Post by thread:making sounds (with a pic and a speaker)

- what is an appropriate (simple) amplifier for a small speaker (or is the 5 volt the limit, and should I use either a higher voltage or a 4 transistor bridge)? Why do you think you need anything more than a single transistor to Ground? - would a piezo buzzer make more noise (when driven with the same voltage)? Yes, put you'd have much less flexibility in the type of sound. - does anyone have experience in generating more interesting sounds via 1 output bit (for instance using PWM)? There exists IBM PC (original) software that can use the internal speaker (which has a single bit, single transistor to ground) for multi-tone music, synthesized voice, and all sorts of stuff. My understanding is that it works by combining a PWM signal with a high frequency signal designed to "bias" the speaker cone near the middle of its travel points (in combination with physical inertia.) BillW

At 09:45 PM 10/8/97 +0100, you wrote: >Im working on a 16C84 based 'laser' (actually IR) gun/target. >When it either shoots or it is hit it should make some interesting noise. >I've connected a PIC pin to a 100uF elco and a small 8 ohm LSP, >but that doesn't make much noise. Adding a BD136/BD137 makes more noise, >but not enough for my taste. Also a simple square wave makes a rather dull >sound. > >Questions: >- what is an appropriate (simple) amplifier for a small speaker (or is > the 5 volt the limit, and should I use either a higher voltage or a > 4 transistor bridge)? >- would a piezo buzzer make more noise (when driven with the same voltage)? >- does anyone have experience in generating more interesting > sounds via 1 output bit (for instance using PWM)? > A simple single NPN transistor with the speaker attached between the collector and positive, the emitter to ground, and the base to a PIC output via a resistor (around 10K, I would estimate) should drive the speaker with around 200 to 400mW of power when you pulse the PIC pin. This may be suitable (it is equivalent to a small radio at full volume). You certainly could get a piezo buzzer that would be louder, but it would probably need a transistor to drive it, the 20mA or so of PIC pin drive probably would not be enough. The voltage is not the primary concern, it is the current. I have done 1 bit complex sound on a PC speaker. It seems to me that you could do the same on a PIC. All you do is to come up with a waveform (you could record some type of sound that you like through a computer sound card and look at the waveform on the computer. See how often it repeats. Lets say it repeats around every 20ms and can be adequately sampled at 8KHz. What you can do is take a 20ms chunk of 8KHz samples and translate each sample (0-255 8 bit value) into lets say 4 levels (0 for 0-63, 1 for 64-127, etc.). Now, write a routene for your PIC which every 1/8000 of a second, looks at the next sample in a row and turns the output pin on for an amount of time which corresponds to the appropriate one of the 4 levels. This would require a timing resolution of 1/24000 of a second. This is quite possible on a 16F84 if you do your timing via instruction timing (1 instruction = 1 microsecond at 4MHz) Here is an example: Original 4 samples: 0 100 200 50 Translated levels: 0 1 3 0 Timing of pin(tickmarks are 1/24000 sec, * is on, _ is off) ___*__***___ |||||||||||| #1 #2 #3 #4 sample #s^ I realize that I didn't do a great job of explaining it, but this is a basic PWM method of doing pretty good 1 bit audio. I am able to play back .WAV files on the PC speaker with good sound. If you want, I can provide you with C source code which does this on a PC. I would reccommend using a look up table to get the timing values for each sample, unless you happen to be able to have 64 or less samples, in which case you can place them in the EEPROM data area of the 16C/F84. Good luck, Sean Sean Breheny,KA3YXM Electrical Engineering Student

At 09:45 PM 10/8/97 +0100, Floortje Hanneman & Wouter van Ooijen <spam_OUTwfTakeThisOuTXS4ALL.NL> wrote: >Im working on a 16C84 based 'laser' (actually IR) gun/target. >When it either shoots or it is hit it should make some interesting noise. >I've connected a PIC pin to a 100uF elco and a small 8 ohm LSP, >but that doesn't make much noise. Adding a BD136/BD137 makes more noise, >but not enough for my taste. Also a simple square wave makes a rather dull >sound. > >Questions: >- what is an appropriate (simple) amplifier for a small speaker (or is > the 5 volt the limit, and should I use either a higher voltage or a > 4 transistor bridge)? >- would a piezo buzzer make more noise (when driven with the same voltage)? >- does anyone have experience in generating more interesting > sounds via 1 output bit (for instance using PWM)? If you can afford an additional pin, you can get twice the voltage amplitude by differential-driving your speaker. Simply output 01 or 10 to the two pins to either apply +V or -V across your speaker, giving a 2*V amplitude swing. Using one pin, you get only 0 to +V across the speaker. PICs have relatively high current sink & source capabilities (I think 25mA & 20mA, respectively, for the 16C84), but 5V into an 8 ohm load can draw up to 625mA, so a simple transistor amplifier could help if you need higher volume than is possible with differential drive. Small speakers have a rather limited frequency response. Most that I have seen are best used between about 400 - 4000 Hz. Piezo elements have higher impedance than speakers, and you should be able to get fairly high volume out of one with even single-ended drive off of one pin, but they have even worse frequency response than miniature speakers. Most have a resonant peak around 3-4 KHz. If you have access to a signal generator, try driving a piezo element like the one you plan to use with a 5Vp-p square wave, and tweak the frequency for maximum volume. Design your PIC circuit to use the same frequency. If you don't have a signal generator, many sources (such as DigiKey) list the resonant frequencies of the piezo elements they sell. If you want some more control over waveform, try using two pins, two resistors, and an op-amp voltage follower circuit to build a simple buffered D/A converter to drive your output device. +------------+ | | | |\ | | | \ | +--|- \ | | >-----+------O Vout +-----------+-----------|+ / | | | / > > |/ D1 | D0 | Vout < 10Kohms < 20Kohms -----+----+------- > > 0 | 0 | 0 | | 0 | 1 | V/3 | | 1 | 0 | 2V/3 --------------------------------- 1 | 1 | V D1 D0 PIC Output Port (where V is your supply voltage) This will allow you to have 4 distinct voltage levels on your output, and the op-amp will give you higher output drive capability, as well. By careful setting of the output levels, you can simulate triangle, sine, sawtooth, and other waveforms, albeit with only two bits of accuracy. However, depending on how many output pins you have free, you could go to 3 or 4 bits by extending this concept. Just make sure that each resistor is twice the value of the previous one. For example, use a 10K on D3, a 20K on D2, 40K on D1, and 80K on D0. Beyond 4 bits, go to a R/2R network or a dedicated D/A chip. Hopefully, my ASCII art circuit is sufficiently legible.... use a fixed-pitch font such as Courier to view it. - Rick "ASCII silly question, get a silly ANSI" Dickinson +---------------------------------+---------------------------+ | Enterprise ArchiTechs Company | | | Lotus Certified Notes | Never underestimate the | | Appl. Design & Administration | bandwidth of a station- | |(818)563-1061 .....rtdKILLspam@spam@notesguy.com | wagon full of tape reels. | | http://www.eArchiTechs.com | | +---------------------------------+---------------------------+

In message <1.5.4.32.19971008204500.00366a04KILLspamxs4all.nl>, Floortje Hanneman & Wouter van Ooijen <.....wfKILLspam.....XS4ALL.NL> writes >Im working on a 16C84 based 'laser' (actually IR) gun/target. >When it either shoots or it is hit it should make some interesting noise. >I've connected a PIC pin to a 100uF elco and a small 8 ohm LSP, >but that doesn't make much noise. Adding a BD136/BD137 makes more noise, >but not enough for my taste. Also a simple square wave makes a rather dull >sound. > >Questions: >- what is an appropriate (simple) amplifier for a small speaker (or is > the 5 volt the limit, and should I use either a higher voltage or a > 4 transistor bridge)? >- would a piezo buzzer make more noise (when driven with the same voltage)? >- does anyone have experience in generating more interesting > sounds via 1 output bit (for instance using PWM)? With a square wave, simply sweeping the frequency (incrementing it on every cycle) makes it much more "interesting"; some of the pedestrian crossings near where I live use this technique, presumably to make people walk faster. Here's a little program I wrote some time ago to try this out. ;WRITTEN BY LEON HELLER ;DATE 20/7/96 ;FILE NAME WARNING.TXT ;PROCESSOR PIC 16C84 ;RESONATOR 4 MHZ XTAL ;CYCLE TIME 1 uS ;WATCHDOG DISABLED ;CODE PROTECTION OFF ;EQUATES PORTB EQU 6 PB0 EQU 0 STATUS EQU 3 COUNT EQU 0x0C TEMP EQU 0x0D ;SIMPLE PROGRAM TO GENERATE A SWEPT FREQUENCY TONE ON PB0 ORG 00 INIT MOVLW 00 ;MAKE PORT B OUTPUTS TRIS PORTB CLRF PORTB CLRW ;SETS START FREQUENCY LOOP CALL CYCLE ;GENERATE ONE CYCLE ADDLW 1 ;INCREMENT FREQUENCY GOTO LOOP ;KEEP GOING CYCLE BSF PORTB,PB0 CALL DELAY NOP NOP BCF PORTB,PB0 CALL DELAY RETURN ;DELAY ROUTINE, CALLED WITH DELAY COUNT IN W DELAY NOP NOP NOP MOVWF COUNT DELAY1 NOP DECFSZ COUNT,1 GOTO DELAY1 GOTO DELAY2 DELAY2 NOP RETURN END Leon -- Leon Heller: EraseMEleonspam_OUTTakeThisOuTlfheller.demon.co.uk http://www.lfheller.demon.co.uk Amateur Radio Callsign G1HSM Tel: +44 (0) 118 947 1424 See http://www.lfheller.demon.co.uk/rcm.htm for details of a low-cost reconfigurable computing module using the XC6216 FPGA

In message <KILLspam199710091925.OAA22250KILLspamdfw-ix5.ix.netcom.com>, Andrew Warren <RemoveMEfastfwdTakeThisOuTIX.NETCOM.COM> writes >Leon Heller <spamBeGonePICLISTspamBeGoneMITVMA.MIT.EDU> wrote: > >> With a square wave, simply sweeping the frequency (incrementing it >> on every cycle) makes it much more "interesting"; some of the >> pedestrian crossings near where I live use this technique, >> presumably to make people walk faster. > >Leon: > >It's more likely that the sounders are there to inform blind >pedestrians that the traffic light is in their favor. > >-Andy Only a few of these crossings have sounders, but this is the only one I've come across with a swept frequency. Leon -- Leon Heller: TakeThisOuTleonEraseMEspam_OUTlfheller.demon.co.uk http://www.lfheller.demon.co.uk Amateur Radio Callsign G1HSM Tel: +44 (0) 118 947 1424 See http://www.lfheller.demon.co.uk/rcm.htm for details of a low-cost reconfigurable computing module using the XC6216 FPGA

Hiya, I just sent this reply back to Paul directly when I meant to send it to the entire list: Paul Webster Wrote: > Someone posted a two-resistor "ladder" array and stated it permitted >four levels or states. Wrong! Three output states for each resistor >makes nine, enough for quite detailed analogue sound output. Paul, could you explain how to do this with a Resistor ladder? The Best that I can do is n + 1 (where "n" is the Number of Resistors): | o0 |--- R0 ---+--- Output | | | R1 | | o1 |----------+ | | | R2 | | o2 |----------+ | With this circuit, I can get: Vcc -> o0 at Vcc, o1 = o2 = High Impedance Gnd -> o0 at Gnd, o1 = o2 = High Impedance V1 -> o0 at Vcc, o1 at Gnd, o2 at High Impedance, V1 = R1 / ( R0 + R1 ) V2 -> o0 at Vcc, o2 at Gnd, o1 at High Impedance, V2 = ( R1 + R2 ) / ( R0 + R1 + R2 ) Now, I suppose that multiple Outputs could be turned on to get parallel resistances, but I would be concerned about this with uneven current drive capabilities with different pins. This method would make it very hard to calculate the actual Voltage Outputs. There must be something I'm missing? Or is a circuit other than a ladder used? myke Check out "Programming and Customizing the PIC Microcontroller" at: http://www.myke.com

Thanks for the tips. I've done some more experiments, mayby someone can (re-) use the results. My aim is to produce a few different sounds (at least one for shooting, one for being hit and one for being killed) which are both loud enough to be heard from some distance and appropriate for their function. I've finished the electronics, now I'll start exprimenting with waveforms (using PWM). ELECTRONICS The most simple solution is to couple a PIC output directly to a speaker, with or without a C. This gives a verly low volume due to the limited drive capability of the PIC. ___ \ / PIC --------------| |----- Gnd Lsp - ___ || \ / PIC ------||------| |----- Gnd C || Lsp - Direct PIC output with and without capacitive coupling: simple, but not much volume due to PIC impedance. The one-transistor solution has the drawback that current always flows through the speaker in the same direction (switched DC), which in the long run could be bad for the speaker, and limits the movement of the conus: it can only move from the middle position to one of the extreme positions. And when you leave the pin in the wrong (is this case: high) position for too long quite a current will flow through the speaker, not a good idea for a battery-powered application. ___ \ / /----| |---- Vcc R ____ |/ - PIC ------|____|----| Lsp |> npn \ Gnd or +------- Vcc |/ PIC ----------------| |> ___ Lsp npn \ \ / +---| |--- Gnd - single npn buffer: more volume, but not optimal (DC problem) That's why I initially choose a npn/pnp pair, which gives a real AC of Vcc Top-Top (effective voltage is lower: 0.5 * Vcc for a square wave, 1/2*sqrt(2) * Vcc for a sine). npn / Vcc |/ +---| ___ | |>\ || \ / PIC ---------+ +---||----| |--- Vcc or Gnd | |</ || - +---| C Lsp |\ pnp \ Gnd complementary buffer: good volume, real AC, low impedance. Can not make use of a Vcc > PIC-Vcc In my application I'll have a 7-9V NiCad which is regulated down to 5V for the PIC and the SFH506 IR receiver, so I can gain some volume by using the NiCad directly. This requires a level shifter (3th transistor). ____ +---|____|------+---- Vcc > Vcc-PIC | |/ +----------+--| ___ ____ |/ | |>\ || \ / PIC --|____|---| | +---||---| |-+ |>\ | |</ || - | npn \ +--| | Gnd |\ Vcc or Gnd \ Gnd complementary buffer with inverting level shifter to make use of a higher Vcc than the PIC's Vcc. Once the impedance of the driver is no longer a problem the volume is determined by the impendance of the speaker and the voltage applied to it ( P = V*V/R ). So to get a high volume it is important to take the lowest impedance speaker which you can get (4 ohms if possible). It is even more important to apply the highest possible voltage. If you only have 5V it will help to use two PIC output pins which are always each others inverse to drive two complementary buffers. If you do not want to use two PIC pins an extra transistor can provide the inversion. And if you want to use a Vcc > PIC-Vcc you'll have to add two level-shifters (no need for two PIC outputs then, the level shifters are also inverters). npn / Vcc |/ +---| ___ | |>\ \ / PIC1 --------+ +---| |-+ | |</ - | +---| | |\ | pnp \ Gnd | | npn / Vcc | |/ | +---| | | |>\ || | PIC2 --------+ +---||--+ (inverse | |</ || of PIC1) +---| |\ pnp \ Gnd dual complementary buffer (bridge) doubles the voltage compared to a single bridge. MECHANICS One non-electric measure which is quite important: provide a good housing for the speaker. A speaker without any housing is just causing local air movement (from the front of the conus to the back and v-v). Ideally a speaker should be mounted in an infinite, non-flexible plate. If you don't have that a box will do.

In message <CMM.0.90.2.876687404.billwEraseME.....puli.cisco.com>, William Chops Westfield <EraseMEbillwCISCO.COM> writes >There are a whole new set of audio amplifiers aimed a miniaturized >appliacnces (cell phones, cd players, etc.) They do away with output caps >AND double the power that would normally be available from a low voltage >supply by driving opposites sides of the speaker from mutually inverted >simple amplifiers (similar to the hack used to drive LCDs directly from >logic.) National Semi in particular has been agresively sending out samples >of their "BOOMER" audio amps like the LM4860/4861... > >BillW I like the Philips TDA7052, which uses this arrangement, and needs *no* external components. Leon -- Leon Heller: RemoveMEleonEraseMEEraseMElfheller.demon.co.uk http://www.lfheller.demon.co.uk Amateur Radio Callsign G1HSM Tel: +44 (0) 118 947 1424 See http://www.lfheller.demon.co.uk/rcm.htm for details of a low-cost reconfigurable computing module using the XC6216 FPGA

Myke, Here's two more, V3 -> o0 at Gnd, o1 at Vcc, o2 at High Impedance, V4 -> o0 at Gnd, o2 at Vcc, o1 at High Impedance, If the resistors are wired as shown here, you can parallel them to get more voltages. Obviously, you must choose your resistors wisely to get unique voltages. | o0 |--- R0 ---+--- Output | | | | | | o1 |----R1----+ | | | | | | o2 |----R2----+ | I am enjoying your book. Nice Job! Paul Date: Sun, 12 Oct 1997 10:27:40 -0400 From: myke predko <RemoveMEmykespam_OUTKILLspamPASSPORT.CA> Subject: Re: making sounds (with a pic and a speaker) Hiya, I just sent this reply back to Paul directly when I meant to send it to the entire list: Paul Webster Wrote: > Someone posted a two-resistor "ladder" array and stated it permitted >four levels or states. Wrong! Three output states for each resistor >makes nine, enough for quite detailed analogue sound output. Paul, could you explain how to do this with a Resistor ladder? The Best that I can do is n + 1 (where "n" is the Number of Resistors): | o0 |--- R0 ---+--- Output | | | R1 | | o1 |----------+ | | | R2 | | o2 |----------+ | With this circuit, I can get: Vcc -> o0 at Vcc, o1 = o2 = High Impedance Gnd -> o0 at Gnd, o1 = o2 = High Impedance V1 -> o0 at Vcc, o1 at Gnd, o2 at High Impedance, V1 = R1 / ( R0 + R1 ) V2 -> o0 at Vcc, o2 at Gnd, o1 at High Impedance, V2 = ( R1 + R2 ) / ( R0 + R1 + R2 ) Now, I suppose that multiple Outputs could be turned on to get parallel resistances, but I would be concerned about this with uneven current drive capabilities with different pins. This method would make it very hard to calculate the actual Voltage Outputs. There must be something I'm missing? Or is a circuit other than a ladder used? myke Check out "Programming and Customizing the PIC Microcontroller" at: http://www.myke.com ------------------------------

Myke Predko wrote: > Hiya, G'day! > I just sent this reply back to Paul directly when I meant to send it > to the entire list: No worries, I caught both, will forward to the list only. >> Three output states for each resistor makes nine, enough for quite >> detailed analogue sound output. > Paul, could you explain how to do this with a Resistor ladder? The > Best that I can do is n + 1 (where "n" is the Number of Resistors): * diagram of rather assymetric circuit bearing * * negligible resemblance to R-2R ladder! * Your circuit is a little confused. Let me propose again the use of two outputs, each with a resistor to a common output point (coupled to the LM386 by a capacitor) which is the centre point of an equal voltage divider between Vcc and ground. I specify that the resistors from the PIC outputs are related in a 3 to 1 ratio, and the values in the voltage divider are each equal to the lesser value of the other. Vcc | | R | | DC Block o1 |----R-----+------||--- Output | | | | o2 |----3R----+-- "common point" | | | R | Gnd If neither PIC output is enabled, the common voltage is ¸Vcc. If the output coupled to the lesser resistor is enabled in either direction, it delivers a voltage bias of Vcc/6 to the common point in that direction, while the other output produces a third of this bias in whichever direction. Calling this lesser bias increment "i" and referring this to ¸Vcc, the effects of the two outputs are: o1 Gnd O/C Vcc Nine states in total, four positive Gnd -4i -i +2i and four negative increments relative o2 O/C -3i 0 +3i to the zero point. If you REALLY want Vcc -2i +i +4i twenty-seven states, use a third output. Even a four bit (81 state) system would fit really nicely packed in a byte using the SWAPF instruction to extract first the port, then the TRIS value, and a lookup table might of course be used to read such values from a "phase" or syllable table. Something tells me an "R-3R" ladder would also be usable for this task. For the moment though, stock resistor values are sufficiently accurate. > Now, I suppose that multiple Outputs could be turned on to get > parallel resistances, but I would be concerned about this with uneven > current drive capabilities with different pins. I wouldn't! The PIC "Maximum Current Sunk" by an output pin being quoted as 20mA (though this is actually a maximum safe rating rather than a claimed drive ability) at 5V suggests an output impedance of 250 ohms. Since the input impedance of the LM386 is about 250k, the "R" above should be picked in the order of 270k (giving 3R a convenient value of 820k) and the current drive capability of the PIC pins is trivial. > This method would make it very hard to calculate the actual Voltage > Outputs. That's what scientific calculators or PCs are for! Cheers, Paul B.