Last update to the CMoy charger.

One last update to the CMoy charger, that I didn't publish earlier. I have stopped updating the power supply section, and I have moved on to other areas. My current research and interest is in the Li-Ion/ Li-Pol battery charging side of things. I guess this is only natural. I have already done a few plans, and I am in talks about publishing some of them with a magazine as a short series. So here is the last image:












In about a months time, I will start to publish some of my designs for charging the Li-Ion batteries using USB on pc's or chargers.

As I previously mentioned, I will be uploading an other schematic. So without further adieu, here it is.












I have shown the original work in progress version, that I posted in my previous post, to Chu Moy at Headwize, that created the CMoy, and was happy to see it and encouraged me to post in head-fi. I hope others will appreciate and like these variations too.

Now, I will explain a little bit the circuit. I researched for about a week for the most suitable diodes that would suit this design. The main objective was to have the lowest Vf @ 100 mA. Well, I thought it would be an easy task, it tourned out to be extremely time consuming.

I ended up with the SB220/240 (not the SB120/140) series in my original design. Now when you are on fire, and your design is not 100% finalised, ideas still tend to creep in to your head, ok even after 100%, there is always something you are not happy with. I had found the SR series by luck on the internet, until then I knew the Germanium diodes were better... for lower Vf. Well that is true, up to a point ;-) they are great in crystal radios for one reason., VERY LOW I or mA say, can someone say 1 mA and under?... yep! when I started viewing their datasheets, I was amazed at how low the Vf was, until I seen finally the charts for Vf against I. Then they did not seem nice!

Your best 'run of the mill' diode that you can find everywhere, is the 1n5817/8/9. Its Vf is: typical/maximum at 0.1A (100 mA) is ~0.28 / 0.34 for the 1N5819. I did start with the 1N5817, since I was looking at only one doubling stage with a '9v' Li-Ion 600mA battery. Then I added the other stages and safety margins and ended up with the 40v parts.

The SR and SB series are from Won-Top Electronics Co., Ltd., a Taiwanese manufacturer. Their products are available from Digi-Key and others, but I got some SR-140 from eBay due to the postage rates that Digi-Key charges on international orders. I w aould have loved to use the SB240, and yes I know they have free shipping at Digikey, but only for orders of $200 or more. Maybe another time.
As a side note, the SR sereies has been replaced by the SB series.

Now as to my comments previously about lasting at least 10 x as longer. It is simple, use 3000 mAh for NiMH batteries (AA size) or 1 x 18650 for Li-Ion at 3800 or 4000 mAh. The TC962 uses about 510µA-700µA. Lets say to make it easier, 1 mA x 4 = 4 mA, plus the LDO,

Now I do have a few other variations that include USB chaging of "Apple" devices as long as you have a USB cable to attach. A simple circuit would be the following:
Altoids Mint Tin Project - Minty Boost - USB Battery Charger. (and yes, there is also YouTube too).
This would be useful for many people. You can incorporate it in your own design and add a switch when you are to start charging.

I would recommend people to buy "Protected" Li-Ion batteries and chargers, as it will save them a lot of headache if they were unprotected and they deeply discharged.
What "protected" batteries are, is that they also contain a small circuit board inside the battery case.
It protects them from overcharge, discharge etc.
The only hard part for the 18650 is finding holders, of which are easy to find in the U.S. through Digi-Key etc. just search for 18650 battery holders. As an example here is a direct link.
Mouser has a similar policy with Digi-Key ;-)

Now, back to the design.

Capacitors, always use larger voltage ratings than what you are using.
If you are tight on space, in the 1st stage, you can use 16v, since it will not see that much, second stage, 25v and third 50v.
If you want it to look all the same, just use 50v versions. Try and find very low esr capacitors or use tantalum.
I do not have an osciloscope to check what is happening with the caps etc. I do recommend to ground pin 6 ALWAYS, so it is at 24 kHz, so it is out of the audio frequency.

Thank you for reading. I hope it has helped other people to get them thinking of designing the power section of their CMoy.

Btw, I recently came across a great page for a CMoy with Bass Boost. It can easily be altered to make an on-off switch for the BB. :-)

Here is the link to the page:

http://diy4fun.blogspot.com/2010/03/cmoy-for-altoids-box.html

He also has schematics and all.

It is a very interesting design.

Until next time.

I will be posting a link to this blog on www.head-fi.org for more peoples opinions. Just go to the Forums section and from there to Misc.-Category Forums and then D.I.Y.

CMoy power stage

Hello Everyone,

A quick post. I have found some time to play with the CMoy power stage.
I have been in contact with Chu Moy, as in the inventor of the CMoy headphone amp original design.
The design is based on using a SINGLE Li-Ion battery or two!
It will run at least 10 times longer compared to using a normal "9v" 200 mAh Ni-MH battery and also run a little more "better" due to the larger voltage available and mA too.

I aslo have a circuit that will be for charging the battery(-ies), although I have a few versions.

There is one for Li-Ion/ Li-Pol and another for LiFePO4 (LFM). The later is MUCH safer, but I am not publishing any of those just yet, due to that they require you to be able to handle QFN packages (for my prefered IC's). There are a few that are larger, and from many companies too. There are also some Chinese made, that are available in SO and DIP, but I do not like chasing down items that are hard to find.

I will present two versions for now, a simple one and a slightly more 'advanced' version with more items to buy.

As for BOM, please write them down from the schematics that I have supplied ;-) as I will not at this stage do that, maybe later.

Enjoy.
Boris

p.s. At this pointin time, I have been unable to assemble the below schematics, but they should work AS-IS.
p.s.2 If you are to use any of these for commercial benefit, then a small royalty will be expected.

• One-OFF: only for personal use will be exempt! :-)
  

p.s.3. If you have found these schematics useful and would not mind donating a small amount because of your excitment, or for whatever reason you feel generous, please leave a small donation.

V6 as shown to Chu "CMoy" Moy.


This is v7. The "simple" version.
















and an updated version to the above is this that gives you two options for input voltage, if under 9v use the left, if more and upto 16v the use the right input:

I will soon be publishing my slightly more advanced version thatwill be more precise and can still work when you reach the battery right down to where it shouldn't go... I will be posting this by end of Sunday here.

I have some information in the schematics for part selection. Please do NOT use 18v all the time as that is the maximum rating for the part and it can degrade after prolonged use at that voltage. A safer voltage will be 15v. That is what my next schematic includes.

It will be different, since it will only use 1 or 2 batteries either in series or parallel for Li-Ion/ Li-Pol/ LFM. You would also be able to use 3-4 Ni-MH batteries in series with my next version.

It is important to use the recomended diodes if you are using up to 9v, since when doubled, it will be just under the 18v (17.4v) if you are at exactly 9v, else you will have a much bigger drop.

WARNING: BE SURE TO READ THE SPECIFICATIONS OF YOUR OP-AMP! SOME ARE SINGLE VOLTAGE, OTHERS ARE NOT RATED UP TO 36V!!!

PROCEEDING ON IMPLEMENTING MY SCHEMATICS WILL BE AT YOUR OWN RISK!!!

If anyone wants to read up on the TC962 they can click the link or go to Microchips website. Here is a link to it on Microchips site: TC962.

Part 2 - Updates & Progress

Hello Everyone,

Finally an update to my previous postings.

I have started constructing my clock after shelving the project for two years.

I have gone the way of modular design. What that means is, that the most important items are in modular form, or compartmentalised. Basically, every section is on its own board.

I have made a simple voltage regulator for 5v (LDO).

I have made the timing circuit on a separate board, the display, on its own board, the logic on its own.

The buttons on their own board with their own MC14490 ic too.

As for the boards I used, I will mention at a latter time, since I am expecting confirmation from the manufacturer to get the OK to mention their name.

In the mean time, enjoy some early photo's. These will be published shortly.

Schemaitcs - Pt.1

Hey :-) How are you all?
Well today I have a small update. I have hand drawn the clock pulse circuit that gives 1Hz. Eventually, I will be drawing the rest of it too, just don't ask me to have it done by tomorrow ;-) It will be added - slowly.

Enjoy the diagram.

This is for a 'solderless' board, where the x's is the centre of the board. It was done on the smallest one, as you would have seen in the video. At the bottom are the two power strips for ground and +5v. Alternately, I hope it is clear enough to remember that the power strip is NOT physically connected, and that if you DO use this layout, it is NOT optimised and you will need to break the circuit along the whole length at the bottom between the ground and live. I did NOT draw it with a pencil, so I can not re-do. It is very easy to do wiring for the power ;-) Easiest is the earth that can be done on the top side and the +5 on the bottom :-P hehe. You will only need on ground wire pin 12.

As for the values, those are what worked for me, since the crystal I have is picky, it did NOT work with a variable cap, so I tried various combinations of what I had at hand. The crystal is a 32768Hz watch style, very small cylindrical with short thin leads.

In case you are wondering what the 1 of 1 is, it is due to the printing of the image through a browser.

The I.C.'s are of the 4000 series. I used the numbers on the chips themselves.
The resistors are 1%, but not exact value according to the multimeter.
Enjoy and have a nice time.

Update to blog.

Hi,

For those interested, I have added the links on the right to the following electronics Manufacturing companies.

Texas Instruments
Maxim / Dallas
National Semiconductors
On Semi (Motorola)
Microchip - Mostly known for their PIC micro controllers.

I will be adding a couple of more at a later time, but not many.

To follow are:

Philips (NXP)
Fairchild
and a few others.

The counter becomes a clock

Hi everyone,
Finally, I was able to have some time to continue slowly on my clock project. Up until recently, it was actually a counter from 00-99, incrementing every second or in its original form, manually. I was finally able to add the 74ls08 to it, so it would reset after the second '192 reaches 6, the B&C output that represent the binary 6 are on the input of the AND gate, of which resets the second counter. For those that don't know, six in binary is "0110". You can see the "truth" table in my previous post.

These images are pre-74ls08
Yes, at the moment, I am using a battery pack. The 78M05 does not have any additional circuitry, like capacitors etc. Its output goes to a special chip (TPS7350 - up to 500mA continuous 8 pin DIP made by T.I.) that regulates to exactly 5v.
On the small board is the timing circuitry. On the larger one, is the power and logic. The single digit by itslef is an INL0394 by Innocor that is pin compatible to a TIL311. The other two are HP 5082-7300. The Innocor is still available and the HP has been updated and is still available too. Both these parts are fairly expensive, since they also include an I.C. that has the same functionality as a BCD to 7-Segment driver. I will try to use a different camera and settings next time, since the images do NOT do justice to the displays. They are much better looking than normal 7 Segment displays.

The oscillator may change later for a device from Dallas-Maxim that gives 32768Hz, and much more stable, or it is suppose to be. Here is a link for it. The part number is: DS32kHz I will try it, if and when I receive a sample from them, since I can not find it here easily in Australia.

The circuit works something like this.
On the big board.
Power->Bring down to under 10v. ->Power 5v. ->(Original circuit up down manual counter)->MC14490 debouncer for switches. 3 switches. Black reset, White increment, yellow decrease.
Small circuit.
Power in-> CD4060 (could be any type of the 4060)->crystal 32768Hz (or 32.768KHz)-> with support circuitry->from CD4060 2Hz out -> CD4013->1Hz out-> to large board to the UP input of the 1st 74LS192.

The 74LS192's are connected in such a way, that if needed, they can also count backwards (down).

In the later stages, the circuit will start being built on proto-boards and small ones of those. I will update with pictures again, once that stage is reached. At the moment, I will add the minute and then I will move the project over to new breadboards with a neater layout. Once that is completed, I will post pictures again.

For the moment, I will be taking it easy, due to my lower back problem. It has gotten worse, and now I have been diagnosed with "L4/5 central spinal canal stenosis". So i must be relaxing.

Until my next update.

My Electronic adventures - Real world.

Hi everyone,
It has been a long time since I updated my blog. Well, let me just tell you, that it has been a very peaceful period, nothing exciting. Now for those very few that know me personally, and for many years, would know that I like tinkering with electronics.

For those that are not familiar with my habits, I should let you know that I go in circles. I might like something, start doing what I like, get bored quickly, and then after some time, I will cycle back to it. Well, it has been a few years now that I had left my hobby electronics. At the time, I was looking in to making a digital volume control, that would include an indication of where the volume was at. Needless to say that it was left in the early stages. I had progressed as far as a reset button, up / down control, single digit display with a TIL311/ INL0397 from Innocor, I actually used the later, and an ON-Semi MC14490 switch de-bouncer. You may say, why did you use that, simple, ease of use, less components.

Below, you will find a guide I wrote for myself, like a journal, so I will know what I am going to be doing. The part about the clock, was entered on the 21^st of June, 2008. I was out all day today, so hopefully I might be able to take some pictures tomorrow.

Take care, everyone and have fun.

Description:

This project will have the following parts divided:

0 stage:

Power supply.

This will be provide by a back for revision 1, by a 100v DIP package bridge rectifier, model 1B4B42 DIL 1A, or a W04 1.2Amp.

Since these are a 1A & 1.2Amp models, they are adequate for all 6 digits, 7 segment displays. Some W04’s are rated up to 2Amps. Please make sure about what you use . A 78M05 voltage regulator would be adequate, just, for the Version one. To be sure though, I would advise a proper 7805 that is 1 Amp. Some “brand’s “ claim that they are up to 1.5A. I can not comment on that, and I will not really want to just try it out for the sake of it. Rule is, if it works, does not get hot, you have proper ‘cooling’ as in USING a heat sink, why experiment? The “L” version in a TO92 package is NOT adequate, since it is only 100mA!!! There are some similar ones that are 250mA. Those are still not adequate… Just use a larger one and get over it. Of course, you can use an LCD module, that works on multiplexing, but I am not looking at it at this moment, but it will drop your power requirements considerably. Unless you have a few of those and want to use one per 7 segment LED! Go ahead, make your life difficult. ;-)

4-bit numbers

Binary D C B A	Decimal	Hex base 16
0 0 0 0 0 0 0 1 0 0 1 0 0 0 1 1 0 1 0 0 0 1 0 1 0 1 1 0 0 1 1 1 1 0 0 0 1 0 0 1 1 0 1 0 1 0 1 1 1 1 0 0 1 1 0 1 1 1 1 0 1 1 1 1	0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15	0 1 2 3 4 5 6 7 8 9 A B C D E F

The table on the right shows the 4-bit numbers and their decimal values.

The labels A,B,C,D are widely used in electronics to represent the four bits:

• A = 1, the 'least significant bit' (LSB)
• B = 2
• C = 4
• D = 8, the 'most significant bit' (MSB)
  
  

Binary Coded Decimal, BCD

Binary Coded Decimal, BCD, is a special version of 4-bit binary where the count resets to zero (0000) after the ninth count (1001). It is used by decade counters and is easily converted to display the decimal digits 0-9 on a 7-segment display.

Several decade counters using BCD can be linked together to separately count the decimal ones, tens, hundreds, and so on. This is much easier than attempting to convert large binary numbers (such as 10110110) to display their decimal value.

Do not confuse BCD which stands for Binary Coded Decimal with the labels A,B,C,D used to represent the four binary digits; it is an unfortunate coincidence that the letters BCD occur in both!

When the tens of seconds reaches 6 (0110), the output is also connected (the B & C ) go to a 2 input AND gate, the output from that, goes to the reset of the one that just reached 6 so it goes to Zero, and is also connected to the Clock input of the 3^rd ls192 and increases the count by 1.

Now, since we are counting in 24hour format, to get that to reset, we will need rom the 5^th ‘ls192 the C output (4 = 0100) and from the 6^th ls192 the B output (2 = 0010). So, when both are active, it will reset the hourly ls192’s.

Included is a day counter.

When the 24h pulse is given, it increments the day counter. Once it reaches day 8, it resets again. So the pulse from the 24h plus the pulse from the number 8 (1000) will reset the counter.

Version one.

This will be the most simplest of them all. Just seven LED’s of either 3mm or 5mm, depending on what you want to use, or even rectangular 2mm x 5mm x 7mm ones, what ever takes or makes your fancy. Also, six 7 segment LED’s for the time.

Version one point one, this will use the TIL311 from TI/ INL0397 from Innocor (now JDSU).

Version one point two, Instead of the TIL311 it will use the HP 5082-7300.

Notice: Both 1.1 & 1.2 use an internal integrated BCD decoder and have their own display, that way, one less I.C. is needed, no need for a BCD to 7 segment decoder. The downside is that you can not create your own display scheme, only the normal 0-9 & A-F where applicable. These display’s average around the 10$US price range up to 30$US EACH. You may be lucky and find them cheaper, just like I was with the HP version. They have a different orientation, package wise. The pins are either vertical or horizontal, what I mean is that the pins are either horizontal to the body or vertical. The TIL311/ INL0397 are vertical and on the HP are on the top and bottom, so they are horizontal. This is the version I have started with on the bread boards.

Version one point five, might exist, if instead of using normal 3mm or what ever LED for each day, I was to use an Alphanumeric Segment display, to show the day of the week. This is very possible to be made.

Version two, will have a considerable more higher power demand! 6 digits for the time, 3 digits for the day of the week, 3 digits for the month and 4 digits for the year. So that makes it a total of 16 digits. At an average of around 1.2mA – 2.5mA per segment, that can get to be a fair amount if you want to try out all the segments of all the LED’s at the same time (for test purposes of course…). But in normal use, either of the two should be just adequate. The leap day will need to be done manually.

Version three, if it reaches to there, will have a logic layout change to compensate for leap years. It will have a start year on a leap year, and a counter, of which when it reaches 4, it will trigger the leap date. That way, when the person is fixing the year, it will know from where to count, and adjust February accordingly. That will need a lot of AND gates and other logic. I may never reach there, even if I have the thoughts in my mind at the moment. So it will have a combination of input for the YEAR and month, and if both are together, then when the date is added, it will add one day extra and reset it to go to March.

Version four, will have an alarm function. It may have rotary BCD switches to encode the alarm code, so when the time is reached, it will go off. This may be implemented before version three.

1^st stage:

Pulse and clock.

Pulse generator. 1 second increment. This is the heart of the project. It will provide a pulse every second to the decade counter an 74ls192. 74ls193 binary counters can also be used in conjunction with AND gates, so when they reach ten (10), they will reset and provide the pulse to the next counter. Once the second counter reaches 6, with the use of an AND gate, it will reset the seconds counter and send a pulse to the minute counter. Again, the logic is the same.

Now for the hours, there is also an AND gate needed, so once the decade counter reaches 0, it will pulse the second hour digit, so once both read 2 & 4, it will reset them to 00 and then will also send a clock signal to the day counter.

2^nd stage:

The Day counter.

The Day counter will receive a pulse every time the clock reaches the 24 hour. The pulse that resets the two digits displaying the 24 hour, will also send a pulse to the day counter.

This will be a 4017B decade counter that has direct output for LED’s to be driven. At this stage, there are two options, either to have seven independent LED’s, with the colour of your choice to represent the day of the week, or have an Alphanumeric display to show the name of the display. In revision one, it will only have LED’s for day of the week, and the project will conclude here.

3^rd stage:

Revision 2 only.

This will advance the circuit even further. It will be getting the pulse from the 24 hour reset. It will not use the 4017 as the heart of it for the months, but a 74ls193 binary counter. The 4017 will be relegated to another choir. It will be used for counting the days (x2) and the year. The ‘193 will be counting the months.

This part will be fully alphanumeric displays, 14/16/17 segment, so it can display letters correctly.

Progress:

I have already started construction of this project. I have bought some rectangular LED’s plus the HP display’s from eBay. One of the later things I bought, was another pair of bread boards since the two I have would not be enough (ok, I hate to make it very tight… on the bread board, and was starting again an old project I had that was a counter, I did not need to change anything, except the voltage regulator to a stronger one. I already had an MC14490 switch de-bouncer for manual counting. So all I did was add the HP I.C.’s and leave the Innocor in place as a reference. It is getting very tangled now with the seconds display. I will now need to put my 1^st AND gate. I have finished doing my 1Hz clock. Using my Multimeter that also has a Hz counter (4 digit resolution), it shows that I am doing well. 32.76KHz (it can not display the whole 32768Hz (or 32.768KHz). As for the seconds pulse, it is NOT 1Hz, but 0.999Hz for some reason. That is good enough for me, since I do not have any fancy electronic devices! Only two Multimeters and components, oh yes, and some screw drivers. Although what I do have, is a lot of love for these things and passion and a lot of hair for pulling out ;-) –NOT! I want my hair, hehe.

Saturday, 21^st June, 2008

I must say, I had a hard time getting the crystal to give me the proper frequency. I needed to see a second design, of the same thing to finally clear it up in my head and get the whole thing working! I thought I actually had a faulty crystal, since it was giving me a smaller frequency, but all it was, was my wiring. First time after the second implementation worked right away! No need for a trim cap for me. I will be taking photo’s at a later time, like later today or tomorrow to add to this blog.

Sunday, 22^nd June, 2008

Over here, you will find a great article on how to do Nixie clocks. There are actually so many of them now on the web, but this one goes in to great detail for many things. Anyhow, this is where I used the part for the timer. There is nothing wrong with it, just for me to be able to conceive it and implement it in a breadboard was not easy, until I came across this one. For some odd reason, I was able to implement his layout easier, even though they are both the same thing! If you want, just open the pdf of the logic part (schematic).