kachew Posted April 20, 2007 Report Share Posted April 20, 2007 dear au,is this what u mean? across means parallel? Quote Link to comment Share on other sites More sharing options...
audioguru Posted April 20, 2007 Report Share Posted April 20, 2007 The 0.1uf ceramic disc capacitor is missing from your schematic.The zener diode and the 0.1uF capacitor are across the power supply pins of the ICs. Quote Link to comment Share on other sites More sharing options...
kachew Posted April 20, 2007 Report Share Posted April 20, 2007 Dear Au,the ceramic disc capacitor is still there....which is parallel with the zener diode ;D ;D Quote Link to comment Share on other sites More sharing options...
audioguru Posted April 21, 2007 Report Share Posted April 21, 2007 You changed where the zener diode and ceramic capacitor are connected. They should be connected directly across the supply pins if the ICs. Now you have them connected across the high current 12V instead which is wrong. Quote Link to comment Share on other sites More sharing options...
kachew Posted April 21, 2007 Report Share Posted April 21, 2007 hi au,is this the configuration u meant? Quote Link to comment Share on other sites More sharing options...
audioguru Posted April 21, 2007 Report Share Posted April 21, 2007 The resistor and ceramic disc capacitor are correct but the zener diode is in series with the ICs instead of being in parallel with the ceramic disc capacitor. Quote Link to comment Share on other sites More sharing options...
MP Posted April 21, 2007 Report Share Posted April 21, 2007 I just noticed an Edit that occured a while ago.Another argument about an ordinary low power Cmos oscillator.Texas Instruments says that the minimum resistor value should be 10k, not 1k.The capacitor has AC across it so it should be non-polarized.1k with 4.7uF is wrong. 47k with 0.1uF is correct for 50Hz and 39k with 0.1uF for 60Hz.The posted values on the project have been bench tested. When someone posts something like the above quote, they should perform some bench testing instead of making such statements from theory and application notes. For the low frequency clock of 50 hz, these values work fine. As the TI application notes indicate, the R only has to be larger than a few hundred ohms. 1K is certainly high enough for this application. The TI suggestion for a higher R value helps with stability of high clock rates. 50 Hz is not a high clock rate. BTW- I did not choose these values. I only corrected the decimal problems in the original schematic. The project reads," Corrected by MP".....Not "re-designed by MP". There are certainly a lot of improvements that can be made of this or any other project posted on this site. However, note that the basic function of this project is correct. It is a simple battery powered square wave inverter. If you don't use the right transformer, you will not have good results. Other members of this community have built this project and are using it to power AC devices with success.As I have mentioned many times, theory is where one begins in the design process, but it is not worth much without the follow up of bench tests. One cannot make a final conclusion from theory and simulation without putting forth the work of actual application on the bench. You can never get enough information from a basic data sheet or application note to forego the bench work. Electronics is a "Hands-On"application. MP Quote Link to comment Share on other sites More sharing options...
audioguru Posted April 21, 2007 Report Share Posted April 21, 2007 As the TI application notes indicate, the R only has to be larger than a few hundred ohms. 1K is certainly high enough for this application.No.TI states in their datasheet for the CD4047, "R must be much larger than the Cmos on resistance in series with it". They recommend a minimum value of 10k so that the calculation doesn't need to include the on resistance which varies from device to device.This is ordinary low power Cmos, not high current and high speed Cmos. The on resistance is hundreds of ohms. My Cmos Cookbook by Don Lancaster also recommends a minimum value of 10k for the resistor.As I have mentioned many times, theory is where one begins in the design process, but it is not worth much without the follow up of bench tests. One cannot make a final conclusion from theory and simulation without putting forth the work of actual application on the bench. You can never get enough information from a basic data sheet or application note to forego the bench work. Electronics is a "Hands-On"application.Why do it wrong when it should be done correctly in the fisrt place. Quote Link to comment Share on other sites More sharing options...
MP Posted April 21, 2007 Report Share Posted April 21, 2007 Once again, the only value of your posts are personal conflict. It is posts like these that make me understand why no one would mentor you so that you could become an Electronic Engineer. As I stated earlier, the values posted in the project by the original author, Ashad Mustufa, have been bench tested and work well in application. The values you have posted will also work well. There is no value produced from this argument. MP Quote Link to comment Share on other sites More sharing options...
audioguru Posted April 21, 2007 Report Share Posted April 21, 2007 A 0.1uF 5% capacitor and a 47k 5% resistor produce a frequency of 45Hz to 55Hz. The Cmos has an on resistance of about 500 ohms in series with the resistor which reduces the calculated frequency by only 1%.Your 4.7uF 20% capacitor and a 1k 5% resistor produce 29Hz to 42Hz because the on resistance is a big part of the total resistance and the tolerance of the huge capacitor is so high.The on resistance could be as high as 1k ohms. Then your frequency is way off.Why not do it correctly?? Quote Link to comment Share on other sites More sharing options...
kachew Posted April 22, 2007 Report Share Posted April 22, 2007 hi au,how do u choose, how u know what value of the resistors and capcitors for voltage spikes? the value of the capacitors and zener diode across the power supply of the ic? any formulas?thanks Quote Link to comment Share on other sites More sharing options...
audioguru Posted April 22, 2007 Report Share Posted April 22, 2007 I peeked at the hard drive of my old 486 computer. It has the original Low Cost 500W Inverter's schematic.I thought so! The original designer used 100 ohms in the oscillator to make the frequency of his inverter 50Hz because the on resistance of the Cmos was too high for his too big capacitor. Two wrongs. Quote Link to comment Share on other sites More sharing options...
audioguru Posted April 22, 2007 Report Share Posted April 22, 2007 how do u choose, how u know what value of the resistors and capcitors for voltage spikes? the value of the capacitors and zener diode across the power supply of the ic? any formulas?The max supply voltage for most CD4047 and CD4001 ICs is 18V but Texas Instruments has 22V. A 16V zener diode has a 5% max voltage of 16.8V so it will keep the supply to the ICs from going higher. The capacitor smooths fast supply pulses and the 47 ohm resistor reduces the max current through the zener diode. Quote Link to comment Share on other sites More sharing options...
MP Posted April 22, 2007 Report Share Posted April 22, 2007 hi au,how do u choose, how u know what value of the resistors and capcitors for voltage spikes? the value of the capacitors and zener diode across the power supply of the ic? any formulas?thanksKachew,Are you looking for the formula for the CD4047? It is 1/(4.4 * R * C). R is in ohms and C is in Farads. There is a lot of tolerance latitude since this is only a square wave inverter. Square wave inverters are not used in circuits that require frequency and voltage precision since the items that require it would be damaged or would not work well with square waves. MP Quote Link to comment Share on other sites More sharing options...
audioguru Posted April 22, 2007 Report Share Posted April 22, 2007 Kachew is not making a square-wave inverter. He is making a modified sine-wave inverter with the correct values of R and C for a CD4047 IC and is using a CD4001 to gate the outputs into a modified sine-wave. Quote Link to comment Share on other sites More sharing options...
kachew Posted April 23, 2007 Report Share Posted April 23, 2007 hi,the capacitors for the voltage spikes is AC caps or Dc caps?thanks Quote Link to comment Share on other sites More sharing options...
audioguru Posted April 23, 2007 Report Share Posted April 23, 2007 the capacitors for the voltage spikes is AC caps or Dc caps?I talked about adding one 0.1uF ceramic disc capacitor across the supply for the ICs. It is ceramic which does not have polarity so it is for AC or for DC. Quote Link to comment Share on other sites More sharing options...
kachew Posted April 23, 2007 Report Share Posted April 23, 2007 hi au,how about the voltage spikes caps near the transformer? izzit the ceramic type too? Quote Link to comment Share on other sites More sharing options...
audioguru Posted April 23, 2007 Report Share Posted April 23, 2007 I forgot about the 0.22uF cap in series with the 100 ohms. A ceramic disc capacitor would work well. Quote Link to comment Share on other sites More sharing options...
ante Posted April 25, 2007 Report Share Posted April 25, 2007 Perhaps TI should employ some 30+years engineer to amend their datasheets since their present brains doesn Quote Link to comment Share on other sites More sharing options...
audioguru Posted April 25, 2007 Report Share Posted April 25, 2007 Hi Ante,Engineers at TI actually engineer many circuits and projects, correctly.I have a new topic in the Theory forum about the problems with the CD4047 oscillator's errors in our 500W inverter project. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.