Hi,
I have a modded Eastern Electric DAC (ES9018) which had its volume control and input selection switch bypassed. These bypasses appear to have increased the output gain a little (too much?).
I tried placing some borrowed Rothwell inline attenuators on my preamp’s input to hopefully make the sound a little less “hot” and the volume control a bit more gradual.
This works OK except I find it dulls the sound. I would like to know how I could slightly attenuate the gain from my DAC’s output. I believe this involves inserting one or two resistors from the board to the DAC’s RCA output jacks.
Could anyone provide some advice as to how this is done and what value resistor(s) would be a good starting point to achieve a -10db attenuation.
Thanks
I have a modded Eastern Electric DAC (ES9018) which had its volume control and input selection switch bypassed. These bypasses appear to have increased the output gain a little (too much?).
I tried placing some borrowed Rothwell inline attenuators on my preamp’s input to hopefully make the sound a little less “hot” and the volume control a bit more gradual.
This works OK except I find it dulls the sound. I would like to know how I could slightly attenuate the gain from my DAC’s output. I believe this involves inserting one or two resistors from the board to the DAC’s RCA output jacks.
Could anyone provide some advice as to how this is done and what value resistor(s) would be a good starting point to achieve a -10db attenuation.
Thanks
Hi you need to use an L pad, which is 4 resistors, 2 for each channel. Using a circuit breadboard arrange a series resistor of approx 3.3k for each channel and a shunt ( to ground ) resistor of approx 12k for each channel. Alternatively an LDR based attenuator will give excellent audio and allow you to vary volume to suit.
Cheers / Chris
Chris,
With my very limited DIY experience I understand the first option but not sure why a breadboard is needed.
You lost me with the LDR based option.
Thanks again,
Mike
Hi Mike
A breadboard might be needed if you are cutting the RCA cable, there is then ability to use 1/4 - 1/2 watt resistors. Of course on the RCA socket internally inside the casework is a bit neater ( but was unsure if you could get to that, hence suggestion of breadboard ).
As you may have observed already even resistors impart a sonic signature. Further up the path not to take, is placing resistors or variable means of attenuation like potentiometers or switched attenuators that involve switching of contacts, hence if you needed to vary the attenuation LDR's are a nice option.
Fixed resistors are simple and should work well for the purpose required.
Cheers / Chris
A breadboard might be needed if you are cutting the RCA cable, there is then ability to use 1/4 - 1/2 watt resistors. Of course on the RCA socket internally inside the casework is a bit neater ( but was unsure if you could get to that, hence suggestion of breadboard ).
As you may have observed already even resistors impart a sonic signature. Further up the path not to take, is placing resistors or variable means of attenuation like potentiometers or switched attenuators that involve switching of contacts, hence if you needed to vary the attenuation LDR's are a nice option.
Fixed resistors are simple and should work well for the purpose required.
Cheers / Chris
Attenuators tend to dull the sound because if put in-line they use a noisy 0V reference - the screen of the connecting cable. To get an attenuator which doesn't squash the dynamics you'll need to use a clean 0V reference and this means opening up the DAC to do the mod. For a person with limited DIY skills I'd not recommend this route.
Alternatively, use a transformer with a 3:1 turns ratio. This provides isolation and accomplishes the attenuation without introducing noise.
Alternatively, use a transformer with a 3:1 turns ratio. This provides isolation and accomplishes the attenuation without introducing noise.
Thanks guys.
I should be able to squeeze an L pad between the board and the RCAs from the inside. The best resistors I know of for this application would probably be naked Vishay or the Texas Instrument equivalents. They are very small and are the closest thing to a piece of wire sound wise. The downside is their price.
I’ve experimented with output transformers in the past to replace opamps on a CS4398 board.
At the time I had a very helpful forum member walk me through the steps. I wouldn’t feel comfortable doing it on my own and unfortunately the fellow (Bill Fuss for those who knew him) passed away a while back.
I should be able to squeeze an L pad between the board and the RCAs from the inside. The best resistors I know of for this application would probably be naked Vishay or the Texas Instrument equivalents. They are very small and are the closest thing to a piece of wire sound wise. The downside is their price.
I’ve experimented with output transformers in the past to replace opamps on a CS4398 board.
At the time I had a very helpful forum member walk me through the steps. I wouldn’t feel comfortable doing it on my own and unfortunately the fellow (Bill Fuss for those who knew him) passed away a while back.
Last edited:
McCrackers said:
Before you get too carried away with choosing resistor brand etc. it would pay to get the values correct. 3.3k series and 12k shunt would give an attenuation of -2.1dB (assuming a low impedance source and high impedance load).Chris Daly said:
Swapping them over gives -13.3dB. Try 6.8k series and 3.3k shunt - this gives -9.7dB.
Note that the best place for an attenuator is at the receiving end of the cable. At the sending end it has to drive the cable capacitance so some HF loss can occur.
LDR attenuators are fine if you want adjustment and can cope with a little distortion (LDRs are slightly non-linear - ordinary resistors are much better).
Hi McCrackers
It would pay to measure with DAC switched off with your multimeter what resistance the DAC already has from its RCA output to Ground, as adding further shunt resistance as suggested with an L pad will then be in parallel. From that figure you can then work out correct shunt resistance you are adding, to attain the 10db required.
Cheers / Chris
It would pay to measure with DAC switched off with your multimeter what resistance the DAC already has from its RCA output to Ground, as adding further shunt resistance as suggested with an L pad will then be in parallel. From that figure you can then work out correct shunt resistance you are adding, to attain the 10db required.
Cheers / Chris
Measuring 'cold' resistance, as Chris Daly suggests, will tell you precisely nothing. The actual 'hot' impedance with the device switched on could be much higher or much lower; it is very unlikely to be the same, circuits don't work like that.
Whatever impedance there is will be in series with the L pad, not parallel as he says.
Whatever impedance there is will be in series with the L pad, not parallel as he says.
Emmm... do you actually suggest
- increasing the output impedance of the DAC. Thus multiplying all the magnetic interference by R_source, which efficiently converts induced magnetic fields into voltage. As a bonus - HF response may drop a little as following preamp's input stages may contain proper RF-suppressing caps to ground. Aaand, if it goes straight to the potentiometer - your LOG curve goes weee, or somewhere else...
- loading DAC's output with 10K L-Pad in addition to the typical 10K-100K load found in average input circuits? (worst common load is 5K). At 5k load, the output stage may struggle.
inline resistors ain't good for anything, unless you put a buffer behind 'em.
Do the things at the right places. Their place is inside the DAC's enclosure, prior to output line drivers, right after LPFs. You might need to recalculate the LPFs with attenuation if there is not enough opamp stages to play with. Yet, most likely there is some sort of gain built-in, and all you need to do is to fix it = turn it into a buffer.
If you broke it, fix it.
- increasing the output impedance of the DAC. Thus multiplying all the magnetic interference by R_source, which efficiently converts induced magnetic fields into voltage. As a bonus - HF response may drop a little as following preamp's input stages may contain proper RF-suppressing caps to ground. Aaand, if it goes straight to the potentiometer - your LOG curve goes weee, or somewhere else...
- loading DAC's output with 10K L-Pad in addition to the typical 10K-100K load found in average input circuits? (worst common load is 5K). At 5k load, the output stage may struggle.
inline resistors ain't good for anything, unless you put a buffer behind 'em.
Do the things at the right places. Their place is inside the DAC's enclosure, prior to output line drivers, right after LPFs. You might need to recalculate the LPFs with attenuation if there is not enough opamp stages to play with. Yet, most likely there is some sort of gain built-in, and all you need to do is to fix it = turn it into a buffer.
If you broke it, fix it.
Hi
All valid points, ideally any attenuation should be done where there is sufficient current, call it buffering and without adding series or shunt resistances or combinations of both twice or three times. The post I think called for a easy fix, but of course there never is as I think Ben Duncan has said in the past As "one door opens and 20 more doors are flung open below it as well.
A buffer stage then capable of attenuation
A digital pot
A LDR attenuator ( with very pleasant distortion )
Transformer
Resistors suitably placed so as not to introduce distortion or compete with other resistors placed
Perhaps if a schematic was available we could rally to the cause of suggesting closer to the DAC itself where attenuation and how to do it, would be best placed.
Cheers / Chris
All valid points, ideally any attenuation should be done where there is sufficient current, call it buffering and without adding series or shunt resistances or combinations of both twice or three times. The post I think called for a easy fix, but of course there never is as I think Ben Duncan has said in the past As "one door opens and 20 more doors are flung open below it as well.
A buffer stage then capable of attenuation
A digital pot
A LDR attenuator ( with very pleasant distortion )
Transformer
Resistors suitably placed so as not to introduce distortion or compete with other resistors placed
Perhaps if a schematic was available we could rally to the cause of suggesting closer to the DAC itself where attenuation and how to do it, would be best placed.
Cheers / Chris
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.