Pic Programming for a pre-amp

[philip davies]

Hi

I'm trying to design a pre-amp by using the 16F877 to control a VFD display, source select buttons and standby button, also I'm trying to use a rotary encoder (optical type) to turn the volume up and down.

I have very little experience with programming pic controllers so any help would be very much apreciated.

Thank you

Phil

[Benj]

Hello Phil,

I would try and work on a small piece at a time to increase your confidence with the software and also so that you don't end up with a very large program that doesn't work.

Maybe working on the encoders would be a good starting point.

Eg can you display a numeric variable to a LCD.

Once you can do this then try getting the encoder to increment and decrement this number and update the LCD accordingly.

Once you have this sorted you can then move onto the next step.

[fotios]

Hi Phil
Why encoder for volume control? Are you intend to use those PGA2311 of T.I. (or something similar) for volume control?
Fotios

[philip davies]

Thank you both for your reply and suggestions.

Yes I am planning on using the PGA2311 chip to control the volume.

So far the program can switch on the pre-amp, show a "Power On" on the display and a "Power Off" message. I've startded working on the source select buttons but I'm trying to figure out how to get the outputs to the relays to latch on untill another source select button has been pressed.

Phil

[Benj]

Hello,

You could do something like this.

Loop Start

Read Switch 0
If Switch = 1 Out = 1

Read Switch 1
If Switch = 1 Out = 2

Read Switch 2
If Switch = 1 Out = 4

Read Switch 3
If Switch = 1 Out = 8

Output = Out

Loop End

[fotios]

Hi Phil
So your project is to control a DIY Hi-Fi audio preamplifier. Are you intend to include and remote control (RC5 of Philips is the most commonly used from DIYers) in your program? Also what type of relays have in your mind to use? Common type or Latching type relays? For source select you can use the same method like in TV remote control handsets (zaping or scroll mode) with just two buttons (UP-DOWN). This can obtained with the use of the internal EEPROM of micro. Please let me know, and i will show you.
Fotios

[philip davies]

Hi,

Yes i'm trying to put together a way of controlling my pre-amp with a pic.

I do have a remote which does use the Philip's RC5 code. It has 13 buttons on it:- power button, mute, 6 buttons labled from 1 - 6 (hopping to use those for selecting 6 inputs, 2 buttons for up and down for volume and 3 buttons labled f1, f2 and f3. I was hopping of having buttons on the pre-amp itsealf in order to control it (source/power on off). I would be using non-latching relays for controlling the input selection.

For volume control i have been looking on the internet and discovered a few projects that use some sort of rotary optical encoder and a PGA2310 to control the volume.

I would like to use a vfd display to show what's going on, i.e which source i've selected and the set volume in -dB (so starting in -oodb all the way to 0dB). so far my program is showing "Power On" when i press the on/off button then it clears to show my name and it then shows "Power Off" when i press the same button.

[fotios]

Hi Phil
I have ready a DIY audio preamplifier managed by a control unit which of the heart is a PIC16F887. Take a look here please: http://www.diyaudio.com/forums/analog-l ... s-fun.html. And here: http://eal.gr/MICROS.htm The only difference is that i use a motorized ALPS Blue-Velvet potentiometer instead PGA2310, Latching relays (are less noisy and they save power) and LCD display instead VFD. The level is shown in percentage (from 0 - 100%) instead dB. You should be carefull about the VFD display to be compatible with the control lines of LCD display, so as it can work with the standard commands of Flowcode for LCD displays. For example, this one is compatible: http://uk.farnell.com/noritake-itron/cu ... tt=1495416. I will upload the whole project here on this forum, i have to prepare some photos and schematics first, this will take me 1 to 2 days from now. For the remote control don't worry, RC5 of Philips is very old and is included in any universal remote control handset. I have included in my program a subroutine to automatically recognize and store the address of remote control (like in TV sets). I bought from a kiosk a Chinese universal mini handset - pendant with 6 buttons for 3 Euros, and it works just fine!
From those that i know from diyAudio forum, there is people that preffer PGA2311 and other people that preffer clasic motorized volume potentiometers. Personally, i preffer potentiometers. In practice the difference is that PGA2311 changes the volume level in precise dB steps, while carbon track potentiometers have a tolerance of +/-10%. OTOH, audio preamplifiers implemented either with modern ICs of National semiconductors, Analog Devices, TI - Burr Brown etc or with discrette devices (like mine preamp) have a bandwidth (frequency response) up to 1,5 - 2 MHz with extrmelly low noise, many times bellow -120dB! Those PGAs restrict the bandwidth at 200 to 300KHz and have a worse noise figure. But this is no issue for mid level Hi-Fi projects. So, you can use without fear PGA2310 or 2311. The issue is that i haven't - to the present - try to develop a mcu code for PGAs. If you are decided to use PGA2310, we will try (with the help of our good mentor Benj) to obtain the right Flow-code for this. To the present, i have to change the commands of my program to suit in common type relays. The next step will be the encoder and PGA2311 control.
Fotios

[philip davies]

Thanks for your reply and th links to the websites.

I already have a VFD display which luckily for me is the same one that I bought from farnell a while back. The reason I wanted to use normal non-latching relays is because I already have quite a few NEC 5v small signal relays lying around. I'm still very tempted to use a rotary encoder for the volume control, once I get the software to work with it that is.

I have been taking a closer look at your own personal design this morning fotios, I like what you did with the volume control. Could an Alps 6 channel variable resistor work for that design?

I did buy a Analog Metric remote control kit very cheaply on ebay a while ago. It has 2 pic microcontrollers, the first one (16F873A) seems to control the main curcuit board and input selector board, the second pic (18F252) is fitted onto the seperate display board. It uses a variable resistor to control the volume and an lcd display to show whats going on. Part of the variable resistor is fed directly to the display board but the display is very jumpy, i.e it flickers slightly when the volume is turned up or down manually or via the remote. Another problem i have with the kit is the number of inputs, it only has 4.

[fotios]

Hi Phil
For the local switches, as i told you, i have ready solution. In practice, you need just 4 buttons: PWRON/STBY - MUTE - INPUT (+) - INPUT (-). There is no reason you use separate switches to select each input (e.g. for 5 inputs 5 switches), "zaping" mode practically offers posibility to select among 256 inputs. I am also very interested for the encoder, so it is a chance and for me to learn it. You wrote that you have an optical encoder (is very expensive! for this work there is no need for so big accuracy IMO, an ordinary cheap encoder can do the job) which must be a rotary incremental encoder with two quadrature outputs (out of phase between them). From those i have seen on the web, almost all people use ordinary encoders with electrical contacts (not optical) and they use just one of the two outputs for count (usually is modified to cause a Low to High transition interrupt in the MCU) while the other is used to determine the direction of count (when is Low incremental and when is High decremental). With this method, only the half steps of a full revolution of encoder are detected, but this is no issue for the current application. By looking carefully at PGA2310 datasheet (as well in DIY remote control kits using PGA on diyAudio forum and Head-Fi.org) i have seen that just 3 control inputs of it are used: SCLK (clock), SDI (Serial Data Input) and MUTE. The format of Serial Data is simple, 16bits of which the first 8 bits represent the volume of Left channel and the second 8 bits the volume of right channel. The issue is that i don't know the way to convert a byte stored in a register of MCU in serial format to be transmitted. I have to study it, maybe i will ask the help of experts, i don't know yet. Maybe is very simple by using successive AND operations (successive Masking of bits). There is also the need of a clock source, that is easy. The formula that uses PGA2310 to determine the gain is: Gain(dB) = 31.5 - [0.5 X (255 - N)] that means when N=1 Gain=-95.5dB and when N=255 Gain=+31.5dB. It should be noted here, that PGA2310 can offers gain, not only attenuation. Usually gain is needed only in professional audio devices, in Hi-Fi not because bandwidth and noise figures are worsen proportionally with increasing of gain.
Another one issue is that Flowcode does not include rotary encoder in its menu of devices for simulation purpose. To this, i did a Flow-chart in which i have simulated the rotary encoder with 2 switches. The first is a momentary switch that simulates the A output of encoder and is connected at the RB.0 of P16F877. The Port interrupt or the IOC = Interrupt On Change state as is reffered from Microchip, is enabled. Each time the switch is pressed, the variable "counter" increments or decrements and it is printed on LCD. The second switch is a toggle switch that simulates the B output of encoder. The variable "counter" increments when switch is at Low position. When the switch is turned at High position the variable "counter" decrements. I have included the EEPROM of P16F877 to store each time the value of counter, this will be usefull later during development of Flowcode, we have to include a lot other things. You could take this Flowcode as the foundation stone on which we will build the whole project.
Stay tuned

[philip davies]

Hi Fotios

First of all thank you for the attachment that you sent me. I have also been loocking on the internet to see if I can find out more regarding the rotary encoder that I have. So far i've came up with the same thing regarding the PGA2310, i.e it has 3 connections to control it and that it can be used to provide gain but I have also seen on some pages that you have to provide some sort of pre-amp inorder for the PGA2310 chip to provide the gain.

Also I think i'm going to take your advice on the "zapping" mode as it has given me a nice idea for the control panel, also it will look much neater than having a load of buttons cluttering the front panel but I would still like to use non-latching relays because i have a few new ones lying around from various other projects.

One thing i forgot to ask you, does the numbers and letters flicker as you turn the volume up and down on your pre-amp design?

Again i would like to say thank's for the help you have given me

[fotios]

One thing i forgot to ask you, does the numbers and letters flicker as you turn the volume up and down on your pre-amp design?

No Phil, i haven't a such issue, by no way. In the beginning, i had flickering of the volume value but i resolved it with improvements of program. By any way, flickering of display it shows a deficient program code.
I am glad that you like the zapping - scroll - mode for input selection. As for relays, don't worry it is very simple to modify the program code to drive common relays. The thing is simple: Latching relay has two coils, set and reset, so you need 2 outputs of mcu to drive it and momentary pulses of 15-20msec to change state (changeover of contacts). Supposing that the set coil is connected at port D.0 and reset at port D.1, the command to SET it is of this form:
1) Output "1" --> D.0
2) Delay "20ms"
3) Output "0" --> D.0
and so on.
Therefore, to drive a common relay instead a latching one simply we have to delete the commands 2 and 3. Tell me something: Have you the possibility to change your PIC16F877 with a PIC16F887? It is really usefull for the current project.
Have a nice day.
Fotis

[fotios]

So far i've came up with the same thing regarding the PGA2310, i.e it has 3 connections to control it and that it can be used to provide gain but I have also seen on some pages that you have to provide some sort of pre-amp inorder for the PGA2310 chip to provide the gain.

Looking at the block diagram of PGA2310, you could see that each channel is formed by a front-end voltage divider network, followed by a non-inverting voltage gain stage. Both are controled from the logic circuit included. The voltage divider simply provides "attenuation", while the non-inverting OPA has a minimum gain of 2, given by the formula: Av = 1 + Rf / Rg where Rf is the feedback resistor and Rg is the resistor connected accross the inverting input of OPA and Gnd. Rf is variable, and it should has a minimum value equal to Rg (that is a law in OPAs) so when Rf=Rg --> Av = 1 + 1 = 2 or +6dB. Given that PGA2310 has a max gain of 31.5dB, it means that the variable Rf should be 38 times greater than Rg. From those that i know bandwidth and THD+Noise figures are worsen proportionally with the gain increasing. OTOH, a gain of 2 (+6dB) it is small to cause derating of PGA2310, unfortunatelly all the plots given in datasheet are refered at 0dBFS (dB Full Scale which is a relative scale, in absolute scales like dBV - refered to 1Vrms - the numbers worsen, i know this very well from my FFT analyzer). I haven't looking carefully at the PGA2310 projects on the web. Probably, the pre-amp stage before PGA2310 that you have seen, will be a voltage follower (Av = 1) buffer stage to provide a low input impedance to it. In datasheet is reffered that the source impedance should be less than 600Ω, unless PGA2310 performance is derated.
BTW, if we had a rotary encoder combined with push button we could use it and for input selection. Thus,we could have a front panel with just 2 buttons, a rotary knob, a display and an IR receiver (the last - maybe - hidden behind the protective transparent cover of display). Later, i will upload the Flowcode of input selection with 2 buttons.
Fotis

[philip davies]

Hi Fotis

Thank you for your reply.

Yes I could change the chip for a PIC16F887, there's no reason why I couldn't. I would be very great full if u could post the code for the 2 button in-put selector.

In your honest opinion, which method of volume control would be best for sound quality, a volume control pot or a PGA2310?.

Thanks.
Phil

[fotios]

Hi Phil
Nice to see that you haven't problem with MCU change. I changed it already in P16F887, later on you will see its benefits. I did not have time - until now - to deal with the Flowcode for Input selector, because i was busy with the encoder program. I have obtained so far a software that counts the encoder steps starting from number "2" = -95dB. I did a trick on program for the avoidance of fractional numbers (e.g. -85.5dB), so only integer dBs are calculated. The trick is that, each step of encoder increments or decrements by "2", e.g. 2 -4 - 6- .... - 250 -252 - 254. According to the formula given on PGA2310 datasheet, those steps correspond exactly to integer dBs: -95, -94, -93 ... +29, +30, +31. Perfect! Then, the program prints on the lower line of display the Gain value of both channels. In the upper line of display is printed - to the present, this will be removed later - the step number of encoder. You could see that the correct printing on display it takes up most of the program and maybe this can resolve your query about display flickering issues, as i told you the code must be writen with big care. I have to add some commands so as numbers and signs will be attached continuously, i.e. no empty squares between them. Going further, inside the main loop is included a calculation box that translates the current decimal value of the variable "counter" to a hex number. This hex number is registered in a new variable named "Gain_h" and is ready for convertion in serial format to be transmitted on the "SDI" line of PGA2310. I attach the FCD, that you can run in simulation mode, it is true!
Regarding your question about volume control, the answeer is relative: I have adopt the use of discrette devices in my designs 5 years ago. You have seen my discrette OPA in diyAudio forum that works with +/-24V supply rails. Although my preamp does not offer gain (except the RIAA stage that has a gain of 1000) it offers an amazing "Headroom" thanks to the high level of supply rails. If you take into account that there is not any IC that can sustain a supply level greater than +/-22V (i would say +/-20V to be safe) i can't use ICs from the hand. Some people make mixed use of discrettes and ICs by reducing supply level at +/-18 to +/-20 Volts. IMHO this reduces the performance of discrette OPA. Almost all famous audio designers, they claim that the use of discrette OPAs is reasonable only in RIAA (Phono preamp) stages because their super low noise, down to -140dB many times (mine OPA has a noise floor of -130dB). But you wonder, why they use in their expensive Hi-End class preamps at whole discrette OPAs? Personally, if i had to make with a preamp implemented with ICs - be it so with e.g. those super LM4562 or LME49720 etc - i would use without hesitation the PGA2310 for volume control. It is perfect for this work, it offers precise steps in dBs that can be displayed very easy on a display. Using an analog logarithmic potentiometer, this work is a real PITA! For this reason i did the trick in my preamp with the second pot and the volume is displayed in percentage. I had a thought (like you) to use a quad log mot pot, to use the 3rd or 4th pot for the tracking of rotation angle. You do or you don't believe, the formula of computing voltage levels logarithmically changed and translated in dBs is a huge subroutine inside the whole program. That is my opinion, nothing more nothing less. I'll be back tomorrow with the FCD of input selection.
Have fun and good night!
Fotis

[philip davies]

Hi Fotis

Thanks for sending me the attachment for the encoder. I shall have a look at it to see how it works.

That was another reason why I thought it would be a good idea to use a PGA2310 chip for the volume control because it adjusts the volume in precise dB steps.

Again thanks very much for the attachment and I look forward to hearing from you tomorrow.

Phil.

[fotios]

Hi Phil
The program for input selection is ready, tested (both in simulator and in active hardware) and working. I posted it here: http://www.matrixmultimedia.com/mmforum ... =29&t=8604. The reason is that, maybe there is and other people out there looking for a similar project and if posted as a simple reply to you, will be hidden thus i thinked that is better to upload it as a new thread. If you have any issues to read it or to download the fcf file, please let me know to help you.
Fotis

[philip davies]

Hi Fotis

Well i need to say a big thank you for all the help you have given me on my project. You have been a huge help for me to get a step closer to having a working project. I have had a play with the FCD of the encoder and the input selector and I would like to ask some questions,

firstly how do I use flowcode to bring the two FCD's together,

secondly can i change the display layout slightly so that the top left hand corner of the display can show the selected input and the top right hand corner can show the set volume and the bottom line can be used to show "Mute"

I did put together a simple FCD for the on/off switch and also I managed to get the display to show "Power On" then it changed to "Philip Davies", also i got the display to show "Power Off" when the on/off switch is pressed again.

Thanks again Fotis for your help.

Phil

[fotios]

You're welcome Phil, that was a chance and for me to learn new things. Have you thinked that PGA2310 offers posibility and for Balance control? Though i consider it not necessary for home audio. For pro-audio yes, it is useful e.g. in mixing desks for Pan-Pot (i was busy for 25 years with professional audio equipments). Please let me know. If you don't need balance, then your proposal about single level set indication is reasonable because is same for both channels. You should be patient, the project is enough complex it is not finished with simply implementing volume control and input selection. We have to incorporate and the remote control function. During Power-On of preamp, there is a preparation of device. The same during Power-Off, or better Stand-By. Printing on display of anything is an easy work using flowcode, the only needed is big care in the sequence of commands. In my preamplifier, i use separate supplies for the digital and analog part. On the back panel, there is a Mains switch and when is turned ON the program puts the preamplifier in Stand-By mode, just a red Led lits on the front panel the analog power supply does not activated. To this, you should press the Power-On/St-By button. Then the mcu activates the preamplifier. If you press again this button, preamp is deactivated and on the display is printed STAND-BY for 2 sec, then display is turned off and lits again the red Led. I will study tomorow your demands, everything is posible. You know, rotary encoders equiped with push-button offer bigger flexibility: Press the knob long, and is activated the MENU. By turning the knob you can navigate now between: Volume set, Input select. Press again the knob to confirm item selection, make your settings etc. Press the knob short, and e.g. mute the device. Etc, etc.
The main issue that bothers me, is the volume control. I don't like neither motorized pots neither PGAs. Not for precission, for resistance. Smallest values of both, is 10KΩ. From my experiments, i found out that volume pots derate the linearity of the following stage if they have resistance greater than 1,5KΩ. Pots of 1KΩ are the best, but you can't find a motorized one. The best solution that combines both precision and custom made resistance, are rotary switches of 24 places driven by a stepper motor. There are many small companies offering such type pots, e.g. GoldPoint without motor and exposed contacts. I don't like it, frequent cleaning of contacts. You have probably seen the different style using 8 relays and R/2R ladder networks. These are noisy due to relays and create many loops due to combining of many resistors in series/parallel connection. I am working 1 month ago, on a different style stepper attenuator using pasive reed relays (activated by natural magnet, so no noise) driven by a stepper motor. I try to make it as small as possible in dimensions. Up to now, i have completed just the mechanical part formed from a spindle and two sliding rods. I have to do a lot of work on this.
Good night
Fotis

[philip davies]

Hi Fotis

Yes I had thought about implementing a Balance control but I was going to use a small pot and hide it around the back of the pre-amplifer because it's something that is offered on various pre-amplifiers and amplifiers but i never use it unlike as you said the "Pan" control on mixing desks. I didn't actually know the PGA2310 offered this function.

I have read on various websites stating that it's very important to have a seperate PSU for the digital circuit and the analog circuit due to the oscilator on the control board. Having a push button encoder to activate different inputs and menus sounds very interesting and I think it will add a very nice touch to the finished design although it's not something i've considered.

I have seen this type of stepper motor driven rotary switch before, if I remember corectly the company that I have seen seling it is called Bent Audio. Also i've seen the relay driven volume control system on a website called Hificollective. I believe a company called Tentlabs sell one as a kit. Your home made stepper attenuator sounds very interesting, reed relays are usually sealed therefore you wont have the drawbacks of having to clean oxidized switch contacts. You can get some reed relays that have a coating of mercury on they're contacts which apparently is good for carying small signals such as audio signals, but I dont know if this is true, I have never used them for audio.

Thanks again for your help

Phil

[philip davies]

Hi fotis

Sorry for troubling you again, I wanted to ask you is it possible to have a look at the code for your own pre-amp design, don't worry if its not available.

Thanks again for your help

Phil

[fotios]

Hi fotis

Sorry for troubling you again, I wanted to ask you is it possible to have a look at the code for your own pre-amp design, don't worry if its not available.

Thanks again for your help

Phil

Hi Phil
I haven't problem to post the FlowCode *fcf file of my preamplifier, just i fear that you will confused because its structure is absolutelly different from what you want. Please, stay tuned with PGA2310 to the present, i am working on this project: http://www.matrixmultimedia.com/mmforum ... =29&t=8634. Please be patient, in 2 days - hope so - i will post an upgrade including Input select.
To the present, you have a good chance to study the project of the above link.
Thanks for your patience
Fotis

[philip davies]

Hi Fotis

Thank your for posting the link, I will have a closer look at it when I get home

Thanks again for your help, it is very much appreciated.

Kind Regards

Phil

[fotios]

Hi Fotis

Thank your for posting the link, I will have a closer look at it when I get home

Thanks again for your help, it is very much appreciated.

Kind Regards

Phil

Hi Phil
A request: Please redirect your replies to the new thread http://www.matrixmultimedia.com/mmforum ... =29&t=8634 that i have open.
Thank you very much.
Fotis

[philip davies]

Hi Fotios,

I will post any replies to your new thread but for now I can't as I have just activated the licence on my copy of Flowcode in order to post on the Flowcode V4 threads.

I have been taking a much closer look at the fcf file you posted up and also I have been reading the article you posted very cearfully in order to try and understand how it works. I do have a couple of questions to ask about how the code works but i'm going to wait untill you have posted the second part of the article and also instead of posting them on here i'll post them on your new thread. For now i'm going to continue experimenting with Flowcode and with the fcf file you posted to try and learn more.

Thanks again Fotios

Phil