Wanted to post a build log for reference if anyone is also in the midst of building one! I'll try to provide as many important tips as I can. Although I'm almost completed with the build I'm posting it as though it were logged as I went.
I'll be constructing the version 2 of the focuser with bluetooth so it's wireless and hands free. I'm building for my Celestron NexStar 8se on the Advanced VX mount. I've been able to get great results unguided so far, however the biggest issue I face on this scope is the mirror shift/flop that occurs as it tracks, which bumps meticulous focusing right out of whack. I've had multiple evenings' data invalidated from this. I'll just have a program autofocus every period of time and this should help with this problem.
I initially had decided to use the NEMA17PG27 stepper connected directly to the focuser shaft (I didn't want or need pulleys or belts because there is no physical impediment to the AVX mount from a focuser protruding from the back). I almost purchased it but after a discussion with Robert (initially about the bluetooth capabilities) and some testing of the amount of torque needed to turn this focus knob, I switched to the 28BYJ-48 motor (much, much lighter and smaller) with ULN2003 driver board. I decided against converting it to bipolar mode because I expected it to be sufficient based on torque testing (https://www.youtube.com/watch?v=14jF8umwJLI). I can always switch boards if this becomes necessary.
All the electronics parts have been ordered on Tayda electronics with which I'm really, really impressed. They offer fast, inexpensive shipping and parts for pennies apiece! Get the jumper wires and other items related to your build. They have project boxes too.
I've ordered the PDB from Aisler so it'll be a nice, organized layout. I also got the hc-05 module from Ebay, the motors and driver boards (a set of 5 of each) cheap on Ebay, and a flexible coupler there as well. It doesn't have a wide enough entry on the other end for the focusing shaft so I'll have to drill it out.
Last edit: Anonymous 2021-02-04
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2018-12-29
Total list of parts and cost:
~ Electronics parts from Tayda (everything from switches and resistors to box): $22
~ PCBs from Aisler: $30
~ 5 sets of motors (28BYJ-48) and driver boards (ULN2003) from Ebay: $14
~ Flexible coupler from Ebay: $5
~ Set of 2 arduino nanos from Ebay: $12
~ HC-05 bluetooth module from Ebay: $8
~ Extra jumper wires: $5
~ Metal plate for mounting at local home depot: $3
Total cost: A Measly $99. Impressive what we can get nowadays online!
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2018-12-29
The coupler and motors were the first to arrive (very fast). After some painful drilling by hand to allow the flexible coupler to have a 1/2" size hole on one end (the outer diameter of the 8se focus shaft when the rubber grip is removed), I finally managed to get it to fit. It uses hex-style set screws to fasten. The 28BYJ-48 motors have a shaft diameter of 5mm at the very top which is already the size of the other end of this coupler.
When the arduinos arrived, I used the simplest code available (from here which causes it to simply rotate continuously) to test the torque on my focusing shaft to make sure it was sufficient without building everything first. I used a breadboard I already owned to hook the arduino to the driver board and the board to the motor.
Last edit: Anonymous 2021-02-04
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2018-12-29
I worked on the mount (physical connection to the telescope) while the electronics were coming in. I simply got a metal tie-down from home depot, cut to length and bent twice. I wanted it to be removable for hand-focusing again, so grooves have been added to slide into the three screws that hold the focuser knob plate in.
I've also tested the motor as described before (continuous rotation) with the new mount, and everything is good!
Last edit: Anonymous 2021-02-04
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2018-12-29
Once the electronics came in (I'm really, really impressed with Tayda, which provided very quick and accurate shipping as well as low prices) I started affixing everything to the project box including the power switch, LEDs, serial connector for the motor, power jack and 2A fuse to protect the boards.
Last edit: Anonymous 2021-02-04
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2018-12-29
I also constructed the voltage power circuit (which includes the L7808CV chip and associated capacitors). An important note on this circuit (such tips are the reason for this long set of posts), this voltage regulator dissipates a lot of heat! To calculate the temperature it will reach for your setup, you need to know your input and output voltage for this chip as well as the current. You then multiply the power it will dissipate by the temperature coefficient to get the temperature rise for that configuration.
In my case, the input voltage was as high as 12.4V, and the output while being used was as low as 7.8V. The maximum current this circuit drew for me was 500mA, in which case the power dissipated by the regulator is (Voltage Drop) x (Current) = (4.6) x (0.5) = 2.3 Watts. Note that if you choose to use coil power, you'll be constantly drawing this current (which is more than when the motor is turning) rather than only when moving the controller. You'll need coil power for fractional stepping, which I did not. Now visit the datasheet of your regulator:
This regulator causes a 53 degree Celsius increase per watt dissipated, meaning that this regulator's temperature will rise (during the 0.5A current draw times) by (53 Degrees/Watt) x (2.3 Watts) = 121.9 Degrees C. Yikes! At room temperature (21 Degrees C) this will run at around 143 Degrees C, which is near but below the max temperature (see datasheet above). This issue is resovled for me because I don't need coil power (so the motor is only drawing current when moving, which is rare), and nights are usually cold enough to remove another 10 degrees from this maximum total. It does, however, have to be shielded from the wires in the box to prevent melting them! See later on for how I handled this.
Here is everything wired together on my breadboard to test the circuit since the PCBs aren't in yet. You can see the voltage regulating circuit on the left in its own separate board.
Last edit: Anonymous 2021-02-04
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi Alex
Not sure how you measured the current draw, was it at the 12V side?
The LM7808 only powers the board and leds. Motor current should be from the 12V source and not supplied by the LM7808. The typical draw is <150mA for the board, a little more when leds are pulsing. See the DV8825 circuit. I suspect you used the 8V out to power the ULN2003? It should work relatively fine on 12V if you do not have coil power on.
The LM7808 should'n't really get hot in normal operation. Coil power on will not affect the current through the LM7808 as its from the 12V supply unless you wired it via the LM7808.
As the ULN2003 PCB does not use a LM7808 I guess you added this? Heat will not be an issue with the other versions that use the LM7808 as stepper power is direct from the 12V supply rail
Regards
Robert
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2019-01-13
Yup you're correct, I power the entire board with the output 8V from the 7808 regulator, and since the power of the ULN2003 board is part of the PCB (VCC), it is also powered in turn by the 7808 chip. Should I run into problems with this setup, I'll look at rerouting the 12V in to the ULN2003 driver board without the 7808. So far so good, however!
I don't see how I would do this though, because the VCC pin on the arduino is directly connected to the ULN power lead on the PCB. I'll investigate this further should any issues arise. Thanks!!
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Very impressive so far, I really liked your detail and photos, awesome!
Looking forward to the next installment. Tayda is also one of my preferred sellers, very good
Regards
Robert
PS. If you draw a drawing of your bracket with dimensions and email it me to I am happy to knock one up for you on a 3D printer free of charge.
Last edit: brownrb 2018-12-29
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2018-12-30
Many thanks Robert! I will keep this in mind for any future parts (I have access to solidworks), however this bracket is great as metal because it behaves like a heat sink for the motor. I actually placed thermal paste between the motor and the bracket.
Do you have your own 3D printer?
Last edit: Anonymous 2018-12-30
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2018-12-31
Much appreciated! I'll keep that in mind for future parts. Thanks.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2018-12-30
I took a while trying to figure out why the motor was not behaving properly. It was skipping steps, buzzing and shaking in place (but not rotating), and stopping early. It was also just vibrating sometimes but not rotating. I tried changing the motor and the board and neither was the culprit. I also tried different power sources. No help. I was growing worried that this motor was not going to work because of some "weakness" in its rotation (I hadn't used this stepper before and now know that it wasn't working anywhere close how well it should).
I was about to give up (until using the printed PCB when it arrived, which I figured could solve the problem- no sense in troubleshooting a prototype too heavily) but decided to do some continuity testing on the circuit and found that one of the jumper wires had literally pushed back through its plastic casing while being pushed into the protoboard, meaning one of the 4 channels into the ULN2003 was literally disconnected! What is odd is that the motor still attempts to work, although unsuccessfully, if this happens. Hopefully this will help someone else in the future with this motor.
Here is the working attempt before the PCB came in. Pardon the blurriness, apparently my phone can't choose a sufficiently quick exposure time.
Last edit: Anonymous 2021-02-04
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2018-12-30
PCB finally arrived from Aisler (very, very high quality boards, although not very quick arrival from Germany: over 3 weeks after ordering). I recommed not only soldering female sockets to the arduino's 30 pins (so it can be removed from the board as desired), but also with the LEDs, switches, power, driver board and bluetooth module. This saved me multiple headaches. Here's a closeup of this strategy:
I'll admit, my biggest fear was getting the bluetooth module working to spec (I've wrestled with items as simple as bluetooth headphones on Windows and was not looking forward to this), so I tested everything before adding the HC-05. Note that if you intend to try this first, you must comment out the bluetooth portion in the firmware file (and any other features you won't be using) You will have to reflash the firmware for each element you decide to add as you go. Here is everything save the bluetooth module.
Once it was working via USB cable, I attached the bluetooth module accordingly and to my strong surprise, it operated perfectly the first time. If you're on Windows and aren't so lucky, power off the focuser after flashing the firmware with the bluetooth functionality uncommented, reset your computer, turn on the focuser, go to "add a bluetooth device" and choose to place in the object's passcode, which by default is 1234 for most of these. When added, the drivers should be installed automatically and a virtual COM port should be added. Check in the device manager for it under "ports (COM & LPT)" if you're on Windows 7. If you're on Windows 10, I'm sorry to hear that! Note the COM port number for this connection
(standard serial over bluetooth link) and use this COM port in the Windows application for the focuser to connect. You may need to update this COM port number in the ASCOM interface so it points to the new wireless location. From there it's plug and play! Here is everything combined:
Last edit: Anonymous 2021-02-04
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2018-12-30
I added an LED for the power indicator (note that the PCB for the power indicator red LED does not have a resistor and one must be added to the LED above the board) and drilled the hole for it, as well as screwed the power supply circuit, PCB and ULN2003 board into the hobby box.
I decided that I'd like to have the bluetooth antenna of the HC-05 protrude from the box to give it the cleanest signal to my laptop. I did tests and had successful connection to the machine even from upstairs, traveling through a ceiling and multiple walls! This is better performance than my bluetooth headphones even get. Regardless, no reason not to have the antenna exit the box, which I accomplished by cutting a slot:
I debated awhile on how to shield the 7808 chip's high surface temperature from any wires that would find their way onto it, so as to prevent melting. It dawned on me to use PVC pipe to physically shield it from other wires! Without contacting the PVC, the chip will not melt it but should also stop wires from touching it.
Last edit: Anonymous 2021-02-04
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2018-12-30
Here's the completed project, using a power LED, Red/Green LEDs, push buttons, power switch, fuse, and bluetooth:
And the 8se OTA still fits in my case! Glad I went with the 28BYJ-48 motor instead of the giant and heavy NEMA choices.
Helpful information and field tests to come. Thanks Robert!
Last edit: Anonymous 2021-02-04
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2018-12-30
Useful notes:
~ The firmware (as of 12-29-18) entitled "Focuserv277_ULN2003H" is the correct choice for the full option, full-stepping implementation of the ULN2003 board with the 28BYJ-48 motor; the H simply implies this firmware is capable of fractional stepping.
~ Coil power will cause a continuous draw of roughly half an amp with this motor and driver board, because the motor is constantly maintaining its position in fractional steps. Although I didn't need this, you'll have to account for the heating of the power supply and motor itself (I placed thermal paste between the motor and metal mounting bracket- so it can act as a heat sink- to help this). Best not to use it if you're full stepping, unless there's another reason to use coil power that I don't know about?
~ The bluetooth signal (with the antenna not enclosed in the box, is strong and solid more than 15ft away, even through multiple walls. I can't move far enough away from it indoors to lose connection! Once the COM port is assigned, AstrophotographyTool (APT) recognizes and operates it easily!
~ Neither Astrophotography tool (APT) nor Sequence Generator Pro (SGPro) allow wireless or WiFi connection to Nikon cameras, in any way, as of 12-29-18. Some Canon models can utilize EOS tool to trick it to work. Boo on you, Nikon software. Therefore the USB cable to the camera will be necessary.
~ The Nikon d3000, d3100, d3200, d3300, d3400, and d3500, among other introductory models, do not offer Software Development Packages (SDKs) and therefore cannot work with these programs in many ways, including bulb and live view modes. Some are not workable altogether.
~ I see why many people have been using gears and pulleys now. Although nothing was easier than my simple mounting bracket directly to the focusing shaft, this prevents the camera from attaching at prime focus to the back of the scope in roughly 120 of the 360 degrees it could before. Look out for how the camera will mount to the scope when designing the mounting bracket. I will see how long I can go shooting at the new allowed angles, although some large targets don't have much freedom in the angle at which the camera is mounted since the sensor/aspect ratio is rectangular and the tube cannot rotate. A few workarounds I'll have to consider include redesigning the attachment, or extending the connection between the camera and the OTA so it clears the focuser bracket when rotating the camera body to those angles. The latter method is always one I try to avoid, however, to assure there's enough focus travel! Moving the imaging focal plane always risks these problems. I'll see if I can get away with the current configuration for awhile, however.
Last edit: Anonymous 2018-12-30
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2019-01-11
Adding the extra tube between the camera and telescope (to extend the distance from the tube it reaches and thereby eliminating the focuser being in the way when I rotate the camera at different angles) proved to be a non-issue; I can still achieve focus with plenty of extra room. Here is the before situation, where the focuser was in the way of the camera if I rotated it to certain angles since I didn't use pulleys or gears:
You can see the camera wouldn't be able to rotate further because the motor is in the way. Adding the following extra distance (tube) solves the problem, and still allows for focusing on my 8se
I can now rotate to any angle with the camera without using any pulleys or gears on the focuser.
Last edit: Anonymous 2021-02-04
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2019-01-11
This focuser is awesome. I've tested it successfully on AstrophotographyTool (APT) and Sequence Generator Pro (SGPro) and it works perfectly on both. I'm super impressed with the bluetooth connectivity with the HC-05 module too; on my machine of choice, I get connectivity through the wall inside (when the scope is outside) roughly 30-40ft away without any disconnection (except if the program itself crashes or has a problem) for hours. I haven't needed "coil power" and therefore this thing will not be drawing much power at all. I did have problems getting the same range on other machines, but I'm fairly sure that this is because of outdated hardware or non-updated drivers on them, although I don't know for sure.
Having a focuser has served two different purposes. I wanted it so I could run auto-focusing algorithms that would compensate for the mirror flop I've experienced on this scope; many nights the focus would become unusably bad after under an hour of imaging, and unfortunately this is due to settling of the primary mirror on its own tracks, not the knob moving. I wanted to periodically correct for this to keep me in focus the whole night.
However, I have found that the algorithms employed by SGPro in particular are so accurate in achieving overall focus (they collect stars thoughout the image and find a global average HFR) that I found the results better than I could even get by hand. I would spend 15 or more minutes at a time trying to get focus manually and it wouldn't even settle there. However, this programs leads to focus tighter than I could ever get by hand. Here is a comparison, showing an attempt at M42 I did many months ago vesus running autofocusing routines on SGPro the other night:
Clearly these stars are tighter and much more focused.
Last edit: Anonymous 2021-02-04
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Wanted to post a build log for reference if anyone is also in the midst of building one! I'll try to provide as many important tips as I can. Although I'm almost completed with the build I'm posting it as though it were logged as I went.
I'll be constructing the version 2 of the focuser with bluetooth so it's wireless and hands free. I'm building for my Celestron NexStar 8se on the Advanced VX mount. I've been able to get great results unguided so far, however the biggest issue I face on this scope is the mirror shift/flop that occurs as it tracks, which bumps meticulous focusing right out of whack. I've had multiple evenings' data invalidated from this. I'll just have a program autofocus every period of time and this should help with this problem.
I initially had decided to use the NEMA17PG27 stepper connected directly to the focuser shaft (I didn't want or need pulleys or belts because there is no physical impediment to the AVX mount from a focuser protruding from the back). I almost purchased it but after a discussion with Robert (initially about the bluetooth capabilities) and some testing of the amount of torque needed to turn this focus knob, I switched to the 28BYJ-48 motor (much, much lighter and smaller) with ULN2003 driver board. I decided against converting it to bipolar mode because I expected it to be sufficient based on torque testing (https://www.youtube.com/watch?v=14jF8umwJLI). I can always switch boards if this becomes necessary.
All the electronics parts have been ordered on Tayda electronics with which I'm really, really impressed. They offer fast, inexpensive shipping and parts for pennies apiece! Get the jumper wires and other items related to your build. They have project boxes too.
I've ordered the PDB from Aisler so it'll be a nice, organized layout. I also got the hc-05 module from Ebay, the motors and driver boards (a set of 5 of each) cheap on Ebay, and a flexible coupler there as well. It doesn't have a wide enough entry on the other end for the focusing shaft so I'll have to drill it out.
Last edit: Anonymous 2021-02-04
Total list of parts and cost:
~ Electronics parts from Tayda (everything from switches and resistors to box): $22
~ PCBs from Aisler: $30
~ 5 sets of motors (28BYJ-48) and driver boards (ULN2003) from Ebay: $14
~ Flexible coupler from Ebay: $5
~ Set of 2 arduino nanos from Ebay: $12
~ HC-05 bluetooth module from Ebay: $8
~ Extra jumper wires: $5
~ Metal plate for mounting at local home depot: $3
Total cost: A Measly $99. Impressive what we can get nowadays online!
The coupler and motors were the first to arrive (very fast). After some painful drilling by hand to allow the flexible coupler to have a 1/2" size hole on one end (the outer diameter of the 8se focus shaft when the rubber grip is removed), I finally managed to get it to fit. It uses hex-style set screws to fasten. The 28BYJ-48 motors have a shaft diameter of 5mm at the very top which is already the size of the other end of this coupler.
When the arduinos arrived, I used the simplest code available (from here which causes it to simply rotate continuously) to test the torque on my focusing shaft to make sure it was sufficient without building everything first. I used a breadboard I already owned to hook the arduino to the driver board and the board to the motor.
Last edit: Anonymous 2021-02-04
I worked on the mount (physical connection to the telescope) while the electronics were coming in. I simply got a metal tie-down from home depot, cut to length and bent twice. I wanted it to be removable for hand-focusing again, so grooves have been added to slide into the three screws that hold the focuser knob plate in.
I've also tested the motor as described before (continuous rotation) with the new mount, and everything is good!
Last edit: Anonymous 2021-02-04
Once the electronics came in (I'm really, really impressed with Tayda, which provided very quick and accurate shipping as well as low prices) I started affixing everything to the project box including the power switch, LEDs, serial connector for the motor, power jack and 2A fuse to protect the boards.
Last edit: Anonymous 2021-02-04
I also constructed the voltage power circuit (which includes the L7808CV chip and associated capacitors). An important note on this circuit (such tips are the reason for this long set of posts), this voltage regulator dissipates a lot of heat! To calculate the temperature it will reach for your setup, you need to know your input and output voltage for this chip as well as the current. You then multiply the power it will dissipate by the temperature coefficient to get the temperature rise for that configuration.
In my case, the input voltage was as high as 12.4V, and the output while being used was as low as 7.8V. The maximum current this circuit drew for me was 500mA, in which case the power dissipated by the regulator is (Voltage Drop) x (Current) = (4.6) x (0.5) = 2.3 Watts. Note that if you choose to use coil power, you'll be constantly drawing this current (which is more than when the motor is turning) rather than only when moving the controller. You'll need coil power for fractional stepping, which I did not. Now visit the datasheet of your regulator:
This regulator causes a 53 degree Celsius increase per watt dissipated, meaning that this regulator's temperature will rise (during the 0.5A current draw times) by (53 Degrees/Watt) x (2.3 Watts) = 121.9 Degrees C. Yikes! At room temperature (21 Degrees C) this will run at around 143 Degrees C, which is near but below the max temperature (see datasheet above). This issue is resovled for me because I don't need coil power (so the motor is only drawing current when moving, which is rare), and nights are usually cold enough to remove another 10 degrees from this maximum total. It does, however, have to be shielded from the wires in the box to prevent melting them! See later on for how I handled this.
Here is everything wired together on my breadboard to test the circuit since the PCBs aren't in yet. You can see the voltage regulating circuit on the left in its own separate board.
Last edit: Anonymous 2021-02-04
Hi Alex
Not sure how you measured the current draw, was it at the 12V side?
The LM7808 only powers the board and leds. Motor current should be from the 12V source and not supplied by the LM7808. The typical draw is <150mA for the board, a little more when leds are pulsing. See the DV8825 circuit. I suspect you used the 8V out to power the ULN2003? It should work relatively fine on 12V if you do not have coil power on.
The LM7808 should'n't really get hot in normal operation. Coil power on will not affect the current through the LM7808 as its from the 12V supply unless you wired it via the LM7808.
As the ULN2003 PCB does not use a LM7808 I guess you added this? Heat will not be an issue with the other versions that use the LM7808 as stepper power is direct from the 12V supply rail
Regards
Robert
Yup you're correct, I power the entire board with the output 8V from the 7808 regulator, and since the power of the ULN2003 board is part of the PCB (VCC), it is also powered in turn by the 7808 chip. Should I run into problems with this setup, I'll look at rerouting the 12V in to the ULN2003 driver board without the 7808. So far so good, however!
I don't see how I would do this though, because the VCC pin on the arduino is directly connected to the ULN power lead on the PCB. I'll investigate this further should any issues arise. Thanks!!
Very impressive so far, I really liked your detail and photos, awesome!
Looking forward to the next installment. Tayda is also one of my preferred sellers, very good
Regards
Robert
PS. If you draw a drawing of your bracket with dimensions and email it me to I am happy to knock one up for you on a 3D printer free of charge.
Last edit: brownrb 2018-12-29
Many thanks Robert! I will keep this in mind for any future parts (I have access to solidworks), however this bracket is great as metal because it behaves like a heat sink for the motor. I actually placed thermal paste between the motor and the bracket.
Do you have your own 3D printer?
Last edit: Anonymous 2018-12-30
Yes, I have a tevo tarantula, with a lot of modifications, once setup its pretty reliable and prints are very acceptable.
https://sourceforge.net/projects/my3dprinter-tevo-tarantula/
Last edit: brownrb 2018-12-30
Much appreciated! I'll keep that in mind for future parts. Thanks.
I took a while trying to figure out why the motor was not behaving properly. It was skipping steps, buzzing and shaking in place (but not rotating), and stopping early. It was also just vibrating sometimes but not rotating. I tried changing the motor and the board and neither was the culprit. I also tried different power sources. No help. I was growing worried that this motor was not going to work because of some "weakness" in its rotation (I hadn't used this stepper before and now know that it wasn't working anywhere close how well it should).
I was about to give up (until using the printed PCB when it arrived, which I figured could solve the problem- no sense in troubleshooting a prototype too heavily) but decided to do some continuity testing on the circuit and found that one of the jumper wires had literally pushed back through its plastic casing while being pushed into the protoboard, meaning one of the 4 channels into the ULN2003 was literally disconnected! What is odd is that the motor still attempts to work, although unsuccessfully, if this happens. Hopefully this will help someone else in the future with this motor.
Here is the working attempt before the PCB came in. Pardon the blurriness, apparently my phone can't choose a sufficiently quick exposure time.
Last edit: Anonymous 2021-02-04
PCB finally arrived from Aisler (very, very high quality boards, although not very quick arrival from Germany: over 3 weeks after ordering). I recommed not only soldering female sockets to the arduino's 30 pins (so it can be removed from the board as desired), but also with the LEDs, switches, power, driver board and bluetooth module. This saved me multiple headaches. Here's a closeup of this strategy:
I'll admit, my biggest fear was getting the bluetooth module working to spec (I've wrestled with items as simple as bluetooth headphones on Windows and was not looking forward to this), so I tested everything before adding the HC-05. Note that if you intend to try this first, you must comment out the bluetooth portion in the firmware file (and any other features you won't be using) You will have to reflash the firmware for each element you decide to add as you go. Here is everything save the bluetooth module.
Once it was working via USB cable, I attached the bluetooth module accordingly and to my strong surprise, it operated perfectly the first time. If you're on Windows and aren't so lucky, power off the focuser after flashing the firmware with the bluetooth functionality uncommented, reset your computer, turn on the focuser, go to "add a bluetooth device" and choose to place in the object's passcode, which by default is 1234 for most of these. When added, the drivers should be installed automatically and a virtual COM port should be added. Check in the device manager for it under "ports (COM & LPT)" if you're on Windows 7. If you're on Windows 10, I'm sorry to hear that! Note the COM port number for this connection
(standard serial over bluetooth link) and use this COM port in the Windows application for the focuser to connect. You may need to update this COM port number in the ASCOM interface so it points to the new wireless location. From there it's plug and play! Here is everything combined:
Last edit: Anonymous 2021-02-04
I added an LED for the power indicator (note that the PCB for the power indicator red LED does not have a resistor and one must be added to the LED above the board) and drilled the hole for it, as well as screwed the power supply circuit, PCB and ULN2003 board into the hobby box.
I decided that I'd like to have the bluetooth antenna of the HC-05 protrude from the box to give it the cleanest signal to my laptop. I did tests and had successful connection to the machine even from upstairs, traveling through a ceiling and multiple walls! This is better performance than my bluetooth headphones even get. Regardless, no reason not to have the antenna exit the box, which I accomplished by cutting a slot:
I debated awhile on how to shield the 7808 chip's high surface temperature from any wires that would find their way onto it, so as to prevent melting. It dawned on me to use PVC pipe to physically shield it from other wires! Without contacting the PVC, the chip will not melt it but should also stop wires from touching it.
Last edit: Anonymous 2021-02-04
Here's the completed project, using a power LED, Red/Green LEDs, push buttons, power switch, fuse, and bluetooth:
And the 8se OTA still fits in my case! Glad I went with the 28BYJ-48 motor instead of the giant and heavy NEMA choices.
Helpful information and field tests to come. Thanks Robert!
Last edit: Anonymous 2021-02-04
Useful notes:
~ The firmware (as of 12-29-18) entitled "Focuserv277_ULN2003H" is the correct choice for the full option, full-stepping implementation of the ULN2003 board with the 28BYJ-48 motor; the H simply implies this firmware is capable of fractional stepping.
~ Coil power will cause a continuous draw of roughly half an amp with this motor and driver board, because the motor is constantly maintaining its position in fractional steps. Although I didn't need this, you'll have to account for the heating of the power supply and motor itself (I placed thermal paste between the motor and metal mounting bracket- so it can act as a heat sink- to help this). Best not to use it if you're full stepping, unless there's another reason to use coil power that I don't know about?
~ The bluetooth signal (with the antenna not enclosed in the box, is strong and solid more than 15ft away, even through multiple walls. I can't move far enough away from it indoors to lose connection! Once the COM port is assigned, AstrophotographyTool (APT) recognizes and operates it easily!
~ Neither Astrophotography tool (APT) nor Sequence Generator Pro (SGPro) allow wireless or WiFi connection to Nikon cameras, in any way, as of 12-29-18. Some Canon models can utilize EOS tool to trick it to work. Boo on you, Nikon software. Therefore the USB cable to the camera will be necessary.
~ The Nikon d3000, d3100, d3200, d3300, d3400, and d3500, among other introductory models, do not offer Software Development Packages (SDKs) and therefore cannot work with these programs in many ways, including bulb and live view modes. Some are not workable altogether.
~ I see why many people have been using gears and pulleys now. Although nothing was easier than my simple mounting bracket directly to the focusing shaft, this prevents the camera from attaching at prime focus to the back of the scope in roughly 120 of the 360 degrees it could before. Look out for how the camera will mount to the scope when designing the mounting bracket. I will see how long I can go shooting at the new allowed angles, although some large targets don't have much freedom in the angle at which the camera is mounted since the sensor/aspect ratio is rectangular and the tube cannot rotate. A few workarounds I'll have to consider include redesigning the attachment, or extending the connection between the camera and the OTA so it clears the focuser bracket when rotating the camera body to those angles. The latter method is always one I try to avoid, however, to assure there's enough focus travel! Moving the imaging focal plane always risks these problems. I'll see if I can get away with the current configuration for awhile, however.
Last edit: Anonymous 2018-12-30
Adding the extra tube between the camera and telescope (to extend the distance from the tube it reaches and thereby eliminating the focuser being in the way when I rotate the camera at different angles) proved to be a non-issue; I can still achieve focus with plenty of extra room. Here is the before situation, where the focuser was in the way of the camera if I rotated it to certain angles since I didn't use pulleys or gears:
You can see the camera wouldn't be able to rotate further because the motor is in the way. Adding the following extra distance (tube) solves the problem, and still allows for focusing on my 8se
I can now rotate to any angle with the camera without using any pulleys or gears on the focuser.
Last edit: Anonymous 2021-02-04
This focuser is awesome. I've tested it successfully on AstrophotographyTool (APT) and Sequence Generator Pro (SGPro) and it works perfectly on both. I'm super impressed with the bluetooth connectivity with the HC-05 module too; on my machine of choice, I get connectivity through the wall inside (when the scope is outside) roughly 30-40ft away without any disconnection (except if the program itself crashes or has a problem) for hours. I haven't needed "coil power" and therefore this thing will not be drawing much power at all. I did have problems getting the same range on other machines, but I'm fairly sure that this is because of outdated hardware or non-updated drivers on them, although I don't know for sure.
Having a focuser has served two different purposes. I wanted it so I could run auto-focusing algorithms that would compensate for the mirror flop I've experienced on this scope; many nights the focus would become unusably bad after under an hour of imaging, and unfortunately this is due to settling of the primary mirror on its own tracks, not the knob moving. I wanted to periodically correct for this to keep me in focus the whole night.
However, I have found that the algorithms employed by SGPro in particular are so accurate in achieving overall focus (they collect stars thoughout the image and find a global average HFR) that I found the results better than I could even get by hand. I would spend 15 or more minutes at a time trying to get focus manually and it wouldn't even settle there. However, this programs leads to focus tighter than I could ever get by hand. Here is a comparison, showing an attempt at M42 I did many months ago vesus running autofocusing routines on SGPro the other night:
Clearly these stars are tighter and much more focused.
Last edit: Anonymous 2021-02-04
Hi Alex
Great shots. Nice to hear things have worked out. Yes motor stuttering is often traced to one faulty stepper wire not making connection.
Awesome writeup, many thanks for your valuable and most intersting contribution
Regards
Robert
Where are you located?
You as well! This build kicks butt.
I'm located here in the States. Lots of snow right now but hopefully more opportunities coming up!