Monday 17 January 2022

CNC Activities

 While the blog may have been a little inactive, the bench has not. Lots of things on the go and thought I’d bring the blog up to date.

Since my last etching a board entry I started to have problems etching boards. I never did figure out what my problem was, but it was very frustrating as I did have this process down pat. No matter what I did I could not get the toner to transfer properly onto the pcb.  In the end I decided that it was time to explore the cnc world.  It was something I had wanted to try out for a while anyway.

I ordered the Prover 3018 along with extra bits and a spoil board.  Assembly went well and naturally took a lot longer than what the instructions said it would.  I was very pleased with the quality of the 3018 and it was a pleasant build.

The learning curve was fairly steep in order to get to actually milling a pcb. Between figuring out how to use 2 new software packages, Flatcam and Candle, and then all the different settings and how they work together, it was several days before milling a board that I could use.

Briefly, the biggest mistake I did was starting out to mill a SC70 size footprint. My thoughts here was if I figure this size out the rest come easy.  Once I went to the larger smd sizes things went much better.  After I learn the in’s and out’s the SC70 size will be next.

The other main issue was in Candle.  The program that uploads the g code to the mill.  It took me hours before I figured out which button to press to get the milling process to start.  Uploading the g code went fine but then the mill would stop. I spent a lot of time learning the g code and viewing videos and basically go nowhere.  Turns out after you hit the send button the mill starts up and goes to a safe position and stops. You then have to hit the Pause button to start the process. Who’d have thought you need to press pause to start things?  Maybe its me but I don’t think so.

I’ll save my adventure with my DCV Frequency Data Logger project for the next post.



73, Peter


Monday 22 March 2021

Arduino DC Volt and Frequency Meter with Data Logger Functionality

 

Having on several occasions the need for another dc voltmeter, and while owning a really good one and a cheap one that was at best intermittent, it was time for another decent meter.   I thought why purchase one when I could build an Arduino based voltmeter with parts I have on hand. Thus born is my latest project, The Arduino based dual input DCV meter.

The project however was quickly a victim of feature creep. Originally just dc voltages, I decided to add another input for frequency and add in a data logger function. 

This came about after a few years of completing several pages of data from characterizing various IC’s and circuits and then entering that data manually into a spreadsheet for further analysis. 

The present specs for this project are:

Input 1 and 2, 0-15vdc

Input 3, up to 30mHz

Data Logging to export data into a spreadsheet automatically on demand

The DCV function went fairly smooth in the scheme of things.  Originally, I was playing around with the idea of using an opamp to invert a negative voltage on one of the inputs for reading negative voltages. I determined however that this was not a practical function and while it gave me a better insight to opamp behavior, the idea was abandoned.

The data logging routine went well and after a couple of days of hacking Arduino code I was able to get the dcv being measured exported to a spreadsheet via the serial port to my computer.  At first it was a constant stream of data and I put in a pushbutton debounce routine that allowed me to send data one line at a time at the push of the button.  Perfect.

Now I can turn my efforts to figuring out how to use the Arduino to read frequencies up to 30mHz.  I have the code working now up to about 4mHz.  The need for frequency divider circuit is apparent and I have never played around with these type of circuits before. 

Spending the last few days watching videos, reading what I can on the subject, not really understanding too much of it yet, I put in an order for a few IC’s that should get me what I need.  The practical experience will help with my understanding of these circuits. 

Along with counter IC’s I ordered a couple of Schmitt triggers as well. Mainly because I have seen where they are sometimes used in these circuits and since its all new to me and they are inexpensive, why not.  Should be here in a day or 2. With any luck I hope to have a working circuit by the weekend.

Another thought I have had lately is to get away from using the Nano board. While they are cheap and convenient, I am thinking about just adding in the couple of components required directly onto the pcb along with the rest of the circuit.  I have breadboarded such circuits years ago using the Atmega 168 IC so I know I can do it. Not tried the 328 though. I have a couple of 328 chips doing nothing so this might be the project to do this.

73, Peter

Monday 21 December 2020

Etching a Breakout Board for SC70-6 and DFN-6 Footprints

 

A while ago I took an interest in experimenting with digital pots and caps. So I ordered the MCP4018 and the NCD2400.

The package size should prove to be a bit of a challenge, especially the NCD2400 which comes as a DFN-6. The MCP4018 is SC-70-6 which should be ok.

While I didn’t have the correct breakout boards for either chip, I could have made do with what I had on hand for the 4018. But there was nothing I had on hand for the NCD2400. So, I decided to try and see if I had the skills to etch my own breakout boards that small.

My 1st attempt (upper left in picture) resulted in a board that while it had no shorts, it was not suitable. The iron was to hot and or I pressed to hard.  My 2nd attempt (upper right in picture) was better, from the perspective of the board quality and I reworked the trace sizes somewhat. The traces leading up to the IC footprint were wider and then reduced leading to the component. The thickness there was the same size as the pin on the chip.  However the board was still not good enough to attempt to solder the chips in place.

My 3rd attempt proved much better. Traces were altered further as well and I made the attempt to solder the chips in place.

The 4018 install went fine, a little crooked but there were no shorts and that's what counts for the moment.  But, the 2400 was a different story.  I tried the tin one pad and reheat the pin and follow up with the rest method, but after a few unsuccessful attempts the trace lifted.

I’m not sure how exactly my next approach to solder a DFN IC will be but I’m still confident I can do it without the use of a solder oven. I do have parts on hand to make a solder oven but who knows if and when I’ll build it.

The MCP4018 is working well and I am experimenting with it at the moment with the UA741 opamp. That will be a posting at a later date.

 


 

Friday 28 August 2020

A Few New/Old Projects on the Go.

 

I am always amazed at how one project can lead to several more.  The latest example of this for me was the Signal Generator project my HB Club undertook last year. 

Once that project was completed, I wanted to be able to control the output. So that lead to incorporating the PE4302 step attenuator chip and how to use it. 

Next was instead of calibrating the SigGen and then selecting an output and calculate the output level, I wanted to measure and display the actual reading.  I also wanted to be able to read at least to one decimal point. This led to exploring an alternative to the Arduino Map function and experimenting with a 16bit ADC, ADS1115 chip. It also has led to utilizing an opamp on the output of the AD8307 and seeing if that can help in reading accuracy especially in the area of the 2nd decimal point reading.

Of course, it does not stop there as now I needed to better understand how I2C communications worked.

With all the different boards laying about and a too numerous to count number of interconnecting cables, test leads, power leads, erratic and intermittent performance, I needed to build a test bed.  So here was another project that this time didn’t contribute directly to the project, but I did end up with a good test bed for any future Arduino projects.

In the Altoids tin is the AD8307 board. Opamp to the immediate left and further left is the ADS1115 and Si5351 boards.  The 2 Nanos, LCDs, ADS1115, and Si5351 boards are all connected via I2C. 

More details to follow shortly about how all this pans out.   




Thursday 21 May 2020

Signl Generator Control Project Update

The software development had been taking up a fair bit of my time mainly because I really don't know a whole lot about Arduino coding and hacking into the existing code for the original SigGen was a bit daunting. Fortunately Kevin, VE3KHH who is doing the same project as I am, has the appropriate skills for the software changes and gave me a few clues on how to approach this. His end of the software changes is completed and I still have a bit of work left to do on the rf sensing code.

I took a break from software to start assembling all the various boards into a repurposed power supply case.

Since the nano for the step attenuator was now redundant, the code is now in the SigGen Nano, I removed the existing pin headers for the nano and installed new ones on the reverse side of the board. I used headers to access the the encoder and the 2 attenuator boards.  I needed to add in a 3.3vdc power supply for the attenuators as the nano was no longer installed. I mounted the regulator right on some unused pins. Unfortunately I let the smoke out of the 1st regulator. I forgot that the tab was also Vout and had soldered it to ground. Good thing I always order several components at a time so I have a few spares.

If all goes well I should be able to disassemble everything and paint the case sometime next week. 




Wednesday 15 April 2020

Change of Pace. Lets try operating cw.

Decided to change things up and actually do some operating instead of building. 

A few of local area hams got together the other night and dusted off older QRP CW rigs and see what happens.  In my case I used an all homebrew station.

My setup consisted of an all hb tuner and audio filter. The keyer and rig are kits. Still hb in my eyes.

CW was rusty, as expected but I did well enough that the Reverse Beacon Network picked up my 2 watts. My signal was heard state side in 1, 8 , and 9 areas.  But the band was very busy and was barely heard at times locally. Need a little more power for the ground wave stuff.

Anyway it was fun and will give it another go but I'll likely use my KX3instead.