Designing a C64 RF Modulator Replacement Phase 1, Part 3

Identify and Source Components

In Parts 1 and 2 of this madness, I decided I was going to design and build an RF Modulator Replacement for my SixtyClone. You should definitely read all that before reading this. And please remember, this is absolutely NOT a tutorial! Just the wild speculations of someone with very little clue what he is doing.

I finished Part 2 with a new schematic, having cut all the bits I don't need from the original:

Now I need to work through every symbol on this schematic, identify what it is from the information provided, and finally, for this part at least, attempt to source the parts from a supplier. It's no good identifying parts I cannot buy coz then I won't be able to build the frickin thing.

In the new schematic, I have, for now, retained the original part numbers for each of the components. This is reasonably important because, if we look back at the original schematic, some helpful information was provided in the notes:

Here, for Diodes (as designated by the D prefix) and Transistors (as designated by the Q prefix) we are provided with the exact parts that we need. Happy days!

Yeah, not so fast there cowboy. Remember, the Service Manual all this information comes from was published in 1992. That's exactly 30 years ago as I write. And as well as being less than kind to my face, time is no friend to the availability of electronic components either.

First things first, I need to remove from those notes the parts we no longer require. This gives us a reduced shopping list for Diodes and Transistors:

1. D1, D2 : MA151K or Equivalent (Diodes)

2. D6        : HZ-7A1 or Equivalent (Zener Diode)

3. Q1, Q2 : 2SC2405 or Equivalent (Transistors)

4. Q5        : 2SC2120Y or Equivalent (Transistor)

The next thing to do then is start searching and first on the list is a MA151K diode. What happens next is a pattern that will repeat itself.

I search Mouser and Digikey for that part and find to my horror that the only thing I can find with that designation is a Surface Mount component.

I wasn't expecting that at all. My soldering skills just will not extend to soldering these. I'm looking for a standard through hole diode so on a hunch I looked back at the service manual, not at the schematic I have been using to date (#251696) but at the older schematic (#251025) and that indicates that the diodes undertaking the same function (referenced as D2 and D3 there) are 1SS119 diodes. If I search for that in Digikey, bingo! These are available, so long as I buy a minimum of 2,929 of them. I shit you not. So obviously thats not happening but I discovered from this that I'm looking for a "high speed switching diode" so searching that exact term provides me with an alternative, 1S2076S7A, which has almost identical specifications and is available in a minimum quantity of 8,795. Good grief.

At this point I just Googled "1SS119 replacement" and found some forums where vague references were made to the 1N4148 diode being used in place of the ISS119. I also found this page, which contained the following:

Comparing specs sheets was hopeless as they really are quite different but I figured I'm not building a rocket here and nobody's life depends on this. So with a wing and a prayer and my fingers crossed they're not too different, I added 1N4148 for parts D1 and D2.

There is very little entertainment to be had from my relating how I went through the rest of this list. I basically repeated the exact process detailed above with all of the Diodes and Transistors till I finished up with an alternative list that has more than a good chance of being mostly garbage.

1. D1, D2 : 1N4148 (Diodes) from Digikey

2. D6        : 1N4582A-1 (Zener Diode) from Digikey

3. Q1, Q2 : 2SC1684T (Transistors) from Ebay

4. Q5        : 2SC2120 (Transistor) from Ebay

At this point I decided to create a list on Digikey as I decided that was where I was going to buy everything (except the transistors which I could only source on Ebay), for no other reason than they seemed to have more in stock than Mouser. 

With the more complex components dealt with, time to move on to the "easier" stuff.

The search for Capacitors was reasonably straightforward - the values are all present in the schematic and unless otherwise indicated, I assumed I needed ceramic. The only thing I needed to watch out for was the voltage rating (I got 50v) and lead spacing, i.e. the distance between the legs as this will be important when we come to design the new PCB. They are all 5mm. There is one electrolytic capacitor at C25 and I just sourced a small 330µF radial capacitor and guessed that a 25 volt rating would be sufficient.

It was a similar story for the Resistors, but here I am going to try both Carbon and Metal Film resistors, just to see if either one makes any difference at all. As it's not specified, I have guessed that I need ¼ watt resistors because that is what populated the majority of the 250466 board, but I have absolutely nothing else to go on, and the original schematics don't advise.

In a potentially embarrassing attempt to get to grips with some of the voltages on this board, I did some more Googling. I very quickly established that our Power Circuit is the well recognised Transistor Series Voltage Regulator.

You can see from the above diagram that I've simply taken our existing schematic, cropped out everything except the Power Circuit and shortened the ground wire to make it visible. I've also added 3 labels:

VOUT     =    Voltage Out
VBE       =    Base-Emitter Voltage
VZ         =    Zener Voltage

If I'm understanding the principle correctly (big if!) then this circuit takes the unregulated 9v input and by using the unique properties of the Zener Diode, creates a reduced, and regulated output voltage. It seems that calculating the Voltage Out in our circuit is a simple formula:

VOUT = VBE - VZ

To obtain the values for VZ and VBE I needed to consult the data sheets for the Zener Diode and Transistor I chose (1N4582A-1 and 2SC2120 respectively). If I am reading them correctly, then the VZ is 6.4V and the VBE is 0.5V (min) to 0.8V (max) which gives us a VOUT of between 5.6V to 5.9V. As the transistor is the one specified by Commodore, and the Zener diode has very similar values to those originally specified by Commodore I can only hope that this is in the ballpark of what they intended! We can attempt to test this later.

In a similar vague manner, when it comes to the two inductors at L1 and L2, the schematics do tell me I need a 22µH inductor at L1 and a 10µH inductor at L2. The only other thing we can glean from the schematic is from the symbol:

If I go back to Google and search for the term "inductor symbol" I find this page, with a graphic that seems to suggest this might be a ferrite core inductor. The symbol used by Commodore 30 years ago has clearly been superseded so again, I'm winging it. I have zero clue what the rest of the values (current rating, resistance, frequency) need to be and despite attempting to understand the current in this circuit (the hfe of the Q5 transistor has something to do with it) I'm just not understanding it, so I simply sought out: through hole, axial, ferrite core with the indicated inductance and hope that they work but honestly, this is all blind guesswork.

I suppose at this stage one might reasonably ask why I haven't sought out direct advice on various forums to help with this sort of thing? And my answer to that is quite simply: if I want to do this, I really need to put some effort in myself. How will I learn anything, if I'm spoon-fed solutions? Perhaps when I've tried a few things and it's failed I will seek out advice, but for now, let's just wing it and see what happens.

So. I now have a schematic and I now have a list of all the components I need to put it together. The next step is to prototype the whole thing, and to this end I will need a few more items:

Once all the components from Digikey arrive (they are presently on the way from the USA as I type) we'll start the first prototype. Read all about that in Part 4.