or Connect
AVS › AVS Forum › Audio › DIY Speakers and Subs › Guide to install pre-outs on an AVR
New Posts  All Forums:Forum Nav:

Guide to install pre-outs on an AVR - Page 2

post #31 of 68
Quote:
Originally Posted by MBentz View Post


900ohm output impedance is pretty high for a line level signal - I would try to scale the resistor values back so that the shunt resistor is more like 100ohms. Also, it's not uncommon for speaker amplifier stages to have a little DC offset - I personally would add a series cap to prevent any potential offset from transferring down stream.

Not as low as one might want but typical power amp input impedances are 10K and above and this will not be an issue.

post #32 of 68
Quote:
Originally Posted by WagBoss View Post

so the behringer DI4000 will take a speaker-level out and convert it to balanced line-level outputs? Or does it only take line-level outs? It looks like it only takes line inputs, but it says it takes up to 3000 watts per channel into it?
is there any advantage to use this over just a standard voltage divider? other then the fact it has balanced outputs.

It takes both inputs, i used the sub out and did a +20db boost for input to a pro amp.

I primarily used it for attenuation of speaker level, so i could use it for pro amp. Since my AVR (1021k) , doesnt have pre outs. It has options for up to -40db attenuation if needed for higher power applications, but i just used -20.
post #33 of 68
Quote:
Originally Posted by WagBoss View Post

so... there is no disadvantage to just use a voltage divider? Since I already bought RCA chassis mounts, mideswell build my own line-level converter instead :P So I'd use just a 1k and a 100 ohm resistor as the divider... or a 4k and 400 ohm? Should I just grab some 1 W resistors to be safe or would 1/2 be fine?
or should I do what http://www.epanorama.net/circuits/speaker_to_line.html says and do a 33k ohm to 1k ohm as its possible for my receiver to output 105 wpc?
Or the best bet seems to be like a 10k and a 470 ohm..?

Well you're gonna have the distortion/noise from the output stage, whereas a true line level output would have one/two less gain stages in the signal path. As far as the divider, you set the shunt to be your desired output impedance and then the series sets the amount of attenuation you want. As far as power handling, that's pretty easy to calculate. 105 watts into an 8 ohm load is about 30Vrms. 30Vrms into a 1k load is 0.84W. I would recommend 2W resistors for that kind of application. Or you can scale up the values if you want to use smaller resistors - generally you want to be better than 10:1 on the input/output impedance between devices. You can live with 1k for a 10k input impedance device, but I personally try to target at least 20:1. Some of your pro gear will be 5k input impedance for line level inputs. I've seen as low as 2k on some products. Obviously you can design for your own gear if you know what it's gonna be. Your EMI immunity on longer unbalanced cable runs will be better with a lower output impedance too.
post #34 of 68
post #35 of 68
Thread Starter 
OK so the two options I am considering is tapping into the line right before the amplifier to get LCR pre-outs, or putting a voltage divider on the LCR speaker-outs. From your posts I conclude that its the same quality doing either, so I will just make a voltage divider. I am going to use a 3900 ohm resistor and a 430 ohm resistor (identical to the picture linked in post #7), both rated at 1W. Will this work fine? I am planning on getting an Emotiva XPA-3 to connect with these pre-outs. Thanks.
post #36 of 68
To make proper pre-outs you really need the schematics.
I am going to do this to a Denon, I think it will work out fine on this one.

Most AVRs will have wires from the pre circuit to the power amps, so this is where you can tap the signal.
Then you should consider:

1. Level must be appropriate - the output stages in an integrated AVR does not necessarily have the usual 27dB-or-so gain.
2. Muting may not work.
3. There must be room for the connectors on the rear panel.

You will of course void any warranty on the AVR, and you should know a little about electronics.
This is not a job I would consider to do for a client if I was a professional.
post #37 of 68
Thread Starter 
Quote:
Originally Posted by Okv View Post

To make proper pre-outs you really need the schematics.
I am going to do this to a Denon, I think it will work out fine on this one.
Most AVRs will have wires from the pre circuit to the power amps, so this is where you can tap the signal.
Then you should consider:
1. Level must be appropriate - the output stages in an integrated AVR does not necessarily have the usual 27dB-or-so gain.
2. Muting may not work.
3. There must be room for the connectors on the rear panel.
You will of course void any warranty on the AVR, and you should know a little about electronics.
This is not a job I would consider to do for a client if I was a professional.

can you reply to the post above yours please?

I am going to make a voltage divider as I just bought this 8 months ago and don't want to void my warranty and there seems to be no disadvantage to a voltage divider.
post #38 of 68
Thread Starter 
for a voltage divider:

I used R1 = 4.7k, R2 = .36k and Vsource = 28V.

So I calculated Itotal = 0.0055 A
V1= 26 V
V2 = 1.98 V
P1 = 0.143 W
P2 = 0.0109 W

So my 1W resistors will work fine with this, assuming my calculations are correct. and the 2V output is exactly what I want going into my amplifier, correct?

I am not that familiar with how an AVR amplifier works, is the output always going to be 28 V, and the current output will vary based on volume, so at any volume level the pre-out will be 2V and the source will be 28V?

thanks for the help guys
post #39 of 68
Quote:
Originally Posted by WagBoss View Post

for a voltage divider:
I used R1 = 4.7k, R2 = .36k and Vsource = 28V.
So I calculated Itotal = 0.0055 A
V1= 26 V
V2 = 1.98 V
P1 = 0.143 W
P2 = 0.0109 W
So my 1W resistors will work fine with this, assuming my calculations are correct.
That looks fine. 1/4W resistors would be fine as you'll not be giving full voltage constantly, unless you use a test tone.
Quote:
Originally Posted by WagBoss View Post

and the 2V output is exactly what I want going into my amplifier, correct?
I am not that familiar with how an AVR amplifier works, is the output always going to be 28 V, and the current output will vary based on volume, so at any volume level the pre-out will be 2V and the source will be 28V?
thanks for the help guys
2V may be correct for full power into your external amp or it may not. You need to look at the voltage sensitivity of the amp - should be in the specs, or it can be calculated from the output power. Find the voltage for that power into the specified load and divide by the gain ratio, which if given in dB, will need to be converted.

The 28V you used as full output voltage will not be correct unless the amps in your AVR are rated at 100W/8R. Anything else will give a different max output voltage.

Amplifiers do not always put out the same voltage, that will vary on the signal level off the disc, and the attenuation from the volume control. The gain (amplification) of a power amp stage is constant.

Lets say you're using a sine wave to test, and you have set the level so that your power amp is putting out 28Vrms. If the amp had a gain of 28x, you'd have 1Vrms at the input of the amp. Before that you have a volume control, which generally only reduce the level of the source. Let's say that's at -6dB (0.5x) then the level of the test tone from the source is 2Vrms, which conveniently is what a DAC typically puts out. Now source material on a disc, whether music or movies varies in level all the time. If you'd changed nothing from the test tone, at the loudest parts of the source, you'd be right near clipping, but the softest parts will be around -80dB or 1/10000th, 2.8mV at the amp output or 100uV at the amp input.

If the amp were connected to a speaker, it would give the same output voltages for the examples above, but because the load is 8R, the current will increase proportionately, so power will be delivered to the load (P=V x I).
post #40 of 68
Thread Starter 
no it is not rated at 100W while all channels running, just when 2 are. But i want to design it for worst case scenario, as I will be playing 2ch music too. Would it be a problem if when I'm using all channels, it's only putting out maybe 14V rms , so the output would only be about 1 V?

the amp I want to get is an emotiva xpa-3, which has a 32 dB gain, and can put out 200 W / ch into 8 ohms.
post #41 of 68
Figured I's throw my two cents in. I just broke open my Denon 1612(lame model, so sue me) and found everything clearly labeled. The HDMI board handles all digital. There is a board that links it to the DAC board, both ends of all connecting traces are labeled. The receiver has sub and rear/height outs, so those are taken care of. There is a group of 10 or 11 wires all going to the amplifier board labeled either labeled GND or with what channel they are. I would have to remove the HDMI board to see if there is any amplification between the DAC and the existing pre-outs.

Overall, this model seems fairly straightforward to mod. The warranty is already void because I installed a power cable with a ground pin to help kill hum. I am due a new receiver soon anyways, so doing some modding to this one couldn't hurt. If I make any progress I will report back. Might take some time though, the on board amplifier is perfectly adequate for the speakers I am running.
post #42 of 68
Quote:
Originally Posted by WagBoss View Post

no it is not rated at 100W while all channels running, just when 2 are. But i want to design it for worst case scenario, as I will be playing 2ch music too.
2ch is the easiest case: none of the other channels will be supplying power to their speakers; surrounds, nothing, and mians only a few mA in the voltage divider. Design for rated power.
Quote:
Originally Posted by WagBoss View Post

Would it be a problem if when I'm using all channels, it's only putting out maybe 14V rms , so the output would only be about 1 V?
As surrounds put out less power than mains most of the time, they will contribute less to power supply sag than the mains and the mains in your case will be contributing nothing negative there as they will only be supplying a few mA to the voltage divider. I'd design for rated power.
Quote:
Originally Posted by WagBoss View Post

the amp I want to get is an emotiva xpa-3, which has a 32 dB gain, and can put out 200 W / ch into 8 ohms.
200W/8R is 40Vrms and 32dB is 40x gain, so 1Vrms in for rated power.
post #43 of 68
i did just that about a year back with my previous pioneer. it's a tall order if you're not into heavy modding, never modded any audio equipment or don't have access to good tools and equipment.
definitely not impossible to do.





most receiver have a clear marking on their boards. and it's fairly easy to identify the digital section, analog preamp section and power amp section. you just need to identify the output of the preamp section going to the poweramp and piggyback the signal. if the signal is high enough you don't need an extra preamp section like i did and just wire it to some rca jacks.
post #44 of 68
Thread Starter 
Quote:
Originally Posted by A9X-308 View Post

2ch is the easiest case: none of the other channels will be supplying power to their speakers; surrounds, nothing, and mians only a few mA in the voltage divider. Design for rated power. As surrounds put out less power than mains most of the time, they will contribute less to power supply sag than the mains and the mains in your case will be contributing nothing negative there as they will only be supplying a few mA to the voltage divider. I'd design for rated power.
200W/8R is 40Vrms and 32dB is 40x gain, so 1Vrms in for rated power.

So if I go with this resistor combo, in stereo, I should keep the gain below 29 dB, since it will be a 2VRMS input into the xpa-3? and with 5 channels going, it could possibly be putting out half power, so 14V RMS, which is 1VRMS on the voltage divider, so I could turn it to full if i wanted?
post #45 of 68
Thread Starter 
Quote:
Originally Posted by paskal9 View Post

i did just that about a year back with my previous pioneer. it's a tall order if you're not into heavy modding, never modded any audio equipment or don't have access to good tools and equipment.
definitely not impossible to do.



most receiver have a clear marking on their boards. and it's fairly easy to identify the digital section, analog preamp section and power amp section. you just need to identify the output of the preamp section going to the poweramp and piggyback the signal. if the signal is high enough you don't need an extra preamp section like i did and just wire it to some rca jacks.

cool, thanks for the pics. What was the voltage on your analog pre-amp signal? It was low so you had to add a pre-amp section?
post #46 of 68
Quote:
Originally Posted by WagBoss View Post

cool, thanks for the pics. What was the voltage on your analog pre-amp signal? It was low so you had to add a pre-amp section?
yes, the output is quite low. the output from the DAC is 2V peak to peak before being buffered, filtered then went into an unmarked chip. output from the unmarked chip goes into another buffer then ends at around 200mV (or issit 400mV can't remember). this signal then goes to the poweramp module. there's some transistors at the poweramp module that looks like the first voltage amplifier but it's easier to piggyback the signal before it went into the poweramp board.

if you've done modification or built some audio equipment be it an amplifier or preamp or DAC then it's pretty much the same thing being utilized in the receiver. there's more channel, but still consists of the same thing like opamp, high pass filters, buffers, etc. pretty easy to recognize.
post #47 of 68
Thread Starter 
Quote:
Originally Posted by paskal9 View Post

yes, the output is quite low. the output from the DAC is 2V peak to peak before being buffered, filtered then went into an unmarked chip. output from the unmarked chip goes into another buffer then ends at around 200mV (or issit 400mV can't remember). this signal then goes to the poweramp module. there's some transistors at the poweramp module that looks like the first voltage amplifier but it's easier to piggyback the signal before it went into the poweramp board.
if you've done modification or built some audio equipment be it an amplifier or preamp or DAC then it's pretty much the same thing being utilized in the receiver. there's more channel, but still consists of the same thing like opamp, high pass filters, buffers, etc. pretty easy to recognize.

So it's probably just easier to put a voltage divider on the output of the amplifier terminals? I don't need to avoid my amp terminals, I just want pre-outs.
post #48 of 68
None of these mods are "easier" than just buying a receiver with preouts...

Doing a pre-out mod is going to sound better than tapping off the speaker outputs, but it requires a different skill set....especially if you want it to actually be better than tapping off the speaker outputs. Getting audio to come out the back is a piece of cake - making it sound right is a bit more involved.

To be honest, I'm surprised that an electrical engineering student isn't designing his own solution - isn't that why you're going to school? wink.gif
(just trying to give you a friendly nudge to dive in)
post #49 of 68
Thread Starter 
Quote:
Originally Posted by MBentz View Post

None of these mods are "easier" than just buying a receiver with preouts...
Doing a pre-out mod is going to sound better than tapping off the speaker outputs, but it requires a different skill set....especially if you want it to actually be better than tapping off the speaker outputs. Getting audio to come out the back is a piece of cake - making it sound right is a bit more involved.
To be honest, I'm surprised that an electrical engineering student isn't designing his own solution - isn't that why you're going to school? wink.gif
(just trying to give you a friendly nudge to dive in)

Well, I designed the voltage divider im gonna put on ? :P

Wouldn't buying a new receiver be "not designing my own solution" ? lol

The stuff we do in school is trivial compared to an AVR, I know how easy it is to mess up electronics, I don't want to wreck my AVR
post #50 of 68
(I think I'm a little late with this comment, but oh well!) Tapping the volume control is not correct. You want the volume control to affect the external amplifiers too! Best to tap the signal off of the output of the (most likely) op amps of the preamp inside the receiver. Though those will probably be harder to find (especially to the untrained eye) than the volume control will be...

P.S. the level of your preouts should be in the 1-2VRMS range. Technically 1V is correct, though you don't need to follow that exactly. Don't go crazy and make it 10V or something. You'll clip the input buffer/preamp of some amps (particularly consumer ones that actually do expect a relatively low voltage signal).
Edited by DonoMan - 7/26/12 at 1:04pm
post #51 of 68
Thread Starter 
Quote:
Originally Posted by DonoMan View Post

(I think I'm a little late with this comment, but oh well!) Tapping the volume control is not correct. You want the volume control to affect the external amplifiers too! Best to tap the signal off of the output of the (most likely) op amps of the preamp inside the receiver. Though those will probably be harder to find (especially to the untrained eye) than the volume control will be...
P.S. the level of your preouts should be in the 1-2VRMS range. Technically 1V is correct, though you don't need to follow that exactly. Don't go crazy and make it 10V or something. You'll clip the input buffer/preamp of some amps (particularly consumer ones that actually do expect a relatively low voltage signal).

so 2V rms at max power rating should be perfect? If i'm listening quietly, it would be something like .4V say, would that be a problem?
post #52 of 68
Quote:
Originally Posted by WagBoss View Post

so 2V rms at max power rating should be perfect? If i'm listening quietly, it would be something like .4V say, would that be a problem?

That is normal. You have the choice to either follow typical consumer line-level standards, to break them in a way that is still compatible with your specific equipment, or by making the gain of your voltage divider adjustable (could use a simple DPDT switch to give you two options for levels and not have to buy a ganged potentiometer)

If you do choose to follow standards for output level, look up the numbers for consumer and pro audio and pick one. I don't actually remember the exact numbers offhand, just that consumer is quite low, and pro is somewhat less low.
post #53 of 68
Thread Starter 
Quote:
Originally Posted by DonoMan View Post

That is normal. You have the choice to either follow typical consumer line-level standards, to break them in a way that is still compatible with your specific equipment, or by making the gain of your voltage divider adjustable (could use a simple DPDT switch to give you two options for levels and not have to buy a ganged potentiometer)
If you do choose to follow standards for output level, look up the numbers for consumer and pro audio and pick one. I don't actually remember the exact numbers offhand, just that consumer is quite low, and pro is somewhat less low.

I was hoping I could just put the resistors on and call it a day. This is fine to do right, you're just saying making it a constant voltage output or switches to control the divider would be better, but not necessary right?
post #54 of 68
Quote:
Originally Posted by WagBoss View Post

I was hoping I could just put the resistors on and call it a day. This is fine to do right, you're just saying making it a constant voltage output or switches to control the divider would be better, but not necessary right?

Well, not a constant voltage as that would mess up the sound. You're looking for your peak output level to be approximately equal to the line-level spec you may choose to adhere to. It's up to you how closely you want to adhere. And besides, plenty of gear will accept a signal higher than the spec, or otherwise it would just limit you to a lower maximum gain on the receiver volume control if you do use a voltage that is too high for your amp.

You could use a DPDT switch to switch between a resistor that gives you consumer level and one that gives you pro level.
post #55 of 68
Quote:
Originally Posted by WagBoss View Post

So it's probably just easier to put a voltage divider on the output of the amplifier terminals? I don't need to avoid my amp terminals, I just want pre-outs.
before messing up with the internals i did tried the voltage divider method at the speaker output. it works. needs some fine tuning but it works. that is the easiest method.
Quote:
Originally Posted by WagBoss View Post

so 2V rms at max power rating should be perfect? If i'm listening quietly, it would be something like .4V say, would that be a problem?
that would not be a problem. start at 2Vrms first, but i'm sure you gonna need to tune it later on to have a proper setting that works for your setup. don't think too much and try it.

2Vrms is just a starting guide. i've measured the sub preout on my current anthem mrx300 and on reference 0dB the signal is 2.65Vrms
Quote:
Originally Posted by WagBoss View Post

I was hoping I could just put the resistors on and call it a day. This is fine to do right, you're just saying making it a constant voltage output or switches to control the divider would be better, but not necessary right?
totally unnecessary. you just need a couple of resistors for each channel. test the values first on a particular channel then use the value for all channel.
post #56 of 68
Quote:
Originally Posted by paskal9 View Post

totally unnecessary. you just need a couple of resistors for each channel. test the values first on a particular channel then use the value for all channel.

Better to use an op amp so that the circuit is load-independent, if you ask me. Though it doesn't HAVE to be done that way. Could also use a line out converter meant for a car, though most of them require a 12v input (some are passive and do not, but they are less good).
post #57 of 68
Quote:
Originally Posted by DonoMan View Post

Better to use an op amp so that the circuit is load-independent, if you ask me.
There is no need whatsoever for any active buffer here.
post #58 of 68
Quote:
Originally Posted by A9X-308 View Post

There is no need whatsoever for any active buffer here.

I didn't say there was a "need," but I do believe that it would be a better option though more of a pain to implement. With just a voltage divider, the output voltage will depend on the input resistance of the amplifier (which would be in parallel with one of the resistors, and would constrain the values of resistor you could use as well). It's not the end of the world but it's not a good thing either. It can be done passively, just that it's not a great solution if you want to be able to swap in any amplifier you want at a later date while having predictable results from the voltage divider.

I think you probably should have given some reasons at least, else anyone who listens to you wouldn't even understand why they're doing so... But anyway, thems the pros and cons, pretty much.


Oh, and I'll respond to something else I saw
Quote:
1. Level must be appropriate - the output stages in an integrated AVR does not necessarily have the usual 27dB-or-so gain.
Indeed... it's in the preamp section, not the amp section. Adding gain in the output stage will reduce the bandwidth at which the output stage can handle, per the slew rate and other factors. It is a very significant effect. Output stages thus almost always have 10dB or less of gain. Often unity, just as a power buffer.
Edited by DonoMan - 7/30/12 at 5:58am
post #59 of 68
Quote:
Originally Posted by DonoMan View Post

I didn't say there was a "need," but I do believe that it would be a better option though more of a pain to implement. With just a voltage divider, the output voltage will depend on the input resistance of the amplifier (which would be in parallel with one of the resistors, and would constrain the values of resistor you could use as well). It's not the end of the world but it's not a good thing either. It can be done passively, just that it's not a great solution if you want to be able to swap in any amplifier you want at a later date while having predictable results from the voltage divider.
Are you actually serious? the OP is purportedly a 4th year EE student, so I'm sure that he's capable of calculating the output voltage of a 3 resistor network. As most SS poweramps are >10k Zin. WB suggested a 360R resistor for the lower in the divider (post 38) which when parallel with a becomes a 347.5R network. WB has also mentioned an Emotiva XPA3 with a Zin of 47kR, so the difference will be even less. If he decides later to change it out later, I'm sure for the time and effort to change 3 10c resistors will be OK. By which time I'm sure he will have stumped up for a newer AVR with pre outs, so it will be moot anyway.

Quote:
Originally Posted by DonoMan View Post

I think you probably should have given some reasons at least, else anyone who listens to you wouldn't even understand why they're doing so... But anyway, thems the pros and cons, pretty much.
The only advantage to using opamps is a lower net Zout, which is irrelevant in this case as it will make such a trivial difference. The added cost ($50-100 for a PSU, case and other components) as well as the added complexity serve no useful purpose. OP is not willing to work inside his AVR, so they will be needed.
Quote:
Originally Posted by DonoMan View Post

Oh, and I'll respond to something else I saw
Even though you quoted me as saying that, note that I did not. It was Okv in post 36.
Quote:
Originally Posted by DonoMan View Post

Indeed... it's in the preamp section, not the amp section. Adding gain in the output stage will reduce the bandwidth at which the output stage can handle, per the slew rate and other factors. It is a very significant effect. Output stages thus almost always have 10dB or less of gain. Often unity, just as a power buffer.
Rubbish. The difference between a lower model AVR w/o pre outs and one with, will be the connectors and rear panel. Why would they change the gain structure of the units to have the line level different from usual levels, say 0dBV?

As for output stages, Okv was obviously talking about the poweramp stages, which given typical line level inputs will indeed have 25-35dB of gain. The actual output stages in any SS amp I have seen in 3 decades are followers, so their gain will always be less than unity.
post #60 of 68
Yeah well I've seen more issues with the passive circuit than the active. The passive LOCs for cars almost always have relatively bad reviews compared to the active ones. OP can figure out which he wants to do.

I'm right about the gain thing. Most to all gain is applied in a/the preamp stage. Goes for both low-end and high-end. It may not be an EE course but I have 4 years of electronics at a tech school (and then I went on to computer science instead). I'm not ignorant of how audio and amplifiers work.

And don't feel the need to stick up for the OP. I'm sure he can stand up for himself if he thought I was trying to be insulting, which I wasn't, and which nobody but you figured I was. Most EE people aren't into audio, anyway. I'm sorry if you just automatically assume that someone with an EE degree knows more than you do and that you are unworthy. I, on the other hand, will still share my knowledge.


PEACE
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: DIY Speakers and Subs
AVS › AVS Forum › Audio › DIY Speakers and Subs › Guide to install pre-outs on an AVR