HollowState said:
When the windings are symmetrical, the electrical center will be at the center point of the windings.
The use of the word "true" isn't really saying anything.
Either the impedances are balanced or they're not. If the impedances are balanced, then it's balanced. Period. Ain't no "quasi" or "pseudo" about it.
No center tap is required for common-mode rejection.
se
There you guys go again, hijacking someone elses thread and confusing the heck out of me again 🙁
I'm beginning think that I can wire this baby up with bailing wire and old tin cans and a. it will still work and b. someone on this site will tell me i did it the right way and another will say i did it all wrong 😉
I'm beginning think that I can wire this baby up with bailing wire and old tin cans and a. it will still work and b. someone on this site will tell me i did it the right way and another will say i did it all wrong 😉
limey222 said:
Sorry. 🙁
Ok, so, in the first paragraph of that mess you posted from the seller of this thing, he says:
This input accept any normal line level, high impedance source (outboard solid state or tube mono or stereo preamplifier, designed for home or studio use, CD player outputs, Tape outputs etc. The volume (gain) of this input (as well as the mic input) is controlled by the master volume control as shown.
Now, if you can control the volume with the master volume control, why do you need some attenuator on the output?
se
Steve Eddy said:
I am assuming (and I don't mean this as a criticism) that you are probably a hifi purist, however in the world of recording studios (and musicians) being able to vary the amount that you drive a tube amp changes the tone by introducing controlled amounts of distortion which many find pleasing to the ear. So it goes that if you drive the tube harder the output signal probably increases thereby potentially overloading the input of the studio tracking console.....hence the need to lower the signal level while maintaining the over-driven condition...hope that makes sense.
limey222 said:
As a guitar player who has owned a number of tube amps over the years (including one I had modified to go to "12" 😀), I know exactly what you're talking about here. I just didn't know that overdrive was something one wanted in a microphone preamp.
Yes, I understand perfectly now.
So, next question. Don't mixers have attenuators on their inputs? Or do the inputs go straight into an active gain stage before any attenuation, thereby requiring that maximum levels be limited ahead of the mixer's input?
se
Steve Eddy said:
Yes, under normal circumstances they do although the less expensive brands are pretty weak in this area. Older. expensive mixrs have nicely- designed attenuation circiuits. What has become common in studios today is the elimination of the mixer altogether and going from an external mic pre directly into a A/D converter like an Apogee or direct into a PCI sound card with onboard converters like a Lynx Studio Lynx2 card, very high end stuff but with digital attentuation of incoming signal (if they have any) which is not the ideal way to go. Most top quality sound cards ($1,000 plus) are designed to deliver the best audio at digital "0".
limey222 said:
Mmmm. Ok.
So, do you think 6dB of attenuation would be sufficient to keep the output level of the Roberts low enough not to overload the inputs to your mixer?
Also, do you have any idea at all as to what sort of maximum output levels you'll be getting out of the thing? That will be very helpful to know if you're going to be using any sort of output transformer.
Digital "0"? 'Splain.
se
Steve Eddy said:
Digital "0" the point at which going above results in terribly harsh sounding distortion. Digital is not as forgiving as analog.
Actually I have no idea of what my output will be, I'm just going by the warning that the seller gave me about not sending too hot a signal from the current unbalanced output jack of the roberts amp. Basically I've been swimming in a sea of confusion ever since. I have an an Alps Blue Velvet 100K, a 500 ohm series resistor, a Jensen 1:1 ratio output transformer (for balanced operation) and an XLR jack and that's the order that i plan to install them in on the output side of the amp. At this point after so much conflicting advice I'm just going to try it and see.
I've got a pretty god handle on what i need to do on the input side of the amp. Feel free to comment/advise/warn etc.
limey222 said:
Yes, I know you don't want to clip digitally. Just that "digital 0" doesn't seem to give any sort of frame of reference. Depending on the ADC, you could reach full scale at virtually any input level and unless you know what that input level is, "digital 0" doesn't seem to be saying anything.
Anyway...
Understood.
Ok.
What I would recommend though would be getting rid of the 500 ohm series resistor and the 100k pot. Instead, get a dual gang 5k audio taper pot (i.e. Mouser 313-2420F-5K) and wire it across the transformer secondary:
An externally hosted image should be here but it was not working when we last tested it.
If you like the overall result, you can toss the cheapie pot and make yourself a nice switched attenuator using precision resistors.
se
What I would recommend though would be getting rid of the 500 ohm series resistor and the 100k pot. Instead, get a dual gang 5k audio taper pot (i.e. Mouser 313-2420F-5K) and wire it across the transformer secondary:
So am i right in assuming that pot you are suggesting becomes a sort of variable H-pad? If that is so, why would you not just stay with the pot instead of using it as a tool to determine what fixed values you would need in a stepped resistor? If the final outcome is an H-pad I assume that this is something that i have to assemble myself. This is the part that confuses me and is the original reason why I was going to put the Alps pot prior to the transformer onthe unbalanced side. This is getting interesting, thanks for staying with the thread.
So am i right in assuming that pot you are suggesting becomes a sort of variable H-pad? If that is so, why would you not just stay with the pot instead of using it as a tool to determine what fixed values you would need in a stepped resistor? If the final outcome is an H-pad I assume that this is something that i have to assemble myself. This is the part that confuses me and is the original reason why I was going to put the Alps pot prior to the transformer onthe unbalanced side. This is getting interesting, thanks for staying with the thread.
limey222 said:
In previous replies from other posters, a great deal of emphasis was placed on presenting the correct impedance to the original "speaker" transformer which is still in the circuit providing the output we are using. In case you haven't read that far back, the seller modded the amps to have only one output, the one coming from the speaker transformer terminating at an unbalanced output jack, hence it is a very hot signal. Doesn't it neeed to see some sort of constant impedance prior to the use of any variable pot. I was under the impression that the seller instructed me to install the 500 ohm series resistor on the unbalanced output side. He sent me a sketch which I can certainly pass on to you if you send me your email address since for some reason it won't attach to these postings.
limey222 said:
Would something like this work?
http://cgi.ebay.com/Stepped-100K-St...4QQihZ002QQcategoryZ73369QQrdZ1QQcmdZViewItem
limey222 said:
No, it's not any sort of H pad. It's just two potentiometers being used to preserve output balance. An H pad maintains the same input and output impedance. With a potentiometer, it has a fixed input impedance but a variable output impedance.
The advantage of the potentiometer is that its worst case (i.e. highest) output impedance, which occurs at the midpoint (resistively, not rotationally unless you're using a linear taper pot), is only one fourth the end to end resistance. With an H pad, its output impedance would always be the same as its input impedance.
So let's say you wanted the output of the Roberts to drive a light, 10k ohm load. If you were to do this with an H pad, then your output impedance would also be 10k. Not good.
On the other hand, if you use a 10k potentiometer, at its worst, the output impedance would be 2.5k (10k divided by four). Above and below the resistive midpoint, it will be less.
Actually, since the guy selling it says it can easily drive even 600 ohm loads, instead of the 5k dual gang pot, you could probably get the 1k version. The Roberts would see about a 2k load and your worst case output impedance would be only 500 ohms.
You're welcome. In spite of some of the confusion, it's been fun overall. 🙂
se
limey222 said:
Yes, I read that far back but it wasn't quite clear to me from what the seller wrote whether the output was coming from the speaker output or from the driver stage.
Well, the pot will present a constant impedance. But if the output is coming from the speaker output, then the question becomes does it need to see an impedance similar to that of the original loudspeakers for best performance? If so, then using a 500 ohm resistor in series with the output won't accomplish that.
Please do. You can EMail it to me at steve@q-audio.com
se
limey222 said:
Well, not quite.
First, it's just a single attenuator, not a dual (it has two decks because it's a ladder type). You'd need a four deck version. Second, it's 100k, which is way too high.
se
Steve Eddy said:
I found a circuit diagram which indicates that the existing output (speaker) transformer is listed as (5Kohm : 8 Ohm) I assume that the 5k side is the primary presented to the tube output and that the 8 ohm side is for direct connection to an external speaker since it terminates in an unblanced output jack. Still don't understand why the circuit diagram shows a 100 ohm resistor strapped across the secondary??
Here's something that might be of interest. because i am extremely wary of tube circuits I tried a little experiment. I hadn't actually tested the mono block amp since I received them so i figured that i should atleast do that before I begin modding them.
So as they exist at this time they are in an open chassis condition. The power lead to the power transformer is just a 2-conductor lead, no ground (I will obviously be changing that when I rack-mount and encase them in enclosures.
So I dug out a Shure plug-in mic transformer and plugged it into the unbalanced mic input jack and connected a mic. On the other side I connect the unbalanced line out to the line input of a JBL powered floor monitor that we use for rehearsals (obviouly the volume was turned way down). I powered the mono block up and allowed it to warm up, then I slowly increased the gain on the mono block until i got a satisfactory voice level out of the JBL's...so far so good, it sounded clean, no hiss or hum. I then tried exactly the same proceedure with the other mono block with exactly the same results. I'm pleased with these preliminary results so far, however there was one area of concern and I don't know if this is normal or not. If I put a voltmeter set at 200VAC range with one probe on the chassis and the other in a spare ground socket in a nearby wall receptacle, i get a reading of 57 volts. I get the same reading if i keep the grounding the same and touch the probe to the mic body in it's clip on the mic stand. The same happens with the second mono block. Is this normal? I can then disconnect the mains power and still get the same 57 volt reading. It is only after 3 or 4 minutes of being disconnected from the mains that both amps no longer show any voltage readings. Now when they are racked and the chassis is grounded at the point of incoming power with the IEC socket I don't expect to see this reading, can anyone tell me if all of this is normal. It would be bad form to electrocute my singer.
So as they exist at this time they are in an open chassis condition. The power lead to the power transformer is just a 2-conductor lead, no ground (I will obviously be changing that when I rack-mount and encase them in enclosures.
So I dug out a Shure plug-in mic transformer and plugged it into the unbalanced mic input jack and connected a mic. On the other side I connect the unbalanced line out to the line input of a JBL powered floor monitor that we use for rehearsals (obviouly the volume was turned way down). I powered the mono block up and allowed it to warm up, then I slowly increased the gain on the mono block until i got a satisfactory voice level out of the JBL's...so far so good, it sounded clean, no hiss or hum. I then tried exactly the same proceedure with the other mono block with exactly the same results. I'm pleased with these preliminary results so far, however there was one area of concern and I don't know if this is normal or not. If I put a voltmeter set at 200VAC range with one probe on the chassis and the other in a spare ground socket in a nearby wall receptacle, i get a reading of 57 volts. I get the same reading if i keep the grounding the same and touch the probe to the mic body in it's clip on the mic stand. The same happens with the second mono block. Is this normal? I can then disconnect the mains power and still get the same 57 volt reading. It is only after 3 or 4 minutes of being disconnected from the mains that both amps no longer show any voltage readings. Now when they are racked and the chassis is grounded at the point of incoming power with the IEC socket I don't expect to see this reading, can anyone tell me if all of this is normal. It would be bad form to electrocute my singer.
limey222 said:
Yeah, 5k is the primary.
Well, if that's the case and the output of this thing really is coming off the speaker output, I guess it would be best to load it with something in the 8 ohm region as the load on the tranformer secondary establishes the plate load for the tube.
You could just get a regular ol' 8 ohm L pad and put it on the output ahead of the output trannie and you should be good to go.
As for the 100 ohm resistor, I'm not really sure, though I seem to recall it being said that you don't want to run a tube amp like this without a load on the secondary so perhaps it's there for protection reasons. It's large enough so that in parallel with 8 ohms it's not going to waste a lot of power, but small enough to provide a sufficient load when not connected to a speaker.
By the way, here's the scan you sent me so others can see it. He's not using the 500 ohm resistor for attenuation purposes, but instead he seems to be using for 600 ohm "impedance matching" which as EC0810 rightly stated preivously went out with the Ark. So I would just forget about that 500 ohm resistor.
An externally hosted image should be here but it was not working when we last tested it.
se
Interconnect wire recomendations?
For short runs between XLR input and Input transformer primary and also bewteen Input transformer secondary and plate would you recommend shield twisted pair, twisted pair or shielded single solid conductor wire. The run lengths are only about 3"- 4"
Any recomendations of particular products? Could I use shielded coax for the single conductor wire? Also do i terminate the shield at both ends ?
For short runs between XLR input and Input transformer primary and also bewteen Input transformer secondary and plate would you recommend shield twisted pair, twisted pair or shielded single solid conductor wire. The run lengths are only about 3"- 4"
Any recomendations of particular products? Could I use shielded coax for the single conductor wire? Also do i terminate the shield at both ends ?
Steve Eddy said:
"Instead, get a dual gang 5k audio taper pot (i.e. Mouser 313-2420F-5K) and wire it across the transformer secondary"
Could i still just use the dual gang pot across the secondary given your above comments since I already have them, instead of the simple L-pad prior to the output transformer?
"
- Status
- Not open for further replies.