|
Usually I use cascoded mosfets, but yeah they need quite a bit of voltage. Even worse, the gate capacitance is very nonlinear until Vds is rather high (>10 volts). So most often when I'm using them for audio I have at LEAST 35v+ across the CCS. Sometimes 50v+. And since it's audio I'm never anywhere near saturation.
|
#
?
Apr 8, 2011 02:33
|
|
|
|
| # ? May 28, 2024 11:19 |
|
|
I did some simulation tests out of curiosity. I sure wish I had all these devices laying around with a frequency analyzer, but hey. All at 25V, ~2.35mA current output. LND150 Depletion MOSFET (1 active component) Ro@1Hz = 7.8G, 76kHz BW Turnon = 2.1V dV/dI = 2.1M 2N5457 JFET (1 active component) Ro@1Hz = 770k, 429kHz BW Turnon = 1.9V dV/dI = 760k 2N2222 BJT mirror (2 active components) Ro@1Hz = 54k, 643kHz BW Turnon = 410mV dV/dI = 54k 2N7000 MOSFET cascode mirror (4 active components) Ro@1Hz = 3.1G, 14Hz BW (  )Turnon = 1.6V dV/dI = 4.4M 2N2222 BJT cascode mirror (4 active components) Ro@1Hz = 6.2M, 30kHz BW Turnon = 1.1V dV/dI = 4.1M@1.1V, 4.9M@10V TL071+2N3055 feedback source (2 active components) Ro@1Hz = 503M, 14Hz BW (again?) Turnon = 1.5V (not really relevant as this is easily changed in this topology) dV/dI = Inf Cascodes really give no fucks about voltage across the output, but the feedback topology Battlemaster is using doesn't even know what voltage is. It's probably a simulator limitation but needless to say it rejects DC changes real good. I still prefer single-component current regulation/simple mirrors for implementation reasons, but if was ever having some PCBs made I'd probably give a cascode a go.
|
|
#
?
Apr 8, 2011 04:55
|
|
|
I feel like I have outgrown my DSO Nano. What is the best bang/buck for an entry level oscilloscope? There seems to be a few units floating in the $300-400 range from Rigol, Owon and Uni-Trend. What is the general opinion for USB-scopes in the same price range? I don't want to "get a used Tek", I don't have the space for it.
|
|
#
?
Apr 8, 2011 17:43
|
|
|
http://www.madelltech.com/m1-4.html I've heard good things about this scope (the CA8016, last one listed at $150) and, assuming they are true, it's a steal. 10MHz isn't fantastic, but it's good enough for most hobby work. Single-channel sucks, but X-Y can be really handy. edit: Also budget digital scopes, like the cheap Rigols, have a lot of issues with crappy update rates. I don't know about the DSO Nano, but I assume it's less than ideal. Even the cheapest analog scopes will certainly not have this problem. sixide fucked around with this message at 17:56 on Apr 8, 2011 |
|
#
?
Apr 8, 2011 17:51
|
|
|
I don't know if I'd EVER buy a scope without trigger holdoff like that 8016 though. Can make it essentially impossible to view any non-continuous signal (periodic digital stuff on a serial data line for instance). A good analog scope has a couple advantages (like viewing squarewave risetimes in the brightness), but then DSO's have some serious advantages as well (lots of math, measurements, single-tracing, sometimes storage, averaging, and more...) There's also Atten, but they have many various models with wildly different specs. Like the ADS1062CA or 1102CA which supposedly are pretty decent and have like 1M sample memory, while the 1062C and 1102C have only 4k, so you have to be careful.
|
|
#
?
Apr 8, 2011 19:06
|
|
|
Rescue Toaster posted:There's also Atten, but they have many various models with wildly different specs. Like the ADS1062CA or 1102CA which supposedly are pretty decent and have like 1M sample memory, while the 1062C and 1102C have only 4k, so you have to be careful. That ADS1062CA looks nice. I mostly tinker with audio and very basic digital stuff so I don't think I need a very high bandwidth but dual channel is a must.
|
|
#
?
Apr 8, 2011 21:41
|
|
|
Is it safe to assume an 1100uf capacitor from a camera flash can kill me if I'm not careful?
|
|
#
?
Apr 8, 2011 21:55
|
|
|
I mean, it depends on the voltage its been charged to, the equivalent series resistance (which is usually rather low for flash caps), and other things. I mean, I'm building annoying circuits at 4KV, and even at nanofarads, they hurt like a big to discharge.
|
|
#
?
Apr 8, 2011 22:19
|
|
|
GWBBQ posted:Is it safe to assume an 1100uf capacitor from a camera flash can kill me if I'm not careful? It is definitely safer to assume that.
|
|
#
?
Apr 8, 2011 22:27
|
|
|
Rescue Toaster posted:Battlemaster - assuming your opamp is supplying the 3055's base current without issue (does your opamp get hot?) it's going to be rock solid and there's really no reason to change your circuit if it's working for you. During my tests the op-amp stays cool but at above two amps or so the 2N3055 heats up at an incredible rate. Which is expected, because with a low-resistance load there is a lot of excess voltage for it to drop. So yeah, everything's A-okay! GWBBQ posted:Is it safe to assume an 1100uf capacitor from a camera flash can kill me if I'm not careful? Capacitors are scary and you should never dick around with them, but I know people who've used camera flash capacitors to deliver painful but non-lethal shocks. They're crazy but still alive. BattleMaster fucked around with this message at 23:06 on Apr 8, 2011 |
|
#
?
Apr 8, 2011 23:03
|
|
|
GWBBQ posted:Is it safe to assume an 1100uf capacitor from a camera flash can kill me if I'm not careful? It's about as far from "safe" as you can get but it's probably not deadly. I've stupidly managed to grab both terminals of a 1000u cap charged to 85V and lived, though if you stuck that on my bare chest I might still be in the hospital. BattleMaster posted:During my tests the op-amp stays cool but at above two amps or so the 2N3055 heats up at an incredible rate. Which is expected, because with a low-resistance load there is a lot of excess voltage for it to drop. It's too bad they suck so much base current or it'd be nice and easy to parallel a pair of them to limit heat dissipation.
|
|
#
?
Apr 8, 2011 23:09
|
|
|
sixide posted:It's too bad they suck so much base current or it'd be nice and easy to parallel a pair of them to limit heat dissipation. Shouldn't be a huge deal, though. When I get a bit more free time I am going to drill out a heatsink from a bar of aluminum and mount the sucker to it. I've also got a bunch of computer heatsinks and fans I can attach to it if that isn't enough. It will be makeshift as hell but should work okay. Yeah, I realize they make TO-3 heatsinks but I already own all the parts for this!
|
|
#
?
Apr 8, 2011 23:15
|
|
|
I'm trying to put together a cmoy and I've got all the parts, but for the switch instead of using the toggle switch they suggested I decided to use a rocker switch instead. The one I chose was this one which has 2 terminals rather than the 3 that the tutorial suggested. Did I gently caress up or will the one I chose work? In case you guys haven't deduced it yet, I'm COMPLETELY new to any kind of electronics stuff. Do I just hook up both ground wires to the one side of the switch?
Dolphin fucked around with this message at 21:41 on Apr 9, 2011 |
|
#
?
Apr 9, 2011 21:37
|
|
|
Dolphin posted:I'm trying to put together a cmoy and I've got all the parts, but for the switch instead of using the toggle switch they suggested I decided to use a rocker switch instead. The one I chose was this one which has 2 terminals rather than the 3 that the tutorial suggested. Did I gently caress up or will the one I chose work? In case you guys haven't deduced it yet, I'm COMPLETELY new to any kind of electronics stuff. Do I just hook up both ground wires to the one side of the switch? They only use two pins of their switch anyways. You should be fine.
|
|
#
?
Apr 9, 2011 21:57
|
|
|
Dolphin posted:I'm trying to put together a cmoy and I've got all the parts, but for the switch instead of using the toggle switch they suggested I decided to use a rocker switch instead. The one I chose was this one which has 2 terminals rather than the 3 that the tutorial suggested. Did I gently caress up or will the one I chose work? In case you guys haven't deduced it yet, I'm COMPLETELY new to any kind of electronics stuff. Do I just hook up both ground wires to the one side of the switch? On an unrelated note, I always considered this to be a horribly designed amplifier. I mean, it might've been neat 20 years ago, but a lot could be done more effectively with a circuit that wasn't a relic.
|
|
#
?
Apr 9, 2011 22:31
|
|
|
therunningman posted:What is the general opinion for USB-scopes in the same price range? This is the USB scope I use, and it's worked out very well for me. It's $250 unless you're a student then there's a discount available. http://www.linkinstruments.com/mso19.htm
|
|
#
?
Apr 10, 2011 01:27
|
|
|
Corla Plankun posted:What's the story behind the pic in your av
|
|
#
?
Apr 10, 2011 04:02
|
|
|
That's GIR from Invader Zim. Anyway, I got the PCB's for my next LED project a couple days ago:  This will be a large 7-segment display board to show competitor #'s for a couple different events (mostly horse shows), will show what class is in progress, who's performing, who's up next, second, third, etc... so people know how long they have. The boards sit back to back connected by pin headers, and the top board has the radio installed and they daisy-chain together with Cat5.
|
|
#
?
Apr 10, 2011 18:29
|
|
|
As an experiment I just built a boost converter using a 555 timer. I set it up so the duty cycle can be varied by a potentiometer but the frequency stays stable. By experimenting with different frequencies and inductors I've been able to get over 300 volts out of 12 volts with it. At higher duty cycles (and thus higher output voltages) the variable power supply I'm using to power the project emits high-frequency whines. Anyone have an idea of what's causing this and if it's dangerous?
|
|
#
?
Apr 11, 2011 01:10
|
|
|
Remember that a boost converter, when 'on' (at high duty cycles as you approach 100%) will start to appear as a short circuit to the supply. As you increase on time, eventually the inductor's magnetic field is completely saturated and it no longer offers any resistance to the flow of electricity. So yes, as you turn up the duty cycle, you will stress the supply more and more until it (most likely) goes into current limiting. Depending on the supply it could overheat or be damaged, though.
|
|
#
?
Apr 11, 2011 01:29
|
|
|
I've had it make similar sounds when I've shorted it by accident so that follows. The weird thing is that the current draw, according to the display, never exceeds 500mA or so at high duty cycles though that may be in error because the current is pulsed. I've noticed that different inductor values cause the whine to start at different duty cycles. I've gotten the best results using a frequency of about 4kHz and a really big 2.2mH inductor but the whine starts before 50% duty cycle!
|
|
#
?
Apr 11, 2011 01:33
|
|
|
High frequency whines are almost always inductor (transformer) saturation, since they can start to mechanically vibrate. Could there be DC through a transformer winding in your supply as a result of the switcher as a load?
|
|
#
?
Apr 11, 2011 01:50
|
|
|
It's a commercial supply rather than one I built myself so I don't know what's going on inside other than that the box said it's a switching supply. It doesn't hit the current limit when powering my boost converter, and if I lower the current to below the draw it emits a progressively lower-pitched sound. I'm just going to leave it alone if there's any cause to believe that I am risking damaging it! I'm not too keen on buying a new supply.
|
|
#
?
Apr 11, 2011 18:49
|
|
|
Say I've got a lovely townhouse-style garage. It's got a 15 amp breaker on it, which I believe may also be shared between some parts of the kitchen. Now say that sometimes--not often, but sometimes--when I turn on my tablesaw, I trip the breaker. Would it be a Very Bad Idea™ to just swap in a 20 amp breaker on it? I don't believe the circuit would ever need to sustain 20 amps, just while my saw spins up and draws more current for all of a second. I realize that running 20 amps continuously could be a problem if the wiring isn't heavy enough, but I can't imagine the difference would be that critical, especially if the bigger draw was only temporary.
|
|
#
?
Apr 11, 2011 20:12
|
|
|
You probably ought to ask the wiring thread instead of the electronics thread, but check what wire that branch circuit has. If it's #14 copper or #10 aluminum you can swap breakers, anything less and you really shouldn't.
|
|
#
?
Apr 11, 2011 20:19
|
|
|
Oh, I didn't even know there was a wiring thread. Thanks. 
|
|
#
?
Apr 11, 2011 20:23
|
|
|
This is not the best thread, but does anyone know much much current the bias voltage output in a computer microphone port might be able to supply? I have a contact microphone with a built-in preamp, and it's behavior might be erratic if the voltage drops too low due to current limiting.
|
|
#
?
Apr 11, 2011 21:18
|
|
|
AFAIK microphone jacks have a serious lack of standardization. Some are +5V bias, some are +3.75V, others are +2.5 and I haven't found any mention of typical limiting current. I'm sure a typical electret up to 2mA or a little more would work fine, but I have no idea about piezo transducers.
|
|
#
?
Apr 12, 2011 01:58
|
|
|
Bad Munki posted:Say I've got a lovely townhouse-style garage. It's got a 15 amp breaker on it, which I believe may also be shared between some parts of the kitchen.
|
|
#
?
Apr 12, 2011 04:05
|
|
|
Yeah, it was suggested I try something like this as a safe alternative. I'll almost certainly go with that instead. Still, I can't wait until I move in a year and can get a place I can actually re-wire a bit to have better service where I need it.
|
|
#
?
Apr 12, 2011 04:11
|
|
|
So the Tesla Orchestra is doing an internet show called The open spark project on may 14, in which we'll be playing music submitted by people all over the world and streaming it live. Hopefully we'll have our 20KW PFCs done by then, so we can do 10 foot arcs instead of five foot ones... If you're interested in the musical aspect of it, then I've started a thread in the Musician's lounge.
|
|
#
?
Apr 12, 2011 13:53
|
|
|
Huh, turns out my configuration for a variable-duty cycle fixed-frequency 555 timer doesn't actually have a fixed frequency. I came up with the circuit myself and was racking my brain trying to figure out why it gives frequencies at minimum duty cycle that are about double the ones at the maximum duty cycle. I thought there was a flaw with it but when I did a search to find ideas it turns out that my circuit isn't as original as I had thought and that it was definitely supposed to work properly. (Not my schematic, but is on my hosting) I'm using exactly the same circuit with different component values and a different output circuit, but I have no idea why the frequency varies so drastically as I turn the potentiometer! The duty cycle changes exactly as expected, but the frequency at all duty cycles should be pretty close to 1/(ln2*C*R) where R is the total series resistance of the pot plus two resistors plus 2 times the wiper resistance. What's going on here? BattleMaster fucked around with this message at 01:49 on Apr 13, 2011 |
|
#
?
Apr 13, 2011 01:46
|
|
|
I'm too tired to figure out what the purpose of D2 is, but I know why D1 is there for (so that the duty cycle can be less than 50%, yes?) Excluding that, your frequency equation may be off. The f in your equation should: R = Ra + Rb = R1 + VRa + VRb + R2 where the series resistance of VR = VRa + VRb and without writing out the rest of the algebra: R = R1 + 2R2 + VR + VRb = 130kOhm + VRb where VRb can vary from 0 to 100kOhm, which means that the full range of of R will vary from 130kOhm to 230kOhm.... You can either Try making R1 and R2 much larger than the total resistance of VR, or connecting up 2 555s (or using one of the multiple 555 ICs), with the first setting the frequency, and the other changing the duty cycle. That said, this all depends on D2 doing nothing. I really don't know why its there.
|
|
#
?
Apr 13, 2011 02:09
|
|
|
Dammit, I thought the resistance wouldn't vary because I thought the diodes cause the capacitor to charge through R1 + one half of the potentiometer, and discharge through R2 and the other half of the potentiometer. Oh well. Thanks.
|
|
#
?
Apr 13, 2011 02:20
|
|
|
D2 prevents the capacitor from charging off the Ra+Rb chain and only charges from the potentiometer wiper through D1. So, the cap charges from the wiper and discharges through Rb. The biggest problem with that schematic is the fact that CV is grounded and ground is AC coupled, but that's a pretty obvious mistake.
|
|
#
?
Apr 13, 2011 02:28
|
|
|
Haha, I didn't notice that. I definitely didn't make that mistake in my setup.
|
|
#
?
Apr 13, 2011 02:30
|
|
|
sixide posted:D2 prevents the capacitor from charging off the Ra+Rb chain and only charges from the potentiometer wiper through D1. Hah, I didn't even notice that. I assume he hooked it up right when trying it, though... I threw the circuit into a (sorta crappy) sim, and the freq variation is much less than I thought, so I couldn't tell you what the problem is now. I might take a better look when sober. Here's the sim link: CLICK
|
|
#
?
Apr 13, 2011 02:32
|
|
|
I tested it in LTSpice and it should be deadly stable, unless perhaps there is something special about the diodes like high ohmic resistance or junction capacitance.
|
|
#
?
Apr 13, 2011 02:35
|
|
|
I've been playing with component values and I see the same problem with any combination. Right now I have it hooked up with a 100k linear pot and R1=R2=10k as per the posted schematic. I'm using a 10nF capacitor. Diodes are 1N4001s (would crappy old diodes even make a difference?). Frequency at miniumum duty cycle 24.9kHz. At maximum it's 11.5kHz. The frequency seems stable at higher duty cycles but completely flies off at under 55-65% duty cycle or so. Edit: Welp looks like I might need to try better diodes.
|
|
#
?
Apr 13, 2011 02:38
|
|
|
|
| # ? May 28, 2024 11:19 |
|
|
4001s are a lot beefier than needed but I don't see why they'd cause any issues. You're looking at maybe 30pF instead of 4pF and significantly less resistance than the 4148. You should be seeing frequencies in the neighborhood of 1kHz with those values. 11.5kHz minimum seems drastically off the mark.
|
|
#
?
Apr 13, 2011 02:50
|
|