Double Bass grace felix doubling question: signal routing

[Jmilitsc]

after dreaming about these things for years i just landed a used original Felix, very excited. haven't gigged with it yet but with just a few minutes of testing, it’s as nice sounding and feature-rich as i hoped for the upright (realist lifeline pickup).

The situation and question:

i have a 2-input amp and used to run electric bass into ch.1 via sansamp vt-bass di pedal (plus ebs compressor) and then upright into ch.2 via hpf-pre3. i never play gigs sending signal to a PA, just small venues with my own amp.

i will experiment with several routing options soon on my own but really curious what other folks do here. the felix only has the one amp output and i really like the tone options with the sansamp, don’t want to lose that.

for the electric:

- do you like the sansamp in front of the felix and then amp output to 1 amp input and forgo the 2 ch. amp inputs?

- if not, do you take the DI out from the felix into the sansamp, then to ch.1 of amp? Mic or line level?

- do we ever like felix DI outputs into amp inputs? Mic level?

- splitter cable for amp-out of Felix?

- will any of this blow anything up?

any other good options that makes use of both felix inputs for doubling gigs, keeping the sansamp in place for the electric? i mean i don’t have to use the felix for the electric but i’d like to see what it brings to the table for it.

I’ve read a lot of Felix-related threads and haven’t found many clues on this, looking fwd to hearing what others are doing.

thx!

- jay

 

[Chris Fitzgerald]

Congrats! The Felix is a really nice unit. I wrote a review on it years ago for BGM that can be found here: https://chrisfitzgeraldmusic.com/do...l/?wpdmdl=894&refresh=693586d4367731765115604

The whole point of the Felix is as a two channel preamp, so trying to rout it into two amp channels would be redundant IMO. Also converting DI outs to amp outs will only degrade the sound and complicate things.

But if you like the sound of the Sansamp, there’s no reason not to use it in front of the Felix. You have to be careful with the input levels, but once you get that set, there should be no issue. Enjoy!

 

[Jmilitsc]

Congrats! The Felix is a really nice unit. I wrote a review on it years ago for BGM that can be found here: https://chrisfitzgeraldmusic.com/do...l/?wpdmdl=894&refresh=693586d4367731765115604

The whole point of the Felix is as a two channel preamp, so trying to rout it into two amp channels would be redundant IMO. Also converting DI outs to amp outs will only degrade the sound and complicate things.

But if you like the sound of the Sansamp, there’s no reason not to use it in front of the Felix. You have to be careful with the input levels, but once you get that set, there should be no issue. Enjoy!

What a great review article, thx for the link and opinions.

 

[Wasnex]

Try the Sansamp into the Felix and straight into the amp. Decide for yourself which option is best; for your own reasons. Keep in mind, a different person might make a different choice. If it works both ways, there is no right or wrong, only what you prefer.

 

[Ric Vice]

While it's dependant on what type of pickup you're using ( I use a Ehrlund EAP that needs the RedEye Twin Preamp) There's
really no need to preamp the Grace Felix. It's a all in one box. You're going to quickly get into a problem with "to much of a good thing" running
Sans Amp preamp into the Grace preamp into your Amp's preamp. So i'd try it with only the Grace and see what you think . IME the Grace is ideal for
a powered speaker. Unfortunately, that's on the expensive side.

 

[Brent Nussey]

While it's dependant on what type of pickup you're using ( I use a Ehrlund EAP that needs the RedEye Twin Preamp) There's
really no need to preamp the Grace Felix. It's a all in one box. You're going to quickly get into a problem with "to much of a good thing" running
Sans Amp preamp into the Grace preamp into your Amp's preamp. So i'd try it with only the Grace and see what you think . IME the Grace is ideal for
a powered speaker. Unfortunately, that's on the expensive side.

He's talking about putting an electric bass into the Felix, after passing it through the SansAmp. The Felix is a super clean preamp, and often that's not what you want for electric bass.

To the OP, I've used the Felix for doubling in a musical before. The house supplied a mic, so I didn't need that channel, and decided to use one channel for electric and one for acoustic. It worked, but the electric side was a super clean sound. Even sound aside, I decided if I do it again, to have two separate preamps, separately DI'd and with separate inputs on the amp. Just easier to control everything. When switching with the Felix, you'll have to step on the mute switch, change, step on the channel switch, and then unmute. With 2 separate setups, you just mute one, switch, then unmute the one you're about to use. And you never have to consider whether adjusting one instrument's volume will affect the other etc. The one thing that was good was that the Felix's switches are really silent, which can be important in some situations.

 

[Povl Carstensen]

Just to say, if you run the gain on the Grace (I have an Alix) high, with the clip indicator going red on high transients, you can get some saturation, crunch, compression.

 

[Jmilitsc]

He's talking about putting an electric bass into the Felix, after passing it through the SansAmp. The Felix is a super clean preamp, and often that's not what you want for electric bass.

To the OP, I've used the Felix for doubling in a musical before. The house supplied a mic, so I didn't need that channel, and decided to use one channel for electric and one for acoustic. It worked, but the electric side was a super clean sound. Even sound aside, I decided if I do it again, to have two separate preamps, separately DI'd and with separate inputs on the amp. Just easier to control everything. When switching with the Felix, you'll have to step on the mute switch, change, step on the channel switch, and then unmute. With 2 separate setups, you just mute one, switch, then unmute the one you're about to use. And you never have to consider whether adjusting one instrument's volume will affect the other etc. The one thing that was good was that the Felix's switches are really silent, which can be important in some situations.

Thx Brent, right it’s the not-super-clean tone of the sansamp that i like on the electric. Super clean is great for some things of course. I am looking fwd to experimenting with different options now that I have the Felix.

Thx for all the replies folks

 

[Jmilitsc]

love this preamp

still haven’t used it for real with the fender but i like it so much with the realist piezo on the upright at one gig with it i decided to try blending the atm350 clip on mic.

wow, finally MBOL at a gig last night. another gig and a demo recording coming up with it next week. love this thing.

i guess i’m derailing my own thread:

the amp output from the felix set to mix when blending is SO much lower than just sending channel 2 to the amp. i have to substantially raise the felix amp volume output and the input gain and master on my actual amp. not a big deal for the little restaurant duo i do, but that’s surprising to me, seek to understand.

for others who blend piezo and mic into a amp, is that expected?

 

[lowfreqgeek]

the amp output from the felix set to mix when blending is SO much lower than just sending channel 2 to the amp. i have to substantially raise the felix amp volume output and the input gain and master on my actual amp. not a big deal for the little restaurant duo i do, but that’s surprising to me, seek to understand.

for others who blend piezo and mic into a amp, is that expected?

Yes, that's expected. My edjumacated guess based on the block diagram and my knowledge of summing circuits is that when the mix/blend circuit is engaged, there is some padding (amplitude loss) from each input to account for the signal increase that occurs when the 2 signals of the same amplitude are summed. In order to sum the 2 channels and keep the same signal level, there would need to be an additional gain added when in mix mode. The block diagram doesn't really specify how the output buffering is set up, but switching another gain circuit via the mix/Ch2 switch would get more complicated.

That's just a guess, so perhaps there's more to the story. FWIW, my Felix behaves the same way.

 

[Wasnex]

Yes, that's expected. My edjumacated guess based on the block diagram and my knowledge of summing circuits is that when the mix/blend circuit is engaged, there is some padding (amplitude loss) from each input to account for the signal increase that occurs when the 2 signals of the same amplitude are summed. In order to sum the 2 channels and keep the same signal level, there would need to be an additional gain added when in mix mode. The block diagram doesn't really specify how the output buffering is set up, but switching another gain circuit via the mix/Ch2 switch would get more complicated.

That's just a guess, so perhaps there's more to the story. FWIW, my Felix behaves the same way.

?

If I understand what you wrote, you are saying the summing amp has a gain factor of <1. Not saying that's not possible, but I don't remember ever seeing it.

With all of the mixers I have run, the signal level at the stereo mix bus gets stronger as additional channels are summed.

A couple of my Genz Benz amps allow summing of two channels. There's always a level bump when the second channel is added.

AFAIK the optimal fix for a two channel design that allows A, B, or AB would be to add a trimmer that allows you to set the AB (sum) level however you want it.

An other explanation for why the level drops is the two pickups are not in absolute phase. AFAIK the Felix has polarity switches on the channels (see 4 and 5 below). There is also a variable phase adjust for channel 2. (see 8 below).

 

[lowfreqgeek]

?

If I understand what you wrote, you are saying the summing amp has a gain factor of <1. Not saying that's not possible, but I don't remember ever seeing it.

I'm not sure that I described it quite right the first time, but I did put more thought, some circuit analysis, and a brief simulation into it.

On the Felix, you can turn on and off the mix bus via hardwired switch. When the mix bus is turned off or bypassed, the amp output is wired directly to Ch2 and thus has a different signal path than when the mix bus is turned on. When the mix bus is turned on, both CH1 and CH2 are blended and summed, but the block diagram doesn't describe the summing circuit. What does happen, though, is that the CH2 signal going into the mix bus is attenuated, regardless of whether there is signal present on CH1.

My contention is that CH2 is attenuated going into the mix bus so that when signal is added from CH1, the overall amplitude doesn't increase beyond what CH2 would be when the mix bus was bypassed. If both CH1 and CH2 had the same input amplitude, the summed output would be +6dB hotter than either individual amplitude. Perhaps my assumptions are incorrect, but it would make sense to ensure that whenever the 2 signals are summed, there is no loss of headroom. Even with high voltage rails, there's a finite amount of headroom to work with, and 6dB is significant (2x).

Hopefully that makes better sense.

 

[Jmilitsc]

Thanks guys, yeah the diagram didn’t seem to account for any trimming, but the explanation makes logical sense if that’s how they designed the summing. I can’t find any documentation on this point from Grace. I might shoot a support email out to hear their explanation out of curiousity.

I’ve definitely played with different phase choices and haven’t noticed significant volume changes, very minimal, but of course definite tonal differences.

Thx again

 

[Jmilitsc]

What does happen, though, is that the CH2 signal going into the mix bus is attenuated when the mix bus is turned on, regardless of whether there is signal present on CH1.

Ahah, yes I noticed this too - mix mode with only signal from Ch2 is way lower amp output volume. This is helpful thx.

(I wrote my earlier response before your last post there lowfreqgeek)

 

[lowfreqgeek]

Ahah, yes I noticed this too - mix mode with only signal from Ch2 is way lower amp output volume. This is helpful thx.

(I wrote my earlier response before your last post there lowfreqgeek)

I'd be curious what Grace says if you do ask them about it. I'm making assumptions, but they seem plausible.

I love my Felix, too. I've tried some other ways to do things and the Felix just does everything so dang well. I often use an AMT-S25B along with my Realist and I really appreciate the flexibility, not to mention the purity and depth of the sound.

 

[Wasnex]

I'm not sure that I described it quite right the first time, but I did put more thought, some circuit analysis, and a brief simulation into it.

On the Felix, you can turn on and off the mix bus via hardwired switch. When the mix bus is turned off or bypassed, the amp output is wired directly to Ch2 and thus has a different signal path than when the mix bus is turned on. When the mix bus is turned on, both CH1 and CH2 are blended and summed, but the block diagram doesn't describe the summing circuit. What does happen, though, is that the CH2 signal going into the mix bus is attenuated, regardless of whether there is signal present on CH1.

My contention is that CH2 is attenuated going into the mix bus so that when signal is added from CH1, the overall amplitude doesn't increase beyond what CH2 would be when the mix bus was bypassed. If both CH1 and CH2 had the same input amplitude, the summed output would be +6dB hotter than either individual amplitude. Perhaps my assumptions are incorrect, but it would make sense to ensure that whenever the 2 signals are summed, there is no loss of headroom. Even with high voltage rails, there's a finite amount of headroom to work with, and 6dB is significant (2x).

Hopefully that makes better sense.

I did not dig too deep into analysis, but that's pretty much what I thought you meant. Also your analysis is consistent with what I think I see...other than I assume the mixing circuit is at unity gain.

AFAIK, mixing circuits usually combine signals at unity gain. With exact duplicates of the signal you get a bit of a bump in the level. If I understand correctly, if you mix two signals that are exactly the same, you get +6db of voltage gain. Essentially the voltage doubles and 20Log(2) = 6.02dB

The problem with mixing at less than unity gain is the question of how much to cut the level. I assume the idea is for both signals to sum without producing the level bump, but results depend on the relative signal strength of signals you are summing (This is a bit of an oversimplification as it depends on some other variables as well). So if you can assume the signals will always be exactly the same, you can just knock the level back ~6dB. However if the signal levels are dissimilar, then they sum to < than 6.02dB. So the 6dB of attenuation will over correct and drop the output a bit...which I believe is what you are saying is happening.


FYI, these points are a bit over my knowledge level, so I could definitely be wrong.

 

[lowfreqgeek]

I did not dig too deep into analysis, but that's pretty much what I thought you meant. Also your analysis is consistent with what I think I see...other than I assume the mixing circuit is at unity gain.

AFAIK, mixing circuits usually combine signals at unity gain. With exact duplicates of the signal you get a bit of a bump in the level. If I understand correctly, if you mix two signals that are exactly the same, you get +6db of voltage gain. Essentially the voltage doubles and 20Log(2) = 6.02dB

That is correct; 2 identical signals perfectly summed together will result in a 6.02dB, or 2x, increase in the signal voltage amplitude. All is not usually equal and things like phase shifts and other differences will reduce that increase, but that's the starting point.

In a traditional analog mix bus where everything is connected all the time, the signal levels are designed such that all the signals combined will maintain some headroom margin. It may or may not be "unity gain", but the intent is to keep all the amplitudes below a certain level where their combined amplitudes would overdrive the bus. Obviously, the more channels you're summing, the more you need to take care to not overdrive that bus.

The problem with mixing at less than unity gain is the question of how much to cut the level. I assume the idea is for both signals to sum without producing the level bump, but results depend on the relative signal strength of signals you are summing (This is a bit of an oversimplification as it depends on some other variables as well). So if you can assume the signals will always be exactly the same, you can just knock the level back ~6dB. However if the signal levels are dissimilar, then they sum to < than 6.02dB. So the 6dB of attenuation will over correct and drop the output a bit...which I believe is what you are saying is happening.

In a nutshell, yes - I'm assuming that Grace attenuates CH2 to make room for the possibility of a 6dB increase when mixing CH1, while maximizing the signal from CH2 when not mixing. The choice of how much to attenuate for any given bus design is dependent on the overall design philosophy and goals of the designer. A 2CH mixer will have different requirements than a 12, 24, or 48Ch mixer, so how the summing is achieved may change significantly.

In the Felix, it's not just a straight summing amp, but a variable blend of the 2 channels. Also, there is a separate level control for the amp output, so the relative balance of each signal and the overall output level are adjustable.

FYI, these points are a bit over my knowledge level, so I could definitely be wrong.

We could both be wrong! As I've stated, I'm making some assumptions myself. Grace may have an entirely different reason that CH2 is attenuated when the Mix button is pressed.

 

[Jmilitsc]

Ok, While they didn’t clarify how the attenuation is applied in the unit, it looks like the theories we discussed here are pretty well confirmed by the good folks at Grace Designs support! I also asked to understand how the mix knob impacts levels as well. Here’s the key bits from our discussion:

On a unit here, with only CH2 driven, mix knob 100% CH2, it is 3.7dB lower on amp out mix source compared to CH2 amp source. With mix knob at 50%, it is 6dB lower on mix source.

looks like we give the mixer 6dB of headroom. Makes sense since mixing two correlated signals gives you a 6dB gain.

So to my ears this backs up what I hear through the amp under these variables and I think what you all have weighed in on. I’ll use the preamp BOOST perhaps if I ever need more headroom on my amp, but for the gigs I do this isn’t an issue.

I’ll end with…. Love the preamp.

Thx for the engaging and helpful chat here folks. I guess that’s what I’m ending with, then.

 

[lowfreqgeek]

Ok, While they didn’t clarify how the attenuation is applied in the unit, it looks like the theories we discussed here are pretty well confirmed by the good folks at Grace Designs support! I also asked to understand how the mix knob impacts levels as well. Here’s the key bits from our discussion:

On a unit here, with only CH2 driven, mix knob 100% CH2, it is 3.7dB lower on amp out mix source compared to CH2 amp source. With mix knob at 50%, it is 6dB lower on mix source.

looks like we give the mixer 6dB of headroom. Makes sense since mixing two correlated signals gives you a 6dB gain.

So to my ears this backs up what I hear through the amp under these variables and I think what you all have weighed in on. I’ll use the preamp BOOST perhaps if I ever need more headroom on my amp, but for the gigs I do this isn’t an issue.

I’ll end with…. Love the preamp.

Thx for the engaging and helpful chat here folks. I guess that’s what I’m ending with, then.

Thanks for checking in with the Grace Designs support folks and getting back to us! Good to know that my assumptions weren't completely off the mark, too.

I'll say that the Felix (and their other preamps) are most definitely designed from a high-end studio gear perspective, including little details like the mixer headroom. Not to mention the feel of the knobs and switches, the robustness, the I/O flexibility, and most importantly - the sound.