Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
Snaily
Mar 5, 2006
Sluggish. Wee!
By the way, is there a trick to knowing what all the xN-series diodes and transistors do, or is it a matter of learning specs by heart? I am going to need some kind of MOSFET to drive my stacks of LEDs, but finding a suitable one that is cheap is like finding the proverbial needle...

Adbot
ADBOT LOVES YOU

Zaxxon
Feb 14, 2004

Wir Tanzen Mekanik

ANIME AKBAR posted:

that's a rather unusual implementation of the 555, though I don't see why it wouldn't work.

just a tip, rather than messing with diodes and transistors in the charging loops, an easier way to make a well behaved PWM with 555 is to use two of them. One as a fixed frequency oscillator in astable with a very high duty cycle (so it gives short negative pulses) clocking another 555 in monostable mode. Frequency is completely controlled by the first while duty cycle is controlled by the pulse width of the second, so your duty cycle varies linearly with the potentiometer setting (with the pot being in the astable 555)

if your willing to do some somewhat delicate biasing you could also use the grounding capacitor of the astable 555 and a comparator.

Snaily
Mar 5, 2006
Sluggish. Wee!

Zaxxon posted:

if your willing to do some somewhat delicate biasing you could also use the grounding capacitor of the astable 555 and a comparator.

Any hope for a schematic of this?

Zaxxon
Feb 14, 2004

Wir Tanzen Mekanik

Snaily posted:

Any hope for a schematic of this?

uh maybe when I get home. I can give something you can plug into the java circuit simulator to see how it goes though.

here is the circuit sim code (just pull up file import and paste this in)
here is the applet circuit sim: http://www.falstad.com/circuit/

code:
$ 1 5.0E-6 10.20027730826997 50 5.0 43
165 80 64 112 64 2 5.0
R 144 0 32 0 0 0 40.0 5.0 0.0 0.0 0.5
w 144 0 144 32 0
w 208 96 208 32 0
w 208 32 144 32 0
w 144 32 48 32 0
r 48 32 48 80 0 1000.0
r 48 96 48 144 0 100.0
w 80 160 48 160 0
w 80 192 48 192 0
w 48 192 48 160 0
w 48 160 48 144 0
w 48 96 80 96 0
w 48 96 48 80 0
c 48 320 48 416 0 1.0E-5 2.1013884759746198
g 48 416 48 432 0
w 48 192 48 320 0
w 48 320 288 320 0
a 320 336 544 336 1 5.0 -5.0
w 288 320 320 320 0
O 544 336 592 336 0
w 144 0 320 0 0
w 320 0 320 256 0
r 320 256 320 352 0 5794.0
g 320 432 320 512 0
r 320 352 320 432 0 10000.0
x 355 260 574 264 2 14 Change Me to change Pulse Width
w 400 272 368 288 0
w 368 288 368 272 0
w 368 288 384 288 0
o 20 64 0 34 10.0 9.765625E-5 0 -1
o 14 64 0 35 5.0 0.05 1 -1
This method does have some drawbacks. One you have to have a 2 pole power supply for the op amp. 2 you have to do some carefull tuning on the voltage source to the op amp or else you get nothing.

However you do get some bonuses. One is that you can voltage control your PWM. The other is that you only need to use one of the 555's resistors to control it's frequency. As well your output voltages can be conrolled by the op amps power rails.

I also want to give the caveat that I have never tried this with real parts, it's just a design similar to anlog synth PWM systems, where you use a sawtooth wave and a comparator.

Delta-Wye
Sep 29, 2005

Snaily posted:

By the way, is there a trick to knowing what all the xN-series diodes and transistors do, or is it a matter of learning specs by heart? I am going to need some kind of MOSFET to drive my stacks of LEDs, but finding a suitable one that is cheap is like finding the proverbial needle...

In my experience it is mostly just learning ones over time - as you get familiar you know the standard components for each job. Assuming your LED stack isn't insane, a standard 2n7000 or 2n7001 should be fine. If I ever have a situation where I need weird specs (high Id current, unusually high Vce, whatever) it's a matter of going to Mouser and typing in the specs and seeing what comes out.

EDIT: One of my professors was regaling us with stories of the good ol' days and apparently she used to have books of datasheets in the bathroom that she would read while on the can. When she needed a particular component she would have to go back through the datasheets she's read by memory and see if she could remember any useful components. Us dang kids and our internets! :argh:

Delta-Wye fucked around with this message at 00:41 on Jan 28, 2009

Snaily
Mar 5, 2006
Sluggish. Wee!
This is the best thread. Thanks.

Happiness Commando
Feb 1, 2002
$$ joy at gunpoint $$

(crossposted from the quick question thread)



It's just the guts of a Wagner heat gun which seems to use a standard type nichrome wire heating element, max power draw 1500 W. I want more control over the temperature in the low range than the panel will give me. My guess is that heat and fan are each just controlled by current limiting, so I need a way to limit current at like 200 - 300 watts (a guess pulled out of my rear end).

Potentiometers are out for not being rated at high enough power levels, and 225 W power resistors are $20 each at digikey, too expensive for this project.

Is this doable at all?

Mill Town
Apr 17, 2006

IfIWereARichMan posted:

(crossposted from the quick question thread)



It's just the guts of a Wagner heat gun which seems to use a standard type nichrome wire heating element, max power draw 1500 W. I want more control over the temperature in the low range than the panel will give me. My guess is that heat and fan are each just controlled by current limiting, so I need a way to limit current at like 200 - 300 watts (a guess pulled out of my rear end).

Potentiometers are out for not being rated at high enough power levels, and 225 W power resistors are $20 each at digikey, too expensive for this project.

Is this doable at all?

Resistors are absolutely NOT the right way to limit current in an application like this. Even if they were affordable or practical, the excess power bot being consumed by the heating element would be dumped by the resistor, generating... wait for it... heat. You might as well just stick part of the heating element outside the gun.

The resistive element is probably just controlled by a simple dimming-type circuit, with feedback from a thermistor to regulate the temperature. You could replace the controller with your own microcontroller, but that's kind of an involved project and probably would cost more than $20 overall.

Actually, maybe you could put another resistor in series with the thermistor to make the controller think the output temperature is higher than it is. That's assuming there really is a thermistor in a feedback loop.
Edit: wait, I think I got this backwards. We want the controller to think it's 350 degrees when it's actually 200, or whatever, which means we need to lower the resistance. So you'd need to put a resistor in parallel with the thermistor. Start with a very high value resistor and work your way down until you get the temperature you want.

If there isn't a thermistor, and it's a straight-up dimming circuit, you could probably remove all the electronic guts and just replace it with a light dimmer.

Mill Town fucked around with this message at 22:53 on Jan 28, 2009

Happiness Commando
Feb 1, 2002
$$ joy at gunpoint $$

I put a 600 watt rotary dimmer inline on the neutral lead from the power cord, allowing me to keep the control panel attached. And it worked without starting any fires! This was the major hurdle for this project, although unfortunately, the dimmer also slows down the fan, which has made it difficult to pin down the range of acceptable positions for the rotary knob.

I think the next step is to rip out the fan and replace it with another one that is powered independently.

Mill Town
Apr 17, 2006

IfIWereARichMan posted:

I put a 600 watt rotary dimmer inline on the neutral lead from the power cord, allowing me to keep the control panel attached. And it worked without starting any fires! This was the major hurdle for this project, although unfortunately, the dimmer also slows down the fan, which has made it difficult to pin down the range of acceptable positions for the rotary knob.

I think the next step is to rip out the fan and replace it with another one that is powered independently.

My advice would be to put the dimmer knob in the body of the heat gun, attached directly to the heating element. Then hook the power cord up to the element/dimmer combo in parallel with the fan. Get rid of the existing circuit board entirely.

Edit: Although, since it's drawing 1500W max, you should get a bigger dimmer.

babyeatingpsychopath
Oct 28, 2000
Forum Veteran


Mill Town posted:

My advice would be to put the dimmer knob in the body of the heat gun, attached directly to the heating element. Then hook the power cord up to the element/dimmer combo in parallel with the fan. Get rid of the existing circuit board entirely.

Edit: Although, since it's drawing 1500W max, you should get a bigger dimmer.

And put the new, bigger dimmer in the line side, not neutral. When your 600W burns out, your board and whatnot will be energized while you're attempting to figure out what went wrong.

Hillridge
Aug 3, 2004

WWheeeeeee!
That control pad leads me to believe that your heat gun is controlled by some type of micro. It's hard to say without a better look at it, but my guess is that it's controlling the heat via some type of PWM. You may get lucky and be able to adjust this, but more likely you'll have to resort to that dimmer (put it on the line side as mentioned).

Zaxxon
Feb 14, 2004

Wir Tanzen Mekanik

Snaily posted:

This is the best thread. Thanks.

did that circuit sim stuff help at all?

Snaily
Mar 5, 2006
Sluggish. Wee!

Zaxxon posted:

did that circuit sim stuff help at all?

Mine or yours? Mine helped my design immensely. Yours introduced me to an awesome tool :shobon: and helped further my understanding of the 555, for which I'm grateful.

Edit: This is what I'm going with:

code:
$ 1 5.0E-6 7.010541234668786 50 5.0 43
165 144 128 208 128 2 5.0
w 144 224 144 256 0
w 272 160 272 96 0
w 272 96 208 96 0
R 208 64 192 64 0 0 40.0 5.0 0.0 0.0 0.5
w 208 96 208 64 0
c 288 288 288 160 0 1.0E-7 -1.6666666666666794
w 288 160 272 160 0
w 144 256 144 320 0
w 144 320 336 320 0
w 208 352 336 352 0
c 336 320 336 352 0 1.0E-7 3.318420245531474
g 208 352 208 368 0
174 336 224 400 192 0 100000.0 0.0743 Resistance
w 272 192 368 192 0
w 336 320 400 320 0
d 336 320 336 224 1 0.1
d 400 224 400 320 1 0.1
w 208 288 288 288 0
c 208 288 208 352 0 1.0E-8 3.3333333333333206
O 368 192 464 192 0
o 20 64 0 34 5.0 9.765625E-5 0 -1
It seems simpler, somehow. At least for a nightlight.

Edit2: raise current speed and play with the Resistance slider to see how it works.

Snaily fucked around with this message at 21:35 on Jan 29, 2009

Hillridge
Aug 3, 2004

WWheeeeeee!
For any of you who need project enclosures, http://www.polycase.com/ is a great place for them. You can even get them customized and silk screened (probably not worth it unless you are making a lot of something and selling it).

Swap_File
Nov 24, 2004
WIN386.SWP
My car stereo has a stereo line level output, but it does not have any options to adjust the volume independently of the other speakers. I would like to add a pot between this line level output and my sub woofer amplifier in my trunk to adjust the volume.

I would like to install the control knob in one of the factory option holes:

(Mine are empty since I do not have traction control, fog lights, or heated seats)

So, I went to the junkyard and got a few buttons I could use:


I think I am going to use the big ASC push button for a door opener (mount the sender in my dash and just extend the button), but I could use either the seat heater or the dimmer knob for volume control.

My first idea was to turn the seat heater buttons into up and down buttons and use them with a simple digital pot, like this:
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/4848

But, I would highly prefer to use the analog dimmer knob. It would allow me to easily vary the speed of my adjustment. I hate stereos that just have button volume controls.

I figure that with the dimmer I will need to use something like this:
http://www.national.com/mpf/LM/LM1036.html

The dimmer itself contains a 50K pot, so I should be able to bypass the circuity inside the dimmer and use it as a "dumb" pot and hook it directly to the LM1036.

Does this sound like a good plan?

Edit: Someone was saying when using digital pots with audio signals a simple 12v regulator will not work to power it. They suggested using a +6v -6v supply, kinda like this:


That should not be needed with the LM1039 because it has its own internal voltage supply (seen in the datasheet), right?

Swap_File fucked around with this message at 03:58 on Jan 30, 2009

ANIME AKBAR
Jan 25, 2007

afu~
I'm a little fuzzy on car audio. Are so you're trying to adjust the signals right before they make it to the speaker (and therefore your circuitry needs to deliver a decent amount of power, correct)? Does your radio or whatever output device have stereo output?

clredwolf
Aug 12, 2006
Remeber to decouple the inputs and outputs out of that thing, are you're going to have some nasty ground loops (or worse!). Use the largest, bestest cap you can get away with and not have it sounds like crap (so Mica, Polystyrene, Polyester, or C0G ceramic...personally I think X5R or X7R ceramic is fine but someone's probably going to beat my rear end with a shoe for saying that). The schematic on the LM1036's page includes these (they are the .47uF caps on the inputs and outputs).

That LM1036 is pretty complex for what you want to do, but it will work. Make sure your dimmer control is the right resistance, you may have to tweak around with the slider resistor until you get it to where you want.

Another question is do you necessarily have to use a digital pot? It may be easier to have the dimmer pot control an analog circuit that controls the volume. An opamp or two (per channel) and you're in business.

Note also for the LM1036, you do NOT need a split, rail power supply. You might want one if you are amplifying it's output, but honestly I'd much rather bias the audio signal up and decouple the output for what you want to do (which you can use a single-rail opamp for).

Swap_File
Nov 24, 2004
WIN386.SWP

clredwolf posted:

Another question is do you necessarily have to use a digital pot? It may be easier to have the dimmer pot control an analog circuit that controls the volume. An opamp or two (per channel) and you're in business.

Would op-amps work with my linear taper dimmer pot?

I thought I needed a log taper pot to do that way.

I mean, I know a linear pot with an op-amp would physically work, but won't the scale be messed up?

clredwolf
Aug 12, 2006
You can do a linear to log conversion with opamps and a diode. Use the dimmer to pass a linear DC control voltage to the opamp, and get a DC logarithmic control voltage out. It may be a little wonky, and you probably will have to play around with it until it gets about right.

Then pass that CV to an opamp-based amplifier with adjustable gain controlled by the CV. You could also do it with a VGA (variable gain amplifier), which is just an IC (and pretty simple to use, just gotta find the right one), and set it up so your CV is between 0V and 1V (or 0V and 2V works too). There are also discrete component VCA (voltage controlled amplifier) schematics floating around. This one looks pretty simple , just another opamp and a few generic NPN transistors (probably 2222As or 3904s). This site has some too, the VCA2 there looks promising.

Opamps are fun!

clredwolf fucked around with this message at 18:35 on Jan 30, 2009

Swap_File
Nov 24, 2004
WIN386.SWP
Yeah, I saw a few op-amp based setups but they all seem like they will require quite a bit more board space and more work to assemble. The LM1036 just seems like a simple all in one solution for $2.75 (especially since it will work with my single 50K linear pot without any modifications).

It will be a bit before I start building anything, so I will continue to look around for options.

Is there really any downside to using the LM1036, other than it being overkill for what I am trying to do?

Swap_File fucked around with this message at 21:53 on Jan 31, 2009

clredwolf
Aug 12, 2006
None at all, datasheet even lists 12volt characteristics. That's probably the smallest way to do it outside of SMD parts. Good luck to ya!

Happiness Commando
Feb 1, 2002
$$ joy at gunpoint $$

Hillridge posted:

That control pad leads me to believe that your heat gun is controlled by some type of micro. It's hard to say without a better look at it, but my guess is that it's controlling the heat via some type of PWM. You may get lucky and be able to adjust this, but more likely you'll have to resort to that dimmer (put it on the line side as mentioned).


Yeah it totally didn't work like I thought it was going to. 1500W dimmers are $50 and variacs are even more. Is that the way I have to go?

ANIME AKBAR
Jan 25, 2007

afu~
keep in mind that you can get logarithmic pots as well. A linear op amp with a logarithmic pot will likely be cheaper than a linear pot with a log amp.

clredwolf posted:

Use the largest, bestest cap you can get away with and not have it sounds like crap (so Mica, Polystyrene, Polyester, or C0G ceramic...personally I think X5R or X7R ceramic is fine but someone's probably going to beat my rear end with a shoe for saying that).

for decoupling X7Rs are fine. For audio, anything better is overkill (for power amplification, a low ESR electrolytic family like HE would probably be fine). Stuff like C0G and polystyrene are for very low noise/high frequency systems.

ANIME AKBAR fucked around with this message at 13:51 on Feb 2, 2009

ANIME AKBAR
Jan 25, 2007

afu~

IfIWereARichMan posted:

Yeah it totally didn't work like I thought it was going to. 1500W dimmers are $50 and variacs are even more. Is that the way I have to go?

Well you could build your own dimmer with a triac and some simple external circuitry. Can't say how expensive the triac would be though.

Mill Town
Apr 17, 2006

ANIME AKBAR posted:

Well you could build your own dimmer with a triac and some simple external circuitry. Can't say how expensive the triac would be though.

The existing circuitry probably has a triac in it already so that's not a big deal.

clredwolf
Aug 12, 2006
Even still triacs are cheap as dirt for what they do.

4 bux, and it's still overkill

Swap_File
Nov 24, 2004
WIN386.SWP

ANIME AKBAR posted:

keep in mind that you can get logarithmic pots as well. A linear op amp with a logarithmic pot will likely be cheaper than a linear pot with a log amp.

It would be nice if BMW had used a log pot, but I don't think they have any other knob controls that fit my dash other than the interior light dimmer. So I am kinda stuck with a 50K linear pot.

Cost really isn't a concern, it looks like the whole project will come under the minimum digikey order amount no matter what route I take.

I did notice a lot of people on diyaudio.com complaining about the LM1036's relatively poor SNR (80db). But according to my car stereo's manual, it only has a SNR of 80db to begin with. I am not sure the LM1036 will make things any worse.

I did run across the TDA1524A which is similar to the LM1036, but it allows a higher input voltage (so I wouldn't need to attenuate the input like I would with the LM1036). But people complain about the TDA1524A too.

http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=568-1131-5-ND

I might just order them both and see what happens. As long as I don't hear a hum, it will be OK by me, this is only for subwoofers.

ANIME AKBAR
Jan 25, 2007

afu~
SNR isn't particularly relevant in a car audio system for obvious reasons... 80db should be fine.

In general audio hobbyists will find a way to complain about anything for any reason.

Swap_File
Nov 24, 2004
WIN386.SWP
Yeah, that's what I figured. I am going to stop spergin' out about it and just get the two cheap and simple chips and try it.

But I do think I found the "proper" way to do it (If anyone is interested):
http://home.comcast.net/~dreite/preamp/volume_control.htm

Like clredwolf suggested earlier, it uses a op-amp with a diode to convert a linear pot to log scale, and then uses a VCA. It looks like it was being used as a microphone amp, but it should work for anything with slight modifications.

the wizards beard
Apr 15, 2007
Reppin

4 LIFE 4 REAL
Most op-amps use the same pin-outs, you can swap one out if you think the cheaper chips aren't up to scratch.

FuzzyBuddha
Dec 7, 2003

Hillridge posted:

I use this chip all the time in designs. What voltage level are the signals from the Garmin? For reference, my applications have 3.3V signals coming into the Maxim part.

You mentioned the cap between 2 and 16. That's just a cap from V+ to Vcc. I usually have V+ (and V-) through a 0.1uF 50V cap to ground. Here's my typical setup:

I apologize for getting back to this so late. According to what I was able to get from an Arduino forum, and from my admittedly cheapo meter, it's around 12v.

I think the part that gets me most is that I've breadboarded it, and it's worked great, most of the time, but it's not consistent. Makes me wonder if I got a shoddy breadboard that's having trouble with its contact points.

I have to admit, electronics is not my forte.

Hillridge
Aug 3, 2004

WWheeeeeee!
Oops, I was thinking backwards. I should have asked what levels your micro wanted.

One thing to watch out for, especially this time of year, is ESD. Always ground yourself before handling ICs. It's easier than you think to kill a chip with a static shock.


----------------------------------
My question:

Do any of you know of a Hi-Speed (2.0) USB repeater chip? Essentially a 1 port USB hub. I'm guessing that even if I find one, a hub IC will still be cheaper due to their popularity.

Delta-Wye
Sep 29, 2005

Hillridge posted:

Oops, I was thinking backwards. I should have asked what levels your micro wanted.

One thing to watch out for, especially this time of year, is ESD. Always ground yourself before handling ICs. It's easier than you think to kill a chip with a static shock.


This is especially true in Fairbanks this time of year. :(

FuzzyBuddha
Dec 7, 2003

Delta-Wye posted:

This is especially true in Fairbanks this time of year. :(

So very true. It's dry and 30 below today. :P

Anyway, I believe the Arduino is expecting a 5V TTL signal, not my ~12V RS232. That's why I tried this circuit. It was the simplest one I could find, and with my solder skills simple is best. :)

BattleMaster
Aug 14, 2000

FuzzyBuddha posted:

So very true. It's dry and 30 below today. :P

Anyway, I believe the Arduino is expecting a 5V TTL signal, not my ~12V RS232. That's why I tried this circuit. It was the simplest one I could find, and with my solder skills simple is best. :)

RS232 uses a negative voltage that can be as low as -15v to represent a logic 1, and a positive voltage as high as +15v to represent a 0. The Atmega microcontroller on the Arduino board expects ~0v for logic 0 and ~5v for logic 1. The MAX232 part converts voltages between TTL levels and RS232 levels.

Anyways, let's get back to your problem. Are you using a MAX232 or a MAX232A? The MAX232 requires 1uF capacitors, the MAX232A requires 0.1uF capacitors. If you accidentally bought a MAX232A and followed that schematic, it may not work.

Hillridge posted:

Do any of you know of a Hi-Speed (2.0) USB repeater chip? Essentially a 1 port USB hub. I'm guessing that even if I find one, a hub IC will still be cheaper due to their popularity.

I am looking at Digikey right now and there are 4-port USB 2.0 HUB ICs (ISP1520, in stock too!) for as low as $7 CDN in small quantities. I don't think you can get much cheaper than that.

BattleMaster fucked around with this message at 18:56 on Feb 5, 2009

Hillridge
Aug 3, 2004

WWheeeeeee!
Yeah, I'm aware of that one, but I forgot to mentioned it. The problem with ISP1520 is that it looks like the manufacturer is discontinuing it. This is for a commercial design, so that makes it a no-go :(.

FuzzyBuddha
Dec 7, 2003

BattleMaster posted:

Anyways, let's get back to your problem. Are you using a MAX232 or a MAX232A? The MAX232 requires 1uF capacitors, the MAX232A requires 0.1uF capacitors. If you accidentally bought a MAX232A and followed that schematic, it may not work.

Hmmm, looking at the chip it says: MAX232ECN.

BattleMaster
Aug 14, 2000

FuzzyBuddha posted:

Hmmm, looking at the chip it says: MAX232ECN.

The MAX232E needs the same 1uF capacitors as the MAX232, so I don't think that's the problem. That schematic you linked has a 10uF bypass capacitor while the MAX232 datasheet recommends a 1uF capacitor for that, as well. I don't know if that would make a massive difference but try replacing it.

If that doesn't work and everything else seems fine, try replacing the MAX232 itself if you have a spare. I guess it's possible that you busted it by playing around with the capacitor values.

Adbot
ADBOT LOVES YOU

armchair
May 19, 2005

creak... creak... creak...
Quick RS-232 thing: unless you're going to be running at high speeds or whatever it's easier to just use a max233 instead of a max232. You just have to short out a couple of the pins, you don't need the 1uF caps all over the place. Makes it a little bit easier to wire up and saves you some scrounging around.

I think the advantages of the max232 are just that it is cheaper and the external caps give some flexibility.

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply