This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
Messages - Kaz Kylheku
« on: May 19, 2013, 03:35:07 AM »
First of all, about wattage. If you look at the ambient-temp versus dissipation graph, it gives 850 as the maximum without a heatsink, which derates above 25 degrees at 6.75 mW per degree celsius. If you want anything more than 850 mW out of these chips, you must stick a heatsink on them.
The thing is also quite sensitive for the amount of headroom. An input well below 100 mV drives to saturation with THD above 10% (grossly audible). See the Po, THD - Vin graph on P. 6. Basically, you can easily clip this thing out by directly plugging a guitar pickup into it, and one that isn't even hot. Something like a DiMarzio PAF will put out 300 mV, supposedly.
It's not clear to what extent if at all this lack of headroom will improve with a heat sink.
If all you want is to amplify the output of a preamp into a small wattage, you basically do not need any voltage gain beyond what the preamp is capable of. You just need an output stage that can drive the speaker (e.g. class B push-pull emitter follower, with a buffer in front of it so the preamp doesn't have to deal with the low-ish impedance of that stage).
About the noise that you get when you connect the circuits together, from the description it sounds as if it might be motorboating: a positive feedback loop through the shared power supply. Motorboating is a common plague of single-supply circuits with multiple AC-coupled stages. The circuits have poor power supply rejection, and so parasitic feedback loops form through the power supply connections. These AC couplings introduce phase shift at their cutoff frequencies, and so you can end up with a low frequency phase shift oscillator.
I see the that the Fetzer is just a JFET based common-source stage. But it is an inverting stage. So if there is feedback into its input from the output currents of the amp, it will be inverted, and so 180 degrees of phase shift is enough, which could be achieved by just two capacitive couplings (90 degrees each).
Can you post the complete circuit diagram: Fetzer plus your AN7511 amp?
« on: May 19, 2013, 02:05:28 AM »
Have a look at the JFET input stage of the ADA MP-1. It's a neat circuit which places an N-channel JFET with a PNP transistor in a kind of hybrid complementary pair, which is then used as a source follower (unity gain: it's not a gain stage!) The JFET provides the high input impedance, and the PNP delivers the current, allowing for a fuller voltage swing. The circuit runs on dual 15/15 rails.
I'm mentioning it because of the LTSpice appearance in this thread, and that circuit happens to be something I've been simulating in LTSpice recently. (The transistor part values are not the real ones.)
And, hey, use good parts in this USB interface! You will never beat the M-Audio or Behringer and whatever *s!!t* if you do what they do: use *s!!t* parts.
I found JRC4558's in the preamp section when I opened a Tascam US-122L audio interface. Ugh!
Also read this document: http://www.rane.com/note151.html
And google "pin 1 problem".
« on: February 18, 2013, 05:49:31 PM »
Bucket brigade delays (BBD's) don't count because they are discrete time analog: the amplitude is represented by capacitor charge, not by 1's and 0's. Digital means discrete time and discrete amplitude.
ADA MP-1 (which, incidentallly, has a BBD as the basis for its chorus) is not a digital signal processor at all. All signal processing is analog. The processor just stores and establishes gain/volume and signal routing settings for various circuits.
The MP-1 can be summarized as:
* Discrete FET input stage followed by a programable attenuation feeding an op-amp buffer with soft diode clipping driving either a tube channel or SS channel
* a four-triode (2 x 12AX87/ECC83) tube board with two FET-switched voicings and a programmable attenuation between the second and third stages.
* solid state channel with NE572 compression chip
* a master volume section shared by both channels feeding into a four-band EQ based around a quad op-amp
* a chorus based on an MN3007 bucket brigade chip, with a NE570 compander around it for noise reduction ("stereo" achieved by inverting the chorused signal into the other channel).
* switchable effects loop after EQ, before chorus.
* all glued together with op-amp buffering running on -15/+15 rails.
« on: February 08, 2013, 09:39:59 PM »
The amp is clearly supposed to be called Symphony, but the front plate actually says Gymphony.
The treble clef is a fancy G, not an S.
« on: January 24, 2013, 08:38:52 PM »
The power amp has a huge influence on sound, solid state or not.
A solid state voltage amplifier (global negative feedback based on output voltage) sounds nothing like a similar (or the same) amplifier in with mixed mode feedback (global negative feedback based on a mixture of the voltage applied to the speaker and the amount of current flowing through the speaker). The difference in response cannot be obtained with an equalizer. Creating the tone of mixed feedback in the preamp and then using a voltage amplifier falls into the territory of emulation, raising issues of whether it is true to life, whether it sounds right at full volume, etc.
Moreover, power amps can have tone controls which are in the feedback loop. If there is current feedback, then these controls interact with the speaker, and so they "do things" that a preamp EQ does not. For example, a bass control in the negative feedback loop will increase speaker resonance, which is not the same thing as equalizing for more bass.
Besides types of negative feedback we could probably talk about other possible effects that could happen in power amps, but that's one big one.
« on: January 23, 2013, 03:52:24 PM »
If the frequencies are integers, like 213 or 5193 cycles per second, then that means that each frequency has an even number of cycles in a one second interval. In other words, each waveform has a period such that one second is a multiple of that period. Therefore, in a mixture of such tones, one second is a multiple of each of their periods, which makes it a common multiple (and not necessarily the lowest common multiple). If we take a period that is a common multiple of the waveforms, and repeat that period, we get a function that is indistinguishable from a continuation of the original. Or, in other words, any phase shift of a signal by a multiple of one second is indistinguishable from the unshifted signal, if the signal contains only integral frequencies.
« on: January 22, 2013, 12:14:00 PM »
You can generate tones with the Free/Open Source program Audacity, using the Generate menu.
I usually make one second's worth of a tone and then loop it by shift-clicking the play button. If the frequency is integral (which it almost always is), the loop is perfectly smooth.
There isn't a one-step way to generate mixtures of waveforms, but you can just generate multiple tracks.
« on: January 22, 2013, 12:00:18 PM »
You're using your Crate head to boost the signal.
Since that is working for you to overcome whatever you think is wrong, it basically means that your 10W amp doesn't have enough gain in its preamp for your tastes.
Lugging around an amp head just to use its preamp as a booster is silly.
You could just use a small overdrive pedal, the 10W amp, and the speaker cab.
« on: December 30, 2012, 02:22:32 PM »
I would add that if you are on a budget, sending the amp off for mod work will be a hell of a lot more expensive than a few pedals.
Also, more than second-hand market prices for SS heads that have tons of "brootalz" to support the new realm of distortion into which you have evolved.
« on: December 29, 2012, 01:36:34 AM »
I don't get this. This TBX150 head had "great amounts of distortion" in the store, but doesn't any more? A year has passed and "The lack of distortion is pretty bad now?"
In the store, did you try it with the same guitar that you are using now?
Maybe you have active pickups and the batteries are dying?
Speaking of which, one way to get more distortion is to get louder passive pickups, active pickups, or an on-board preamplifier for your guitar to boost passive pickups.
Sometimes the reason for weak distortion is silly, like the pickup being way too far away from the strings.
Then there are poor contacts, cold solder joints, worn volume pots ... These things tend to rob your signal of bass, where you get much of the drive for distortion.
And then there are these factors: your guitar setup/action, the condition of your hands and fingers (calluses, strength) and the grooming of your fingernails. Anything that interferes with your guitar playability makes it feel like there is less distortion. When the guitar feels great, you don't need that much distortion to shred.
Soft fingertips rob your tone of sustain. And if your fingers hurt, you cannot press down the strings properly. If you take a three month vacation away from the guitar, you will not sound like Gojira (or whoever) when you first plug in simply because your fingers will not be conditioned for guitar.
Sometimes to overcome a "bad tone" day, all I have to do is slightly
file my left hand fingernails from underneath, so they don't dig into the flesh as much. The whole feeling of the fretboard changes. I also find nails that are too long on the right hand somehow interfere with picking in a way that just adds "nonspecific suckage". (I need some right hand nail length for fingerstyle guitar!). Over the years I have always noticed that when I trim the overgrown right hand fingernails, the flat-picking becomes cleaner and more accurate.
One more thing is that if you play with heavy distortion all the time, your ears and fingers somehow get used to it, and it seems like less.
Develop a taste for other tones. Crunches, bluesy tones, jazzy cleans, etc. These will get the "distortion adaptation" out of your head, so that when you come back to it, you will hear properly how heavy it is.
I have a MIDI pedal board, through which I select among a bank of 10 different tones. It's like having a 10 channel amp. Switching among these keeps it fresh.
EQ is very important, as J M Fahey notes, and you will find that your ears like different EQ at different times. After a good night's sleep, you may find your ears are sensitive to highs and you need to back off on the "presence".
Adding highs to tone can make it seem like there is more distortion. If you take a heavy, crisp, distorted tone and EQ out all the highs, it becomes a wet noodle.
« on: December 24, 2012, 02:28:42 PM »
The aluminum panel is complete, but I'm having difficulties transferring the artwork via toner transfer.
So far, the best results (though not nearly good enough) were achieved by heating the whole thing in an oven after ironing over the transfer paper, and then ironing some more. One problem is that the 1/8" slab of aluminum acts like a big heat sink with regard to the clothing iron.
The surface is wet-sanded up to 400 grit paper.
I installed the power switch the right way: I turned it 180 degrees so that you push the rocker upward to turn it on. Also, I discovered that the original hex bolts that attach the panel to the body were poorly threaded: very difficult to turn. I replaced them with attractive stainless steel machine screws with a 6x32 diameter/pitch which screw in perfectly easily.
Turns out there is a reasonably priced water jet cutting shop near our work; next time I will have this kind of thing neatly cut by machine rather than huffing and puffing.
« on: December 21, 2012, 04:58:30 PM »
Interesting that this thing has a "quasi-complementary" output stage: both transistors are NPN.
The MC33078's op-amps an all NPN output stage also. They might make a good "wine pairing" with this power amp.
Thanks for that; I will read the paper. I can see in the diagram how the current mirror sucks the same current through the 40K and 20K, so the top of the 20K is approximately V/2.
« on: December 19, 2012, 10:02:53 PM »
Yes, the regulators do drop a lot of voltage. This is inherited from the device. The original circuit board used the same type of voltage regulators (without any series resistor). The transformer is such that the filter caps charge to 30 volts, for a R-to-R of 60! Totally crazy drop-out down to the desired +/- 15V.
I played it loose by not stealing the heat sinks from the very beginning.
The original board has 12 IC's, most of which are RC4559's. But my four NE5532's can easily have about the same total current draw as all the IC's on that board.
Anyway, the wrap up of this story is that although the heatsinks helped, there was still a problem. I noticed a degradation in sound quality that happened after warmup. I could easily reproduce it by turning the unit off for a few minutes and then on again. Within a minute or two something would change in the tone (though not as drastic as before I put on the sinks). So I concluded that perhaps one or both of the regulators are shot.
I ended up transplanting the regulators from the original board. They are Motorola-branded units! Vintage.
(This was built about six years before Motorola stopped making semiconductors and spun off On Semi.) With these regulators, it sounds fine.
The heatsinks I took from the original unit had legs and were soldered to pads, but I broke those off. The sinks are light and the regs can just hold them up. On the other hand, I did something better: I screwed the sinks tightly to the regulators with small nuts and bolts. In the original unit, they were just slipped on and held by a very light leaf-spring tension. The heat transfer is better with the tight bolting. The matching hole for a bolt was already manufactured in the heat sinks, just not used.
If those particular voltage regulators survived 19 years with those heat sinks and that current draw, I figure they are proven in those conditions.
But if I build another one of these units, for sure I will for sure avoid a transformer that puts anything over 20 volts on the filter caps.
« on: December 19, 2012, 09:44:00 PM »
This could also be a case of poor common mode rejection in the amplifier. Even if a guitar is properly wired, grounded and shielded, with humbucking pickups, it's still a source of common mode noise: crap that appears on both the ground and signal wires in equal voltages.
If it's happening with numerous guitars that behave properly in other rigs, it points to the amplifier.
« on: December 19, 2012, 12:27:14 AM »
A few days ago I had a "bad tone day", which turned into two days, then three ...
Man, I was going crazy. Why is the sound so harsh?
I adjusted things. Unplugged jacks, plugged them back in.
Then I traced it to the SMF-1. I noticed the frequency knobs seemed to have become less effective. Turning down the presence somehow wouldn't filter out all the harshness.
Pop it open. Oops, voltage regulators are hot to the touch. Putting a finger on them soon brings it near the threshold of pain. Voltages seem okay though.
So, long story short, put heatsinks on them, and back to awesome tone. Creamy raging sustain; three dimensional cleans, etc.
The regs must have been near their limit, and so probably kicking some current regulation in and out, causing distortion.
Absolute night and day difference, the imaginary equivalent of which would require the combined confirmation bias of 1000 audiophiles.
On a different topic: making progress on the faceplate. It is designed, and I have procured the 19" by 1 3/4" aluminum. I am going to try printing the artwork on a color laser printer, and transferring it via heat, then clear-coating some acrylic over that.