A little more about the electronics on my homemade equatorial mount!
I mentioned in my previous blog about my homemade mount that my good friend Tim Duke had been instrumental in making the electronics part of the project happen and I have so much to thank him for! I wanted to give Tim the opportunity to put in his own words how all this came about, the history of the design process and the changes made etc. The following text is Tim’s account written in his own words.
But first – Some of the links on this page may be affiliate links where I make a small amount of money from any purchases you may make after clicking one of those links. You won’t be paying any extra for any purchases you make as the prices all stay the same. This is only to earn a few pennies to help go towards the fees to keep this site up and running.
It was by chance that my involvement in this project came about. Having been a member of Castle Point Astronomy Club in Essex, UK since around 1996, I found myself sat at one of our club nights next to a chap that I hadn’t met before, waiting for the weekly lecture to begin. This was in 2004.
During the half time interval, Matt and I got talking, discussing the Astro kit that we had in the past and present and the projects that we were thinking of starting. Matt mentioned that he had currently got a 10″ Dobsonian, a mount and scope that he had built himself from scratch, but he was looking to make an own designed homemade equatorial mount with the ability of doing astrophotography. Unfortunately, Matt had no knowledge of electronics or how to make the mount drive at the correct speed. “What a coincidence” I thought! I have an HNC in electronics and software engineering, love designing circuits for my own use and love to help others out when I can.
So the agreement was made. I told Matt that I would be happy to design the electronics to help drive his new mount. A few months passed whilst Matt designed the mount and during this time we had our weekly club meetings, where the pub meet afterwards were spent discussing the integration of the electronics, including what could and couldn’t be done. After a few more weeks, Matt was making real progress with the mount and provided me with the sizes of the large wheel and smaller drive wheel. From this, I was able to calculate the drive speed for the stepper motor and 500:1 gearbox that we decided to use. To hold a star in a fixed position in the eyepiece, the telescope must be driven in the opposite direction to the Earth’s rotation. It must also take into account the fact that not only is the Earth rotating, but it is also orbiting the sun. Because of the latter, the rate of rotation is not 24 hours, but 23 hours 56 minutes and 4 seconds.
Due to my knowledge of PIC microcontrollers, studied during an HNC course in electronics and software engineering, I decided to use two 16F84 chips to control both the Right Ascension and Declination controls. Each axis would have speed control in the form of slowing down and speeding up of the drive, via a hand controller. As it was just a case of working out the maths, an extra switch would be added to the electronics to allow a lunar tracking mode (Matt, we should think about adding a solar rate for the next version).
Work got underway to build the electronics, which started with stepper motors and gearboxes. These were purchased from RS components. The stepper motors were of the 1.8 degree variety, so 200 steps is equal to one revolution. In combination with the 500:1 gearboxes, 100,000 steps would be needed to rotate the gearbox shaft once. The ratio of the main and drive wheels on the mount were approximately 5:1, so this meant that 500,000 stepper motor pulses were needed to rotate the scope once.
There are 23 hours, 56 minutes and 4 seconds in a sidereal day. This is the same as 86,164 seconds. We want the mount to revolve once in this time when the scope is set to sidereal mode. So simple maths determines that we need to drive the Right Ascension motor at 5.803 pulses per second (or 1 pulse every 0.172 seconds). This would be fixed in the chip software that drives the RA axis and if some adjustment was needed, this could be reprogrammed into the chip.
The main control circuit has two connections for the motors, one for the hand controller and one for power, it also has a push button for sidereal/lunar rate selection. The hand controller consisted of a small handheld box with five push buttons. Four for direction control (RA+, RA-, Dec+ and Dec-) and the fifth for fast speed selection for slewing the scope. Once complete, the electronics were handed over to Matt for fitting to the mount. Several timing tests were performed and a few iterations of RA software programmed into the chip before we were both reasonably happy with the results.
As you may have seen in Matt’s homemade equatorial mount blog, he had an issue with using the timing belt as a surface on one of the drive wheels. So when he changed it over to the fine grit wet and dry paper, the chip was tweaked again to adjust the timing.
Due to work commitments, lack of time and the fact that we live in different counties, most software upgrades and testing was carried out at the yearly Kelling Heath star party. Refinements were not only made to the sidereal and lunar drive rates, but Matt also wanted a faster slew speed. So tests were carried out to see how fast we could drive the motors. Once this limit was discovered, the values were backed off and set in the software.
One other issue that was discovered was when Matt tried to autoguide. The backlash in the gearboxes caused problems when the motor changed directions. For the RA axis, this was cured by just pausing the motor if it needed to move in the east direction, allowing the star to drift back to the correct position rather than driving the motor and waiting for the backlash to be taken up first.
Over the next few years of use and the experiences that Matt had gained with the scope, I really wanted to correct some of the issues and add further features by producing a MkII version of the electronics. The motor and gearboxes would stay the same, but with a change of control unit and hand controller.
Work started on this new unit early in 2014 in the hope that it would be ready for that year’s star party. My own design brief was as follows:
- New up to date chips should be used, removing the need for some external components, saving space and reducing the overall size of the control unit.
- A built in shoestring autoguide unit integrated on the main circuit board, so that everything is kept together in one place and extra cables are not required.
- A built in four port USB hub integrated into the design. This would mean that only one USB cable and one power cable runs to the scope with a single laptop connection. USB connections for the webcam, DSLR and autoguider unit are all connected locally to the control box with less chances of cables getting tangled or tripped over.
- In circuit programming, so that the chips don’t have to be removed from the circuit board if a change in software is required. This would enable Matt to program the chip himself after receiving updated software from me via e-mail.
- DIP switches on the mainboard to finely trim the RA drive. Four switches would provide up to 16 fine speed settings to allow Matt the ability to tweak the drive without needing to reprogram the RA chip.
These new features were added and the completed control unit handed to Matt in the spring, well before the Astro camp in September. Matt mounted the electronics and immediately discovered a small software bug which was quickly ironed out (it being the DIP switches for the fine control not working correctly). The star party came around quickly and a couple of good clear nights enabled Matt to put the scope and electronics through their paces. Testing went well, and some good images obtained, but with a long focal length scope, star trails were still apparent. The PHD graphs still show tracking accuracies that may be due to alignment, backlash or flex between the imaging and guide scope. Hopefully, further tests will help iron these issues out.
I have plans for a MkIII system. I think Matt and I need to look more closely at the motor/gearbox system used, to try and reduce the amount of backlash present. What I would also like to do is to produce a control box that has the extra features of a couple of computerised dew control ports, along with the ability to trim the drive speed using an LCD panel and buttons rather than the DIP switches that are internal to the unit. This would make the system much more user friendly.
So, what next for the control system? PEC? Goto? Backlash compensation? Have I hit the limit of my design knowledge? Who knows, keep coming back to find out…..