I realize that but strange things...or actually stranger things can happen during power spikes, and or glitches!! Put it all together and let us know how it works out. I may give this a try too!! Come to think of it....we could be switching all those relays with a cheap selector switch too.......
Mark
Mark
Here is my version of a ADC8004 based attenuator. The schematics are 95% finished at this time, but i thourght I'd show it to you anyway. Need to do the relay switches and add protection diodes etc.
The ADC0804 is a uP compatible ADC, but supports 'free running' mode. This means the ADC can run without a uP, but needs to be started up. The interface pins CS RD and WR does that. CS must be tied permanently to ground. RD and WR must be grounded shortly to make the ADC start the input sampling and output. This is what T1 and T2 (BC547) does.
I use all 8 bit from the ADC. Sometimes there is a instability in the LSB and the ADC can't decide wether to turn the LSB on or off when the pot is 'between values'. This is solved by feeding the LSB to a Flip-Flop (4027) and a small portion of the output from the FF back to the pot (got the idea from another thread). How much is fed back is determined by the ratio of R6 and R15. The FF will then add a small voltage to the input every second time the LSB shifts, enough to make the signal stable. Ofcourse this will add a instability by it self, but will settle within 2 or 3 'shifts'. I haven't calculated the values yet.
The ADC0804 is a uP compatible ADC, but supports 'free running' mode. This means the ADC can run without a uP, but needs to be started up. The interface pins CS RD and WR does that. CS must be tied permanently to ground. RD and WR must be grounded shortly to make the ADC start the input sampling and output. This is what T1 and T2 (BC547) does.
I use all 8 bit from the ADC. Sometimes there is a instability in the LSB and the ADC can't decide wether to turn the LSB on or off when the pot is 'between values'. This is solved by feeding the LSB to a Flip-Flop (4027) and a small portion of the output from the FF back to the pot (got the idea from another thread). How much is fed back is determined by the ratio of R6 and R15. The FF will then add a small voltage to the input every second time the LSB shifts, enough to make the signal stable. Ofcourse this will add a instability by it self, but will settle within 2 or 3 'shifts'. I haven't calculated the values yet.
use a ULN2803 relay driver -- less real estate and cheap. in fact, you could probably use any of the NPN quad transistor chips.
secondly, you need a capacitor on the Vcc of the ADC -- I use 4.7uF tantalum.
third, consider bypassing the wiper of the potentiometer with a 100nf cap or something. you will see if you can very carefully connect a scope probe to the analog input that there is a lot of noise as you move the wiper. (to really see this you might want to take a piece of coax and solder it directly to the analog input, with the braid soldered to the analog ground, unless you have one of those very expensive tektronix probes.)
I have to run off to a meeting, but will explain how to turn a linear pot into a log (a real, not quasi-log) when I return.
secondly, you need a capacitor on the Vcc of the ADC -- I use 4.7uF tantalum.
third, consider bypassing the wiper of the potentiometer with a 100nf cap or something. you will see if you can very carefully connect a scope probe to the analog input that there is a lot of noise as you move the wiper. (to really see this you might want to take a piece of coax and solder it directly to the analog input, with the braid soldered to the analog ground, unless you have one of those very expensive tektronix probes.)
I have to run off to a meeting, but will explain how to turn a linear pot into a log (a real, not quasi-log) when I return.
Relays
The ADC0804 costs about 3.5$, at least here in DK. So i guess the relays will be the major expence.
Speaking of relays, any ideas? I have my eyes on some realy nice Omrons all capable of switching 10uA with gold plated silver contacts. I think I will by from Elfa, as their selection and prizes seems ok. Anyway, there are a few to choose from with almost identical data, but some cost nearly twice as much. Any advice?
The ADC0804 costs about 3.5$, at least here in DK. So i guess the relays will be the major expence.
Speaking of relays, any ideas? I have my eyes on some realy nice Omrons all capable of switching 10uA with gold plated silver contacts. I think I will by from Elfa, as their selection and prizes seems ok. Anyway, there are a few to choose from with almost identical data, but some cost nearly twice as much. Any advice?
Has any of you given any thought to visual level indication? The design I have been thinkig about incorporates two BCD counters and two 7-segment displays. Level adjustment is done with a rotary “giver thingy” (sorry don’t know what it’s called over there), alternately with two switches. It’s kinda on hold right now cause I cant find a way to get any more then 44db of attenuation, and the attenuation is fare from logarithmic. (I’m not sure I calculate it correct though. Am I right in assuming that if the relays are switched like in the aleph P, it forms a voltage divider?)
Hi there,
I saw a lot of useful tips and designs so far. A simple digital volume control is now in reach.
Now i get a lot of new ideas : a led matrix display for source indication and volume levels and maybe even a remote controlled option. A friend of mine is really crazy about ATMEL processors and he told me that my goal could be realised with such a microcontroller. Alreay built a test board end serial programmer that works.
The only problem is that programming isn't my thing.
Etchting a pcb and soldering works are no problem.
Anyone having no trouble with programming but with etching or so.... maybe a trabe is possible ???
To all you DIY'ers : good luck with your projects.
Greetz,
Nick
I saw a lot of useful tips and designs so far. A simple digital volume control is now in reach.
Now i get a lot of new ideas : a led matrix display for source indication and volume levels and maybe even a remote controlled option. A friend of mine is really crazy about ATMEL processors and he told me that my goal could be realised with such a microcontroller. Alreay built a test board end serial programmer that works.
The only problem is that programming isn't my thing.
Etchting a pcb and soldering works are no problem.
Anyone having no trouble with programming but with etching or so.... maybe a trabe is possible ???
To all you DIY'ers : good luck with your projects.
Greetz,
Nick
Re: Relays
Don't take my word for it but there seems to be quite a lot of agreement on the importance of the quality of a relay used for switching audio. AFAIK relays usually don't have any problems switching higher current but when switching lower current the contact is less reliable. You need to use a relay intended for low level signals and not one designed for higher currents. Omron has quite a few of these and some of the Elfa relays state use for Audio. Otherwise Schuro has really good prices on relays they claim to be good for the purpose. Select carefully.
Thomas said:
Don't take my word for it but there seems to be quite a lot of agreement on the importance of the quality of a relay used for switching audio. AFAIK relays usually don't have any problems switching higher current but when switching lower current the contact is less reliable. You need to use a relay intended for low level signals and not one designed for higher currents. Omron has quite a few of these and some of the Elfa relays state use for Audio. Otherwise Schuro has really good prices on relays they claim to be good for the purpose. Select carefully.
Mark A. Gulbrandsen said:
No, but it should be simple enough. I'm almost finished with the schematic. I think I can send the final schematic tomorrow and maby also the PCB layout. Don't worry, this one will be single sided.
By the way, I'm not using the relay driver chip, but good ol' transistors n' diodes. Almost any (NPN) small signal transistor can be used, but mind the pin config.
Well, less talk, more circuit design.
Thomas,
Am looking foreword to seeing the rest of your project so I can build it. This weekend I am going to build(hand wire) the Jens version. It is also very inexpensive to build....I spent less than $5.00 U.S.!... But I also happen to have a nice 256 ppr shaft encoder laying around that I can use for it. I will also build yours next, then compare the two. I can use the same relays as I'm building that seperately for testing purposes. I don't like the VU meter chip idea so much as I've never seen one that the LED is either on or off. They always seem to fade in or out a bit....that effect might cause problems.
Mark
Am looking foreword to seeing the rest of your project so I can build it. This weekend I am going to build(hand wire) the Jens version. It is also very inexpensive to build....I spent less than $5.00 U.S.!... But I also happen to have a nice 256 ppr shaft encoder laying around that I can use for it. I will also build yours next, then compare the two. I can use the same relays as I'm building that seperately for testing purposes. I don't like the VU meter chip idea so much as I've never seen one that the LED is either on or off. They always seem to fade in or out a bit....that effect might cause problems.
Mark
Ok, here it is!
I hereby release the 'finished' schematics. Quotaion signs because I actualy never tested if it works. This means I don't recomend you to build this, but you are invited to do so on your own. If you can wait a day or two, I will produce tested PCB artwork for this project to be published here.
Component values should not be critical. Most of them are taken from datasheets. The ratio of the 4027 'feedback' resistors should be around 1 Million. If you make a 6 or 7-bit version (by not installing all relays) you can omit the 4027, R6, R15, R16 intirely.
Calculate the relay output resistors (R17-R32) for your own use. Look up the Aleph manual if that's what you are going to use it for.
It's almost 3:30 AM and I'm going to sleep now. I need some sleep now. I desperate need some sleep! I think I'm just going to bed and sleep some.
I hereby release the 'finished' schematics. Quotaion signs because I actualy never tested if it works. This means I don't recomend you to build this, but you are invited to do so on your own. If you can wait a day or two, I will produce tested PCB artwork for this project to be published here.
Component values should not be critical. Most of them are taken from datasheets. The ratio of the 4027 'feedback' resistors should be around 1 Million. If you make a 6 or 7-bit version (by not installing all relays) you can omit the 4027, R6, R15, R16 intirely.
Calculate the relay output resistors (R17-R32) for your own use. Look up the Aleph manual if that's what you are going to use it for.
It's almost 3:30 AM and I'm going to sleep now. I need some sleep now. I desperate need some sleep! I think I'm just going to bed and sleep some.
Relay Driver and LCD
to "gild" the lilly, consider using the LCD+ from www.basicx.com in this project. it'sa serial 4*20 LCD with a relay driver (ULN2803 - used in robotics among other things). I have used a couple of these, they are extremely easy to implement.
In fact, you could drive your entire preamp setup with a pair of ULN2803 relay drivers and a BX24 microprocessor.
BasicX is a little like Visual Basic -- their basic microprocessor has 4 - 10 bit ADC's so you could poll 4 potentiometers. It also has a LOG function, so you could take the output of the potentiometer, take it's logarithm to derive the correct relay settings, and you could display it all on the LCD.
The LCD+ costs $59, the BasicX BX24 microprocessor costs $49, but you should get the development kit which runs $100.
While some folks cringe at the price of the Basic Stamp BS2, BX24, Atoms the fact that you can reprogram them on the flyis a real advantage in a one-off project. (I use a PICStart with my 16F8XX chips)
to "gild" the lilly, consider using the LCD+ from www.basicx.com in this project. it'sa serial 4*20 LCD with a relay driver (ULN2803 - used in robotics among other things). I have used a couple of these, they are extremely easy to implement.
In fact, you could drive your entire preamp setup with a pair of ULN2803 relay drivers and a BX24 microprocessor.
BasicX is a little like Visual Basic -- their basic microprocessor has 4 - 10 bit ADC's so you could poll 4 potentiometers. It also has a LOG function, so you could take the output of the potentiometer, take it's logarithm to derive the correct relay settings, and you could display it all on the LCD.
The LCD+ costs $59, the BasicX BX24 microprocessor costs $49, but you should get the development kit which runs $100.
While some folks cringe at the price of the Basic Stamp BS2, BX24, Atoms the fact that you can reprogram them on the flyis a real advantage in a one-off project. (I use a PICStart with my 16F8XX chips)
Hey Jackinj,
Can you start a thread about implementing one of these? I'd be very intrested in trying that out.......my only negative thought is all the noise the processor generates, and then having that in with the audio stuff. I think other people on this thread have designed these other controllers because of that very problem. Still it would be an interesting thread in and of itself that many would find useful.
Mark
Can you start a thread about implementing one of these? I'd be very intrested in trying that out.......my only negative thought is all the noise the processor generates, and then having that in with the audio stuff. I think other people on this thread have designed these other controllers because of that very problem. Still it would be an interesting thread in and of itself that many would find useful.
Mark
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.