I have built a nice little circuit that runs from the speaker output to my laptop input. There is a 100W resistor dissipating the power from the amp. I use an isolation transformer after stepping down the voltage to isolate the amp and computer's ground from each other. The signal is about 2-4V at the secondary of the transformer (not noisy at all). I use some more resistance to step the signal down to 1V or so to run into my computer. I have the mic input volume on maximum and when I play through the amp, it does not peak out. However, there is a significant amount of distortion when I play loud (still not peaking).
Any suggestions? I was told by a professor that 1V, 600 Ohms was the standard for audio input.
Any suggestions? I was told by a professor that 1V, 600 Ohms was the standard for audio input.
Sounds like you haven't really got a clue as to what you're doing.
Forget the transformer.
Look up 'voltage divider' on Wikipedia. Make Z1 1k and Z2 100R. You'll need one pair for left and one for right. Turn down the amplifier volume if necessary.
Use the battery when recording, although chances are that the computer's brick supply is isolated anyway. You can check this with a DMM. See if there is continuity from any of the mains plug pins to either side of the DC jack.
If you keep the amp turned down to ~10V peak, you can use quarter watt resistors. To put it another way, turn it down if they start to smoke
Forget the transformer.
Look up 'voltage divider' on Wikipedia. Make Z1 1k and Z2 100R. You'll need one pair for left and one for right. Turn down the amplifier volume if necessary.
Use the battery when recording, although chances are that the computer's brick supply is isolated anyway. You can check this with a DMM. See if there is continuity from any of the mains plug pins to either side of the DC jack.
If you keep the amp turned down to ~10V peak, you can use quarter watt resistors. To put it another way, turn it down if they start to smoke
Last edited:
Thats not a very good answer. There are a few "standards" for audio. For professional equipment the line level is +4 dbu (around 1 volt) for nominal level, for consumer gear its -10 dbu or about 200mv (but can vary quite a bit)and for mic inputs its anywhere from -60dbu to -20dbu. The 600ohms is for mic inputs only, and that can vary also, line inputs are usually around 50k ohms. These are nominal levels which dont include headroom, which is typically another 10 to 20 db, so the actual signal the input can handle is 20db hotter than the 200mv (for consumer gear ) or around 2 volts. (Unfortunately the loudness wars have pushed the normal level of CDs so high there is no such thing as headroom anymore.)
The macbook inputs are line level ( i cant find any exact numbers) probably consumer level. And no telling how much headroom they have.
Do what counter culture said and then use the mac meter to set the amp volume so the meter tops out a couple of segments below the max.
Last edited:
With all due respect, I wouldn't have attempted to hook thousands of dollars worth of equipment together without putting some thought into this, or without knowing what a voltage divider was. The circuit replaces the speaker cabinet, so even if I used higher resistances rated at a quarter watt, the power from the amp has to go somewhere or the amp gets damaged. The goal is to silently record out of my half stack (which really only sounds decent at high volumes). The transformer handles audio frequencies quite well and isn't doing any harm, a small price to pay on the off chance that there is a ground loop. The amp is in mono, so I only need to worry about one channel.
The question was about noise at the input jack. If I have adressed any concerns of signal impedance, is there some other reason why the signal is distorting even though it reports no peaking.
Thank you for your response.
The question was about noise at the input jack. If I have adressed any concerns of signal impedance, is there some other reason why the signal is distorting even though it reports no peaking.
Thank you for your response.
I know the distortion is coming from the computer. My setup is keyboard->amp->circuit->computer. My computer seems to be switching between two 'modes' (for lack of a better word). In one mode, the signal input appears quiet, peaking around -10dB. It distorts when I hit the keys hard. Without touching my circuit or amp, the computer will switch to a second mode (I don't know what causes it) where the signal looks as I would expect it to, meaning that it does peak out a little on loud notes (and only then it distorts) and the overall sound is much louder. It is almost as if the computer is preparing itself for two different kinds of inputs and it is switching between the two kinds on its own. Has anyone experienced this?
LOL.
These are the symptoms of a bad joint, almost certainly in your 'circuit', but it could be anywhere right up to the input jack on the laptop.
If you told us what value resistors you're using and which transformer there might be some hope of telling what's going on. You haven't even told us how the parts are connected.
As it is, you can leave the 100W resistor loading the output and use a resistive divider as I have suggested. Most solid-state amplifiers will work fine without the loading though.
A transformer is just looking for trouble, however. There are just so many wrong transformers you could pick and none that are intended for this purpose. If you want DC isolation, use a capacitor. You'll have to get somebody with more patience than I to explain how to dimension and select it though. Better still, use the battery or check the PSU.
These are the symptoms of a bad joint, almost certainly in your 'circuit', but it could be anywhere right up to the input jack on the laptop.
If you told us what value resistors you're using and which transformer there might be some hope of telling what's going on. You haven't even told us how the parts are connected.
As it is, you can leave the 100W resistor loading the output and use a resistive divider as I have suggested. Most solid-state amplifiers will work fine without the loading though.
A transformer is just looking for trouble, however. There are just so many wrong transformers you could pick and none that are intended for this purpose. If you want DC isolation, use a capacitor. You'll have to get somebody with more patience than I to explain how to dimension and select it though. Better still, use the battery or check the PSU.
You still have far too much voltage going into the Mac. A 100W resistor connected in series with the input is useless unless it is part of a voltage divider. Instead a 15K and a 220 ohm resistor, both 1/4W will do fine with the Macbook connected across the 220 ohm resistor. Also note that the Macbook needs only about 100mV of signal to work on its line level audio input.
Or you may just want to get a speaker to line level converter box like the Radial JDX
Or you may just want to get a speaker to line level converter box like the Radial JDX
First of all, can you get at the signal pre poweramp and put a line level signal into the mac? (and not add the amps distortion and noise to your recording)
Unless your amp is a TUBE AMP no load will not damage your amp. The way your speakers only sound good at high levels has nothing to do with the way the amp sound at low levels (most SS class A/B amps will actually sound better at low levels while they are still in class A (unless noise is an issue). Get rid of the transformer, it wont improve anything (except isolation) which you probably wont need. Try it , if you get a ground hum (60hz + harmonics) run the mac of the battery.
This is the circuit. The 16 Ohm is the speaker. All of the right side resistors are subject to change. The 600 Ohm is the laptop. I am going to measure its actual impedance soon. I want to put a pot in series with the 600 Ohm for volume control. Is there a better way to do this?
View image: Screen Shot 2012 03 23 at 6 57 37 PM
My amp is a tube amp, by the way. I am not opening up my amp. I don't think it's necessary as I am deducing that the distorting is coming from the computer because I am giving it too much voltage. Do computers sense for too high of an input and add additional input impedance as a safety feature?
View image: Screen Shot 2012 03 23 at 6 57 37 PM
My amp is a tube amp, by the way. I am not opening up my amp. I don't think it's necessary as I am deducing that the distorting is coming from the computer because I am giving it too much voltage. Do computers sense for too high of an input and add additional input impedance as a safety feature?
It probably was, back in the days.
These 600 Ohm stories really seem to stick...
Your circuit suits the demands for impedance matching between interconnected devices. In essence you employ an L Pad, assuming the input impedance of the computer to be around 600 Ohm.
An amplifier and a line level input are not to be impedance matched, as they don't possess a common impedance level, nor a predefined impedance ratio. Instead the source impedance (of the amplifier) can be assumed as sufficiently low (in the single digit ohm range) and the input impedance of the sink can be assumed as sufficiently high (in the multiple kilo-Ohm range). By extension the input impedance is large compared to the output impedance. Thus, simple voltage divider rules apply. No point in optimizing the network for an impedance level of, say, 600 Ohm.
A voltage divider does the trick, all you need to know is the estimated output voltage level (full scale) of the amplifier and the acceptable input voltage level (full scale) of the line input. No impedance calculation involved, it's a simple voltage ratio problem.
Regarding galvanic isolation, it's neither required (as safety is not an issue, no dangerous voltages are involved) nor helping (because a transformer can easily overload a laptop computer line input, too).
Here's your contradictory assumption. You'll find that the actual input impedance is far from 600 Ohm. Amplifier and Speaker together then form a low enough source impedance.
Cheers,
Sebastian.
PS: I know some people who wrecked their MacBook Pro's audio circuitry by applying too large voltage levels, particularly DC. The full scale input level is probably below 1V. Be careful.
Last edited:
That's right. Your actual input drive level is simply ill-defined due to the interaction of the amp, the transformer and the laptop input impedance.
The System Preferences contain a Sound setting. This one has an Input tab with a level meter. Use this to check levels and overload conditions.
How would they? They can automatically adjust recording levels in software. They have no means to foresee electrocution, though.
Computers just aren't built like that...
What amp exactly? I find it unlikely that it shouldn't have an auxiliary output or an effect send/return loop.
Last edited:
Thanks for all the advice. I stuck a 56k resistor in and the voltage is around 50-100mV. Things seem to be working much nicer. I want to build a peak detector so I know what level the input is at before I plug it in. I know how to do that, but my next question is if the voltage before and after the computer is connected is different (due to the computer's impedance which I found to be a little over 1kOhm), what is the best way to account for that difference? I believe this would mean that my circuit must have low impedance, but in order to keep the voltage across the laptop low, I had to use a the 56k resistor, which made the output impedance fairly large.
If it works, it works. If it doesn't that's a different matter.
If you want an explanation of how your circuit is working, then you have to post EXACT details of what you've got. It's no good saying 'I got a 56k resistor in there'. We're not clairvoyant. What else have you got? How is it all connected?
Draw a diagram, photograph it, and post the photo if you can't do any better.
If you want an explanation of how your circuit is working, then you have to post EXACT details of what you've got. It's no good saying 'I got a 56k resistor in there'. We're not clairvoyant. What else have you got? How is it all connected?
Draw a diagram, photograph it, and post the photo if you can't do any better.
Sorry about that.
View image: Screen Shot 2012 03 24 at 12 06 48 AM
I would be using a voltage buffer to connect it to a peak detector, so I have no impedance concerns for that part. It would be ideal to be able to tell if the voltage clears some threshold whether the load is attached or not. As it is, that can't be done because the load changes the voltage.
View image: Screen Shot 2012 03 24 at 12 06 48 AM
I would be using a voltage buffer to connect it to a peak detector, so I have no impedance concerns for that part. It would be ideal to be able to tell if the voltage clears some threshold whether the load is attached or not. As it is, that can't be done because the load changes the voltage.
Seems rather low. Sure you didnt measure the input resistance? Acording to Apple its "greater than 20k ohms" and will take up to 3 volts.
iMac (Mid 2011): External features, ports, and connectors
The macbooks will be the same as the Imacs
If your trying to record the output of a keyboard (an instrument) you DONT NEED the power amp or any circuit, just a cable!
If you are using the amp as an effects box, thats a different story.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.