I was not sure in which section of the forum I should post this, but I ended up posting it in the tube section because the circuit is tube- based.
I've done some electronics projects before, built a simple tube preamp for a carbon mic. I have a basic understanding of electronics and passive and active components, can read schematics and I'm quite good at soldering. I've never designed any circuits of my own though, and need help in order to make some decisions and design some relay control circuitry.
The audio circuit is based on a EF 86 and two ECC83's, and has a few different switchable pre-emphasis curves. Instead of using rotary switches, I want to use relays to switch between the different curves. Has anyone on this forum done something like that before?
I'm afraid I can't post the whole circuit right now, because the designer is a quirky little old man, and might object if I do so without his consent. Anyway, I attached a section of the schematic to illustrate my "problem".
partofphonopreschem by mrkrabbz, on Flickr
In the original design, a five-pole rotary switch with five positions is used to simultaneously switch at five points in the circuit (a - e) . My first idea was to cascade four DPDT relays for points a and b, as well c and d. The first relay would switch between a disconnected pin and the next relay. The second relay would switch between the next two relays, which would yield the required 4 different paths.
I would switch the relays with some transistors controlled by an Arduino board. The Arduino will also use a LCD display to tell what preemphasis curve is in use!
An electronics shop in my town has these SMD relays on sale:
MEISEI PS-5
DATASHEET
Are these relays good enough to be used in a sensitive circuit like this phono preamp? Are there any general considerations that would forbid using (cascaded) relays and/or the Arduino in this context?
I will appreciate any kind of input!
cheers,
Bob
I've done some electronics projects before, built a simple tube preamp for a carbon mic. I have a basic understanding of electronics and passive and active components, can read schematics and I'm quite good at soldering. I've never designed any circuits of my own though, and need help in order to make some decisions and design some relay control circuitry.
The audio circuit is based on a EF 86 and two ECC83's, and has a few different switchable pre-emphasis curves. Instead of using rotary switches, I want to use relays to switch between the different curves. Has anyone on this forum done something like that before?
I'm afraid I can't post the whole circuit right now, because the designer is a quirky little old man, and might object if I do so without his consent. Anyway, I attached a section of the schematic to illustrate my "problem".
partofphonopreschem by mrkrabbz, on Flickr
In the original design, a five-pole rotary switch with five positions is used to simultaneously switch at five points in the circuit (a - e) . My first idea was to cascade four DPDT relays for points a and b, as well c and d. The first relay would switch between a disconnected pin and the next relay. The second relay would switch between the next two relays, which would yield the required 4 different paths.
I would switch the relays with some transistors controlled by an Arduino board. The Arduino will also use a LCD display to tell what preemphasis curve is in use!
An electronics shop in my town has these SMD relays on sale:
MEISEI PS-5
DATASHEET
Are these relays good enough to be used in a sensitive circuit like this phono preamp? Are there any general considerations that would forbid using (cascaded) relays and/or the Arduino in this context?
I will appreciate any kind of input!
cheers,
Bob
Looking at the datasheet side-by-side with the datasheet for a (randomly selected, Jameco) multipole rotary switch I see that the relay contact resistance is worst case 50 m-ohm. Contact resistance for the rotary switch is 20 m-ohm.
The relay shows a distribution graph for the resistance with the majority of the samples being just over 40m-ohms. The rotary switch shows no distribution for this parameter which probably says something about how reliably this parameter is characterized for the 2 parts.
Obviously it would be preferable to have a contact resistance of 20 m-ohm as opposed to up to 5 series resistances with a worst case total of 250 m-ohm, but the schematic extract shows associated resistances in that area of the circuit in the tens and hundreds of k-ohms, so the contact resistance is probably of no significance.
Capacitances are, however as low as 100pF. You should make every effort therefore to make the layout as compact and neat as possible to minimize strays. If the circuit works satisfactorily with 5-way rotary switches which will inevitably have some point-to-point wiring and consequent strays associated with them, then it will probably be OK with the relays, but it doesn't hurt to take care. Sticking to SMT parts will help to keep things compact.
One consideration might be that rotary switches come in make-before-break and break-before-make flavours and this may affect whether there is 'popping' as the settings are changed. You can obviously arrange how the switch timings occur using a uproc.
You do have to be careful with comparatively high-impedance circuits like these to keep the uproc from getting noise into them. Being high impedance, comparatively small currents can induce detectable voltages. Noise at radio frequencies can still intermodulate with audio. Close attention to supply bypassing and keeping signal/ground return loop areas small will minimize these problems, and separate digital and analog grounds joined at a single judiciously chosen point should be employed. Keep the 2 sections physically separated as far as possible.
The relay shows a distribution graph for the resistance with the majority of the samples being just over 40m-ohms. The rotary switch shows no distribution for this parameter which probably says something about how reliably this parameter is characterized for the 2 parts.
Obviously it would be preferable to have a contact resistance of 20 m-ohm as opposed to up to 5 series resistances with a worst case total of 250 m-ohm, but the schematic extract shows associated resistances in that area of the circuit in the tens and hundreds of k-ohms, so the contact resistance is probably of no significance.
Capacitances are, however as low as 100pF. You should make every effort therefore to make the layout as compact and neat as possible to minimize strays. If the circuit works satisfactorily with 5-way rotary switches which will inevitably have some point-to-point wiring and consequent strays associated with them, then it will probably be OK with the relays, but it doesn't hurt to take care. Sticking to SMT parts will help to keep things compact.
One consideration might be that rotary switches come in make-before-break and break-before-make flavours and this may affect whether there is 'popping' as the settings are changed. You can obviously arrange how the switch timings occur using a uproc.
You do have to be careful with comparatively high-impedance circuits like these to keep the uproc from getting noise into them. Being high impedance, comparatively small currents can induce detectable voltages. Noise at radio frequencies can still intermodulate with audio. Close attention to supply bypassing and keeping signal/ground return loop areas small will minimize these problems, and separate digital and analog grounds joined at a single judiciously chosen point should be employed. Keep the 2 sections physically separated as far as possible.
Hi counter culture,
thank you for your suggestions! I will continue my quest with the relays and get back to report and/or ask some more! 🙂
cheers,
Bob
thank you for your suggestions! I will continue my quest with the relays and get back to report and/or ask some more! 🙂
cheers,
Bob
I just built a preamp with relay switching for the inputs. All I did was get some 12VDC relays and used the rotary switch to switch each relay directly. The other deck of the switch is used to switch LED indicators.
I would not use a microprocessor because it is just another source of noise and adds more complexity than it is worth, but your application sounds interesting.
I would not use a microprocessor because it is just another source of noise and adds more complexity than it is worth, but your application sounds interesting.
- Status
- Not open for further replies.