers,
Wreno
Last edited:
I'm worried (and therefore haven't bought anything yet, soz BB).
This accelerometer, it will be able to tell which way up it is, not just if it is accelerating in any direction. It keeps saying it can measure acceleration - is that what the boffins who made it are calling gravity these days?
http://www.active-robots.com/products/sensors/sparkfun/8658.shtml
EDIT: Also, not sure where to buy it. If I order from above link I get the breakout board. and it come too a total of £11.17.
If I order from rapid I can get it for £16.39 but I don't get the breakout board. How hard would it be do make a board for this component with it's footprint (and solder it)?
This accelerometer, it will be able to tell which way up it is, not just if it is accelerating in any direction. It keeps saying it can measure acceleration - is that what the boffins who made it are calling gravity these days?
http://www.active-robots.com/products/sensors/sparkfun/8658.shtml
EDIT: Also, not sure where to buy it. If I order from above link I get the breakout board. and it come too a total of £11.17.
If I order from rapid I can get it for £16.39 but I don't get the breakout board. How hard would it be do make a board for this component with it's footprint (and solder it)?
Last edited:
Alband.
Acceleration and Gravity?
"is that what the boffins who made it are calling gravity these days?"
Very seriously, take a look at this site.
http://openlearn.open.ac.uk/course/view.php?id=1735
For your second question, have a scan of this thread.
http://www.picaxeforum.co.uk/showthread.php?t=9751
e
Acceleration and Gravity?
"is that what the boffins who made it are calling gravity these days?"
Very seriously, take a look at this site.
http://openlearn.open.ac.uk/course/view.php?id=1735
For your second question, have a scan of this thread.
http://www.picaxeforum.co.uk/showthread.php?t=9751
e