Test mode A asymmetry?
SPK L <-> Output Switch RIGHT:A don't have its corresponding connection for SPK R.
Symmetry is paramount.
/
SPK L <-> Output Switch RIGHT:A don't have its corresponding connection for SPK R.
Symmetry is paramount.
/
thanks, tomorrow I'll post schematics
look, BC is cooking something similar 😀
CordellAudio.com - A Level-Matched Speaker/Amplifier A-B Switch
look, BC is cooking something similar 😀
CordellAudio.com - A Level-Matched Speaker/Amplifier A-B Switch
If you are serious about this make it an A-B-X box. Make a setting where you switch but don't know if you go to A or B.
Then, after familiarizing yourself with A and B, switch to 'X' and try to decide whether it is A or B. Do that a bunch of times, then count the times you got it right.
If you get it right consistently, you can be pretty sure that there is an audible difference.
Jan
Then, after familiarizing yourself with A and B, switch to 'X' and try to decide whether it is A or B. Do that a bunch of times, then count the times you got it right.
If you get it right consistently, you can be pretty sure that there is an audible difference.
Jan
I think ABX is a bit tricky. When you switch A to B, you can hear relay clicks (in defined sequence). Switching A->A (B->B) is not audible, or at least generates different sound sequence, so the listener can easily guess.
One can use SSR or fake relay (not in signal path) to overcom this problem.
SWITCH A->B
1. mute input A (B is already muted)
2. switch input to B
3. switch output to B
4. unmute input B
SWITCH A->A
1. mute input A
2. unmute input A
SWITCH A->A
1. mute input A
2. switch fake input relays
3. switch fake output relays
4. unmute input A
or another idea, generate some sound/noise to mask relay clicks
One can use SSR or fake relay (not in signal path) to overcom this problem.
SWITCH A->B
1. mute input A (B is already muted)
2. switch input to B
3. switch output to B
4. unmute input B
SWITCH A->A
1. mute input A
2. unmute input A
SWITCH A->A
1. mute input A
2. switch fake input relays
3. switch fake output relays
4. unmute input A
or another idea, generate some sound/noise to mask relay clicks
Last edited:
Obviously you seem to be on top of things. Do the clicks come from the actual relays or via the signal chain and speakers?
//
//
Another way to do make an A-B switchbox is to use the PGA2311 preamp kits that are available on eBay, along with an Arduino CPU to replace the controller for those kits. The PGA2311 provides the volume compensation for each speaker selection, and there are plenty of extra outputs on the Arduino chip to toggle the relays. There is a "relay bus" output for controlling up to 4 separate relays, or you could encode/decode the 4 lines to control more relays or other devices.
The software and more info is at: CPU for PGA2311 Preamp – Audiodevelopers Reborn
It's worth noting that the relay contact resistance needs to be very low for valid A/B testing. I usually use good quality 40A relays to ensure low resistance. The smaller 5 or 10A relays can have contact resistance in the .1 to .05 ohm range, which is not good enough. The resistance results in a sometimes non-linear voltage drop that is superimposed on the signal. If the speaker is 4 ohm and you want that voltage drop to be 60dB down, the resistance needs to be less than .005 ohms. For that you need a relay with a very large contact area and low coil resistance to push hard on the armature. There shouldn't be a "click" when you switch relays; it should be a resounding "clunk" 🙂
I can post the schematics and PCB's for the boards in the relay box if there is interest. The boards were made many years ago and the files are for ExpressPCB. They should probably be updated to use a Bluetooth interface rather than a "relay bus", but I'll let someone else do that work...
The software and more info is at: CPU for PGA2311 Preamp – Audiodevelopers Reborn
It's worth noting that the relay contact resistance needs to be very low for valid A/B testing. I usually use good quality 40A relays to ensure low resistance. The smaller 5 or 10A relays can have contact resistance in the .1 to .05 ohm range, which is not good enough. The resistance results in a sometimes non-linear voltage drop that is superimposed on the signal. If the speaker is 4 ohm and you want that voltage drop to be 60dB down, the resistance needs to be less than .005 ohms. For that you need a relay with a very large contact area and low coil resistance to push hard on the armature. There shouldn't be a "click" when you switch relays; it should be a resounding "clunk" 🙂
I can post the schematics and PCB's for the boards in the relay box if there is interest. The boards were made many years ago and the files are for ExpressPCB. They should probably be updated to use a Bluetooth interface rather than a "relay bus", but I'll let someone else do that work...
The control unit board is under manufacturing. I'm planning to implement ABX test too. The randomized path selection will be based on functions random(), randomSeed() and two analog inputs floating data. The selected path info will be sent through usb port to the connected PC.
Attachments
Last edited:
Nice work. I recently built an Arduino project for some voltage measurements. I don't want to send this thread off track, but do mind letting me know what software you used for your board layout? Getting a board fabricated is my next step and it is new to me.
- Status
- Not open for further replies.