Post by thread:4 digit auto-ranging frequency-meter.

part 0 988 bytes content-type:application/octet-stream; name="meter.gif"I ran into a very interesting PIC16C84 project: a 4 digit auto-ranging frequency-meter (made by Simone Benvenuti & Andrea Geniola) which uses virtually no parts. Unfortunately the schematics provided lack some information. The type of transistors is not mentioned, neither the majority of resistor values, nor the type of display. I'm enclosing the schematic in case anybody can add some light to it. Thank you very much. -------------------------------------------- "I must not fear. Fear is the mind-killer. Fear is the little death that brings total obliteration. I will face my fear... I will permit it to pass over me and through me." Kwisatz Haderach - Dune netQ <spam_OUTnetquakeTakeThisOuTinnocent.com> Content-Type: application/octet-stream; name="meter.gif" Content-Description: meter.gif (Imagen de Corel PHOTO-PAINT 7.0) Content-Disposition: attachment; filename="meter.gif" Attachment converted: wonderland:meter.gif (GIFf/JVWR) (0001967C)

Hi, The transistors should be any NPN general purpose transistor (such as PN2222). The resistors on the LEDs should be in the range of 100 to 300 ohms, depending on the needed brightness and the supply voltage. The resistors on the bases of the transistors should be around 10k. I know these values because I have made similar display circuits before. Good luck! Looks like a nice project! Sean At 09:27 PM 10/6/98 -0300, you wrote: {Quote hidden}>Hi! > >I ran into a very interesting PIC16C84 project: a 4 digit auto-ranging >frequency-meter >(made by Simone Benvenuti & Andrea Geniola) which uses virtually no parts. >Unfortunately the schematics provided lack some information. The type of >transistors >is not mentioned, neither the majority of resistor values, nor the type of >display. >I'm enclosing the schematic in case anybody can add some light to it. > >Thank you very much. > >-------------------------------------------- >"I must not fear. Fear is the mind-killer. >Fear is the little death that brings total >obliteration. I will face my fear... I will >permit it to pass over me and through me." > >Kwisatz Haderach - Dune > > netQ ><.....netquakeKILLspam@spam@innocent.com> >Attachment Converted: "c:\bawin\winba\eudora\attach\meter.gif" > +--------------------------------+ | Sean Breheny | | Amateur Radio Callsign: KA3YXM | | Electrical Engineering Student | +--------------------------------+ Save lives, please look at http://www.all.org Personal page: http://www.people.cornell.edu/pages/shb7 shb7KILLspamcornell.edu Phone(USA): (607) 253-0315 ICQ #: 3329174

Oops! I forgot the one resistor, the one attached to MCLR. Its value is not critical at all, in fact, you could just hook MCLR to V+. But, if you prefer, you could add any resistor around 10k or so. Sean At 09:27 PM 10/6/98 -0300, you wrote: {Quote hidden}>Hi! > >I ran into a very interesting PIC16C84 project: a 4 digit auto-ranging >frequency-meter >(made by Simone Benvenuti & Andrea Geniola) which uses virtually no parts. >Unfortunately the schematics provided lack some information. The type of >transistors >is not mentioned, neither the majority of resistor values, nor the type of >display. >I'm enclosing the schematic in case anybody can add some light to it. > >Thank you very much. > >-------------------------------------------- >"I must not fear. Fear is the mind-killer. >Fear is the little death that brings total >obliteration. I will face my fear... I will >permit it to pass over me and through me." > >Kwisatz Haderach - Dune > > netQ ><.....netquakeKILLspam.....innocent.com> >Attachment Converted: "c:\bawin\winba\eudora\attach\meter.gif" > +--------------------------------+ | Sean Breheny | | Amateur Radio Callsign: KA3YXM | | Electrical Engineering Student | +--------------------------------+ Save lives, please look at http://www.all.org Personal page: http://www.people.cornell.edu/pages/shb7 EraseMEshb7spam_OUTTakeThisOuTcornell.edu Phone(USA): (607) 253-0315 ICQ #: 3329174

>I ran into a very interesting PIC16C84 project: a 4 digit auto-ranging >frequency-meter >(made by Simone Benvenuti & Andrea Geniola) which uses virtually no parts. >Unfortunately the schematics provided lack some information. The type of >transistors >is not mentioned, neither the majority of resistor values, nor the type of >display. The displays are seven segment common cathode, it looks like a 4 digit multiplexed array, but four singles in paralell would work as well. The base resistors arent' critical, 470 ohm or 1k should work. The segment resistors are also uncritical, try 220-470. Depends much on the display efficiency. It's a cute project all right, hard to get less parts.

do you have any code for this? I would be interested in building it.... netQ wrote: {Quote hidden}> > Hi! > > I ran into a very interesting PIC16C84 project: a 4 digit auto-ranging > frequency-meter > (made by Simone Benvenuti & Andrea Geniola) which uses virtually no parts. > Unfortunately the schematics provided lack some information. The type of > transistors > is not mentioned, neither the majority of resistor values, nor the type of > display. > I'm enclosing the schematic in case anybody can add some light to it. > > Thank you very much. > > -------------------------------------------- > "I must not fear. Fear is the mind-killer. > Fear is the little death that brings total > obliteration. I will face my fear... I will > permit it to pass over me and through me." > > Kwisatz Haderach - Dune > > netQ > <netquakespam_OUTinnocent.com> > > ------------------------------------------------------------------------ > > Name: meter.gif > meter.gif Type: GIF Image (image/gif) > Encoding: base64 > Description: meter.gif (Imagen de Corel PHOTO-PAINT 7.0)

hi ! >From the schematic I can tell you this. For transistors you can use any NPN as B C547, BC237 etc., resistors value between portA and BASEs of transistors are from 1K to 10K ohm, resistors value for segments are around 390 ohm(even less to 270 ohm 5V-1,8V(V -LED)=3,2V so 3,2V/270ohm=11,8mA for each segment) and finaly display can be any with common c atode as HD1133X the resistor from MCLR to +V is arround 1Kohm. All values are not critical. Valter netQ wrote: {Quote hidden}> Hi! > > I ran into a very interesting PIC16C84 project: a 4 digit auto-ranging > frequency-meter > (made by Simone Benvenuti & Andrea Geniola) which uses virtually no parts. > Unfortunately the schematics provided lack some information. The type of > transistors > is not mentioned, neither the majority of resistor values, nor the type of > display. > I'm enclosing the schematic in case anybody can add some light to it. > > Thank you very much. > > -------------------------------------------- > "I must not fear. Fear is the mind-killer. > Fear is the little death that brings total > obliteration. I will face my fear... I will > permit it to pass over me and through me." > > Kwisatz Haderach - Dune > > netQ > <@spam@netquakeKILLspaminnocent.com> > > ------------------------------------------------------------------------ > > Name: meter.gif > meter.gif Type: Corel PHOTO-PAINT 8.0 Image (application/x-unknown -content-type-CorelPhotoPaint.Image.8) > Encoding: base64 > Description: meter.gif (Imagen de Corel PHOTO-PAINT 7.0)

At 09:29 PM 10/6/98 -0500, you wrote: > >The displays are seven segment common cathode, it looks like a 4 digit >multiplexed array, but four singles in paralell would work as well. > >The base resistors arent' critical, 470 ohm or 1k should work. Do you really think THAT low? Even if the transistors were completely saturated, I would expect the base resistors to be about 10 times the segment resistors. If we used 470 ohm resistors on the bases, with a 5V supply, we would have a base current of about 8mA. It would work fine, but it would seem to me that for efficiency's sake, a few K would be better. It would still throw the transistors into saturation and save some power. >The segment resistors are also uncritical, try 220-470. Depends much >on the display efficiency. > >It's a cute project all right, hard to get less parts. > Sean +--------------------------------+ | Sean Breheny | | Amateur Radio Callsign: KA3YXM | | Electrical Engineering Student | +--------------------------------+ Save lives, please look at http://www.all.org Personal page: http://www.people.cornell.edu/pages/shb7 KILLspamshb7KILLspamcornell.edu Phone(USA): (607) 253-0315 ICQ #: 3329174

Dave VanHorn wrote: > The displays are seven segment common cathode, it looks like a 4 digit > multiplexed array, but four singles in paralell would work as well. > > The base resistors arent' critical, 470 ohm or 1k should work. > The segment resistors are also uncritical, try 220-470. Depends much > on the display efficiency. > > It's a cute project all right, hard to get less parts. You can still minimize the parts by having the 7segment common anode, make all the colllectors of the NPN transistors tie to the +5V supply and reverse/invert drive to the LED (to the negative side). This way the base resistors is not needed anymore. Reggie

>>The base resistors arent' critical, 470 ohm or 1k should work. > >Do you really think THAT low? Even if the transistors were completely >saturated, I would expect the base resistors to be about 10 times the >segment resistors. If we used 470 ohm resistors on the bases, with a 5V >supply, we would have a base current of about 8mA. It would work fine, but >it would seem to me that for efficiency's sake, a few K would be better. It >would still throw the transistors into saturation and save some power. You have to allow for all seven segments to be on, plus worstcase of the pic output drive.. A higher R will probably work. I'd use high eff red displays then, and do the math. This is an off-the cuff shot.

>You can still minimize the parts by having the 7segment common anode, >make all the colllectors of the NPN transistors tie to the +5V supply >and reverse/invert drive to the LED (to the negative side). This way >the base resistors is not needed anymore. > >Reggie Are you SURE they will turn off? Look at the pin specs for minimum high output voltage, and leakage current. I've got my doubts about the current design, but IIRC, they can drive closer to ground than to VCC.

Looks familiar..... When I was at Technicon, I did a course on the PIC. One of the problems in the handbook was a Frequency Meter just like the one you send. We didn't came around doing the problem, but I've got some info on how it works. If any one is interested let me know and I'll see if I can get the info onto the pc. Cheers Herman *********************************************** Herman Theron Hermanus Magnetic Observatory (CSIR) PO Box 32 Hermanus 7200 South Africa Email: RemoveMEhetheronTakeThisOuTcsir.co.za Tel: +27 283 21196 Fax: +27 283 22039 Cel: 0832973358 ***********************************************

What would happen in the schematic shown if the input voltage when negative with respect to the gnd of the PIC ? my guess is that it wouldnt be good for the PIC. in a similar project ( from memory), they use a coil, capacitor, signal diodes, and a transistor to filter and switch the incoming frequency to make it nicer for the PIC to count. i built the project, and it seemed to give accurate results in comparison to a bench frequency counter at a frequency of 40Mhz using a 10Mhz 16cf84. (i used an fm RC transmitter to calibrate it) -- i don't have a frequency generator higher than that so i can't test its high end response. regards Pete

Peter Gee wrote: > > What would happen in the schematic shown if the input voltage when negative > with respect to the gnd of the PIC ? The protections diodes may protect it from damage but excessive current may destroy it ultimately. > my guess is that it wouldnt be good for the PIC. > in a similar project ( from memory), they use a coil, capacitor, signal > diodes, and a transistor to filter and switch the incoming frequency to make > it nicer for the PIC to count. The posted circuit is incomplete, for a more sensitive frequency counter, it needs a pre-amplifier and convert its output to TTL level. > i built the project, and it seemed to give accurate results in comparison to > a bench frequency counter at a frequency of 40Mhz using a 10Mhz 16cf84. (i > used an fm RC transmitter to calibrate it) -- i don't have a frequency > generator higher than that so i can't test its high end response. The PIC doesn't need to be 10MHz, the input is captured in the RTCC using it's prescaler which is rated at 50MHz. The pre-scaler cannot be read directly, that is why RB0 is connected at RA4 to self clock the pre-scaler to get it's value. The 470 ohm resistor is to prevent contention from the input and RB0. Reggie

Dave VanHorn wrote: > > >You can still minimize the parts by having the 7segment common anode, > >make all the colllectors of the NPN transistors tie to the +5V supply > >and reverse/invert drive to the LED (to the negative side). This way > >the base resistors is not needed anymore. > > > >Reggie > > Are you SURE they will turn off? Look at the pin specs for minimum > high output voltage, and leakage current. I've got my doubts about the > current design, but IIRC, they can drive closer to ground than to VCC. Yes I'm sure, 25 ma for sinking, V(OL) = 0.6V . I had used this configuration on some of my designs. In case you misunderstood my point, here is an ascii schematic of what I meant. O Vdd | O---------------O---------------O---------------O | | | | | | | | |/ |/ |/ |/ RA0 O---B| RA1 O---B| RA2 O---B| RA2 O---B| |\ |\ |\ |\ V V V V | | | | O O O O digit select 1 digit select 2 digit select 3 digit select 4 RB1 O---/\/\/\-----O Segment A RB2 O---/\/\/\-----O Segment B RB3 O---/\/\/\-----O Segment C RB4 O---/\/\/\-----O Segment D RB5 O---/\/\/\-----O Segment E RB6 O---/\/\/\-----O Segment F RB7 O---/\/\/\-----O Segment G Reggie

Hi All, At 11:38 PM 10/6/98 -0500, you wrote: >>>The base resistors arent' critical, 470 ohm or 1k should work. >> >>Do you really think THAT low? Even if the transistors were completely >>saturated, I would expect the base resistors to be about 10 times the >>segment resistors. If we used 470 ohm resistors on the bases, with a >5V >>supply, we would have a base current of about 8mA. It would work >fine, but >>it would seem to me that for efficiency's sake, a few K would be >better. It >>would still throw the transistors into saturation and save some >power. > > >You have to allow for all seven segments to be on, plus worstcase of >the pic output drive.. >A higher R will probably work. I'd use high eff red displays then, and >do the math. This is an off-the cuff shot. I must stand corrected. I was tired and not thinking of multiple segments on at once. Dave, you are right, an R in the 1K range seems correct for best/most consistent light output. Thanks, Sean +--------------------------------+ | Sean Breheny | | Amateur Radio Callsign: KA3YXM | | Electrical Engineering Student | +--------------------------------+ Save lives, please look at http://www.all.org Personal page: http://www.people.cornell.edu/pages/shb7 spamBeGoneshb7spamBeGonecornell.edu Phone(USA): (607) 253-0315 ICQ #: 3329174

>The posted circuit is incomplete, for a more sensitive frequency >counter, it needs a pre-amplifier and convert its output to TTL level. Does anyone have a nice and simple pre-amplifier schematic for this kind of application? Regards, Octavio ======================================================= Octavio Nogueira - TakeThisOuTnogueiraEraseMEspam_OUTmandic.com.br ICQ# 19841898 >From the creator of ProPic, ProPic 2 now much better * http://members.tripod.com/ProPic * * PIC Programmer for Windows with down to earth price * =======================================================

Octavio Nogueira wrote: > > >The posted circuit is incomplete, for a more sensitive frequency > >counter, it needs a pre-amplifier and convert its output to TTL level. > > Does anyone have a nice and simple pre-amplifier schematic for this kind > of application? I'll try on ascii: +---------+---O +5V | | 3K3 1K ohms ohms | | +---+ +-----+ +| | | +| 2.2uF = 10K 820 = 2.2uF | ohms ohms | GND | | GND | | OUT | +-----||----+----O TTL | | 10nF | IN 10nF 10nF | |/ | O----||----+--+---||---+-------| MPSH10 10K adjust for DC | | | |\ Ohms level be in the 1N4148 V - 3K3 v trimmer middle of TTL X2 - ^ ohms | | thresholds | | | +----+ | GND GND GND | | GND 120 |+ ohms = 10nF + 2.2 uF in parallel | | GND GND This circuit could work from to 0.02-50 MHz. Reggie

how exactly does the pic capture frequencies of up to 50 MHz with only a 4 MHz ocillator? what is the mechanism? jon Regulus Berdin wrote: {Quote hidden}> > Peter Gee wrote: > > > > What would happen in the schematic shown if the input voltage when negative > > with respect to the gnd of the PIC ? > The protections diodes may protect it from damage but excessive current > may destroy it ultimately. > > > my guess is that it wouldnt be good for the PIC. > > in a similar project ( from memory), they use a coil, capacitor, signal > > diodes, and a transistor to filter and switch the incoming frequency to make > > it nicer for the PIC to count. > The posted circuit is incomplete, for a more sensitive frequency > counter, it needs a pre-amplifier and convert its output to TTL level. > > > i built the project, and it seemed to give accurate results in comparison to > > a bench frequency counter at a frequency of 40Mhz using a 10Mhz 16cf84. (i > > used an fm RC transmitter to calibrate it) -- i don't have a frequency > > generator higher than that so i can't test its high end response. > The PIC doesn't need to be 10MHz, the input is captured in the RTCC > using it's prescaler which is rated at 50MHz. The pre-scaler cannot be > read directly, that is why RB0 is connected at RA4 to self clock the > pre-scaler to get it's value. The 470 ohm resistor is to prevent > contention from the input and RB0. > > Reggie

PLUS . . . if you tie the collectors of the transistors to the UNREGULATED supply, the 5V regulator is nice and small and cheap. >Dave VanHorn wrote: >> The displays are seven segment common cathode, it looks like a 4 digit >> multiplexed array, but four singles in paralell would work as well. > >You can still minimize the parts by having the 7segment common anode, >make all the colllectors of the NPN transistors tie to the +5V supply >and reverse/invert drive to the LED (to the negative side). This way >the base resistors is not needed anymore. > >Reggie dwayne Dwayne Reid <RemoveMEdwaynerTakeThisOuTplanet.eon.net> Trinity Electronics Systems Ltd Edmonton, AB, CANADA (403) 489-3199 voice (403) 487-6397 fax

Regulus Berdin wrote: > Try http://www.picpoint.com . I had seen it once posted there. You're right! Apparently so had I, as I already had it on file! Boy I must start looking through my 500MB of files! This one I must build for interest, but I'm fancy doing those little optimisations on the drivers! -- Cheers, Paul B.

I don't know if this was discussed before, my question to you is resolution at 4.0 Mhz. If I were to display decimal frequency in the range of 0 to 120 Hz, the max resolution I can achieve is per bit 0.48 Hz. It means if I were to display a decimal point, each display advancement is 0.48Hz. Is there a way to improve this. Norayr

"Norayr S. Elmayan" wrote: > > I don't know if this was discussed before, my question to you is resolution > at 4.0 Mhz. If I were to display decimal frequency in the range of 0 to > 120 Hz, the max resolution I can achieve is per bit 0.48 Hz. It means if I > were to display a decimal point, each display advancement is 0.48Hz. Is > there a way to improve this. You can improve the resolution by increasing the capture time. A 1 second capture time can give you 1 Hz resolution and 10 seconds will give 0.1 Hz. But I think you misunderstood how the system works, the count range is not only 8 bit. The minimal range is 16 bits and can be expanded by polling the TMR0 overflows which is easy because it will only occur at minimum every 1/760 second. The frequency is read via RA4/TMR0 which a pre-scaler/divider is set in between the TMR0 register. Reading the TMR0 for each second or a fixed period, will give you the frequency per time period divided by the prescaler value. To make it simple, the prescaler is set to 256 divisions, so the TMR0 can be the high byte of the 16 bit number. Reading the pre-scaler is tricky because it is not mapped as a register. The trick is self clocking, RB0 is connected to RA4 and is clocked until TMR0 changes. The value of the pre-scaler will be 256 - no. of clock pulses. I hope this answers your question. Reggie

Put me down for a copy. Regards, John. ---------- > From: Bob Blick <EraseMEbobTED.NET> > To: RemoveMEPICLISTEraseMEEraseMEMITVMA.MIT.EDU > Subject: Re: 4 digit auto-ranging frequency-meter. > Date: Friday, 9 October 1998 3:44 > > If anyone wants the English version of the text file for this project let > me know, I can email it out tonight. Please respond to me, rather than the > list, so your request will stick out from the rest of the piclist > messages. > > It's rather big, so I won't send it to the list in general. > > -Bob