Well happy new year from the UK... I might see if the wife has regained conciousness

@Marla whoa. xD

you do it every day when driving, or even playing ball

sounds like I should practice some more?

@Shalvenay think about what you do when deciding (calculating) how to slow down, how to avoid collisions, etc. Our minds do calculus and differential equations without conscious though

@Marla ah :)

@W5VO my father (age 89) just gave me his Kenwood TS940S transceiver. I may have to get a license again. Haven had one since 1975, WN0ZED (and you probably recognize it was a novice)

@Marla wow. I think I might sit for the codeless Tech as a graduation gift to myself

@Shalvenay I just took the "practice" test for technician and for general, passed the tech test, failed the general test. Just because I didn't know the rules. The technical part is easy

@Marla yeah, the technical part is not hard

I know people who have passed Tech without studying

morse code not required now in USA

and radio rules aren't too hard as well

I can still after 40 years still do about 10 WPM of morse code

@Marla yeah, I'll have to try learning Morse again sometime though

@W5VO I missed on the IRLP and ITU regions, and APRS packet questions

among similar others

because you're not just memorizing the privileges that you have in total, but also what is the exact part of the band that is Extra only, or which part does an Advanced class have as an added privilege?

@Marla I've used one of those, it's a really nice operating radio

ah, sigh, we do need to know those, but any responsible ham will look it up before transmitting

@Marla that's my point.... you always have a band plan

I have several SW receivers, but the Kenwood beats them all, and it is 25 years old

I actually began my re-birth by my interest in antennas.

well, that was a top of the line radio in its day

worked for Motorola in VHF transmitter design

I think I mentioned it before, we replaced Mocom 70 with Mitrek

I have a Yaesu FT-980 that has spent time on the healing bench

I've used some of the newer radios, and they (can) make it harder to get stuff done

The Kenwood has a feature I never knew of. Able to change selectivity of IF above and below

oh yes, that's nice during contests

my dad found out I was experimenting in SDR and offered his rig to encourage me back into Ham

I want to build a SDR, I started on one but I ran out of free time

My fantasy is something like this: hackaday.io/project/2507-ethernet-to-radio-adapter

I built the antenna analyzer by K6BEZ. That really helps when doing 100 kHz to 2 GHz

decoupling the operator from the radio in a way that hasn't been done yet

actually only is good to 40 Mhz, but you knew what I meant

sure, I've used the MFJ one and I like it pretty well

oh, just looked up your link. I will read in depth later

More ramblings of a crazy person

what ? what ? I thought we all were

I'm actually getting a dev board to work on it a bit more - changing the main MCU

but still haven't gotten to work on the meat of it - the radio

@W5VO I am not a member of the Amateur SE site (beta), but I notice that the main drive on that SE is antenna design. I guess now that you can't really design a better rig on your own (cost effective) it makes sense. So I like that since my interest in is antenna design

I left Ham radio just as SSTV began

(slow scan televison) for those not knowing

also, am surprised that SSTV is still of interest. @W5VO , do you know why SSTV might still be relevant ?

Noobie question : I once had a coil that I loved - - - It's current was was below and above. I asked on the Exchange - - They said I was insane - - - so I'd settle for an energy exchange.

Come on, I expect immediate results. This is EE SE, you guys are supposed to know all the answers, I don't have a schematic or data sheet, you should be able to answer my question

@Marla You can use your sound card to send pictures - kinda fun

low equipment cost, doesn't hurt to listen to

@W5VO tell me a little tiny how a sound card contains pictures.

I guess memory can contain anything

oh, did you mean sound card regarding SSTV ?

:p you can use the sound card in your computer to send and receive SSTV

I was tickled with an App on my Android. Tuned in morse code, the App uses microphone as input, prints the code on my phone.

I made similar on my Coommodore 64 back in the 80's

@W5VO - also, back in the 70's , I was novice, and could only find crystal for 3.4 or something, can't remember the exact frequency. Old trick told to me, use lead pencil on crystal to increase frequency, it brought me into band. The things we did

So I was able to get on 80 meters

I haven't found anything relevant as regards using lead pencil to increase frequency of crystal. I can't remember where I heard that(yikes 40 years ago). But as I swiped the pencil across crystal, the frequency did increase.

FT243 holders

Novice by rule of FCC had to use crystal

Hope everyone had a good NYE!

in Australia, it is going well, here in Denver Colorado USA , still have 4 hours to go

Ahh yes forgot it was still in progress elsewhere!

the world turns

but salutations are accepted in any event

I saw the fireworks display from Sydney, video. spectacular

Happy New Year!

@tabchas same to you and have a hat

Could someone help me please? I don't understand what is going on here in my circuit. I am getting some weird oscillations whenever I press my force sensors labeled (Rs0, Rs1, Rs2, Rs3). When nothing is pressed, I get a stable clean signal. But right when I press the sensor, I get oscillations. I have added a 220pF cap which acts as a low pass filter, but I don't understand why its not working when I press the sensor.

The transistors are to select which force sensor is active. Please note that these sensors are high impedance (>5 M ohm when unpressed, and ~50k ohm when fully pressed)

Does it depend on the value of the cap I chose since the sensor's resistance changes as it is pressed?

what kind of oscillations? what frequency?

does it happen if you touch it very lightly (e.g. you wouldn't expect a significant resistance change)?

@W5VO Here's a screenshot of my oscilloscope.

@W5VO wait i'm trying to understand something first. My Rf is actually a digital pot, so if I vary the resistance, my low pass filter changes its cutoff frequency right?

@W5VO since the cutoff frequency = 1 / (2*piRC)

so where I put my 220pF cap at, my filter changes if I change the digital pot since it varies the resistance?

It will also change the DC response as well

at least at first glance that's what I see

@W5VO yes that was the original intention. I wanted a way to control the range and resolution of the force sensors. Using a digital pot in this configuration does this.

^without any capacitor, I see this

and my Rf is 130K ohms

If I vary Rf to 800K ohms, I see this:

I don't understand why I am getting these different oscillations and different responses. The expected output is a stable ~50mV when no sensor is pressed.

Is the timebase the same between the two?

I'd guess that all changing Rf is doing is changing the frequency of oscillation

@W5VO heres better shots

Rf = 800K (No Cap):

Rf= 130K (No Cap):

Sorry, scratch that one sec

@W5VO here are the correct ones:

Rf = 800K (No Cap):

Rf = 130K (No Cap):

Hey, @Kortuk. Is it already 2016 on your end?

@NickAlexeev normally it'd be in a few

@tabchas what's the "no load" resistance again? 5MΩ?

@W5VO >5 M ohm

but it doesn't actually go open circuit, right?

@W5VO no it does not

Your output is very close to ground, and I don't think it should be

@W5VO Yes the first op amp has a bias of 500mV, so Vout1 will always be >500mV, but the second op amp subtracts the offset and amplifies to 0-5V. So it should be about ~50mV

@W5VO I added a 0.1uF ceramic cap instead of the 220pF in my circuit

Here's the output with an unloaded sensor.

@W5VO Heres with my pressing on the sensor:

I think it's your second amplifier

wait...

nvm

I'd say ditch the second amplifier - you aren't gaining that much resolution

for something that's really easy to manage in software

especially if you're putting in a variable feedback R

but don't I effectively cut the range since I am using a bias voltage?

sure. how much range are you losing? Have you calculated that?

well depending on the bias voltage, in my case I am using 500mV so my output can only be 500mV to 5V

but I can use as little of a bias voltage as I want, right?

so if I get the right pair of resistors, I can voltage divide something very small (50mV?)

@W5VO The manufacturer had recommended it, ^thats where I got it from

and I'm saying that it gives you marginal benefit for the joy of playing with fire at 0V

@W5VO what do you mean by playing with fire at 0V?

@W5VO just trying to understand why you think the second op amp could be a cause of the problems I'm seeing (for my learning :) )

I'm messing around with some LTSpice sims for fun... My initial reaction was based on seeing oscillations on the output of the second op-amp... and then I noticed that I neglected to connect VDD on the second amp

your amplifier response changes near 0V because you don't have a negative voltage rail

and if you're saying you get a good response off the first op-amp

@W5VO thats really interesting why you came to that conclusion

@W5VO the recommended circuit from the manufacturer uses a negative voltage rail

could that be why they recommend using the above circuit?

@tabchas it's a variant

@W5VO ah ok, a lot of people told me to stay away from negative voltages tho since most microcontrollers don't operate using negative voltages

You actually won't see any benefit from the level-shifting stage (other than buffering) though since the output of the first stage clips at +5V

@tabchas reasonable advice, though in this case the negative voltage isn't really necessary

the reason i think they advise it is because its the fastest way (less components) to get the full scale of 0-5V

whereas to get similar range (0-5V) would require two op amps with a positive voltage circuit

@W5VO ^

Seems to be a lot of extra circuitry for minimal improvement

though I suppose it's not a whole lot more if you're starting with a dual op-amp

For this circuit, I would rather get a better ADC than deal with the second op-amp

if the resolution was really needed

@W5VO right. Well its worth a try :). Need to just solder one more wire and I'll get the response of just the first op amp

@W5VO Almost identical response

The beginning and end are when the sensor is unloaded

and the jump in signal is when I press on the sensor

Rf is ~150K ohm

and the cap is 0.1 uF (without the cap, I get the biggest oscillations ever)

well, I'm running out of steam here - probably going to call it a night

@tabchas does it oscillate without the cap if you don't touch it?

@W5VO yes, here is the shot:

@W5VO it seems it is filtering the signal properly when the sensor is unloaded. When the sensor is pressed it is doing similar oscillations.

@W5VO just not sure where to do more filtering to fix this :(

@tabchas try pressing it with a non-conductive stick and compare results

do you have a bypass capacitor on the op-amp power supply?

past that, I'm out of guesses

@W5VO tried non-conductive stick

@W5VO same oscillations

@W5VO there is a bypass cap on the op amp power supply

@W5VO no worries, i really appreciate the help

did we eliminate the lights last time?

@W5VO yea lights are eliminated

anyways, happy new year, and good luck

@W5VO thanks and you too!

@NickAlexeev For 1 hours and 12 minutes now. :)

Happy new year all. :)

@Kortuk Have a good one!

@PeterJ And you also!

Happy 2016!

Happy new arbitrary point in earth orbit day

best wishes to all and Mr Westvleteren aka @Kortuk

We should really organize a beer fest for all EE.SE members near Eindhoven.

Where is this clabacchio character lately anyway, he was the driving force for the local beer fest

@jippie I really enjoyed the modern art museum there. :)

@PlasmaHH Do you mean "good morning"?

Hi everyone, could someone help me understand how to read the sn754410 datasheet (ti.com/lit/ds/symlink/sn754410.pdf) At the absolute maximum rating section (7.1) it mentions Ip (Peak output current) and Io (Continous output current). I want to understand if the power source that Vcc2 is connected to provide a current of 1A would also mean that Io = 1A? Does the current at Vcc2 affect the peak output current and/or continuous output current?

@user2899444 not quite.

THe continuous output (assuming adequate heatsinking and operating conditions...) is 1A as you stated BUT a 1A PSU doesn't mean the peak output of this chip will be 1Amp

you should have some decoupling capacitors not only for ... HF decoupling but tank capacitance for a Hbridge

this will provide a high-ish peak current pulse capability

@JonRB Okay. I want to better understand how the current value at the output pins are determined and the max current this IC could take.

This chip's output is governed by its supply (psu & capacitance) but also the load.

Okay, I see on the datasheet it recommends a 0.1uF capacitor at Vcc1 and Vcc2, is this what you're talking about when you mention decoupling capacitors?

if you were to short the output and have a 1A supply + 1F of capacitance... quite a large amount of current could flow

I really want to have a 0.5~0.7A current going through the electromagnets.

whats the load characteristic of this "electromagnet"

how much resistance?

how much inductance ?

It sounds like you will need to regulate the load curent to 0.5-->0.7A via PWM and current feedback

They're hand-wound electromagnets. The resistance is small like 5 Ohms.

ok 5Ohm, what voltage PSU?

I'm not sure about inductance and getting those values.

I was originally using a 3.0V that was giving 0.85A current even though I know that's well over the recommended minimum. I actually want to use a 12V power supply.

3.0V because I thought the 1A would mess up the chip.

the chip... could be fine you would need to manage it

but yes... PWM... 12V and 25% duty

you could use a 555timer for this and run it open-loop (reasonable if the voltage will stay at 12 and the load won't change)

the switching freqency will depend on hte inductance of your electromag...

300uH ~ 2kHz for instance

OR you close the current loop.

hysteresis control would be good for this.

why use a Hbridge chip tho when a simple FET and diode would do for the powerstage?

I'm not too sure, but here's what I think I understand: In order to have a 0.5A output (or my desired current at the output) I need to use PWM at vcc2, best achieved with a 555, yeah?

Oh, I tried doing a BJT h-bridge and one of my transistors overheated. Then I was told that it's hard to find proper h-bridge schematics... I don't really know what I need to do.

I'm a beginner.

"proper H-bridge schematics" are a dime a dozen. The issue is there are a number of REALLY BAD topologies out there that "might" work for a very specific case...

Also you don't need a H-bridge UNLESS you need positive AND negative current.

I need it to go in both directions, yeah.

you have an electromag... it doesn't care about the polarity of the current, just that there are amp-turns

why do you need bidirectional current for an electromag?

I need it to repel/attract and also stop.

I'm trying to do something like this: www.e-alexander.net/magic_slippers.php

ok so you already have an magnetic flux source available. changing the polarity will do something. so yes Hbridge

IF you do intend to change direction you need some means to ... change direcvtion :)

some additional logic. BUt yes you need PWM to regulate the current in the load and you will need ~ 25% duty to keep the current around 0.6A

I'm using the Arduino to change the logic/direction.

How do you determine the PWM and current?

( the control and what the Arduino needs todo to ensure correct direction aside...) As I mentioned a fagpacket calc is 25%

Is the PWM at Vcc2? or rather the output pins?

the frequency is the important part... THIS is going to be driven by the load inductance... The higher it is the lower the freq can be. If the load inductance is too low and thus the switching freq would need to go up (to balance V = Ldi/dt) then well... you need to check thermals and capability

The PWM would be at 1-4A to then appear at 1-4Y

- personally I would potentially only PWM one pin at a time.

and HI to the equivelent other

significantly cut down on the switching losses. regulate the current considerably better (via the zero volt loop)

I was letting that sink in... Let me see if I get it, I put the PWM at the input pins (1A, 2A, 3A, 4A) and the pwm/current at the input pins will be seen at the output pins 1Y-4Y.

yup

If let's say 1A is low and 2A is high (arduino setting), the circuit should PWM at 2A and do no PWM at 1A?

The duty cycle at pin 2A would then determine the current at the output 2Y?

IF you do it via hysteresis control, yes

I would recomend hystersis control for something like this :)

but oyu need a suitable current sense means

you could use the load itself via a diffAmp

as you say it is 5Ohms

I'll look up hystersis control.

I could get arduino to do the PWM at those pins, right?

instead of a 555

easily. I wasn't aware of what you had.

BUT you need a current sense - measure the voltage of the load BUT you will need +-12V as you will get -amps

and then bias the output so it can go into a 0-5V ADC

Oh, I think I get what you mean. You mean in my arduino code or whatever I need to inspect the voltage at pins 1Y through 4Y.

Wait, is this in the case of an AC supply? What if I'm using a 12V DC?

nothing todo with how you generate your supply (at some point you will need DC to this chip).

but yes the Arduino will need to know the voltage of the load to infer load current to facilitate current regulation

I'm using a DC supply.

thats fine

Okay at arduino part.

Is there a way I could determine inductance of my electromagnets since it seems important in this case?

Or does it matter as long as the arduino could detect the voltage at the output pins?

its still important to roughly know because say the inductance is 1nH ... PWM control will be ... interesting

you can either use an LCR meter (meh...) do some calcs (core permubility, number of turns...) or if you have an osciloscope... pulse it and measure hte current rise

What to do with the current rise value?

I'm watching a video to understand that part.

@JonRB, if I don't use PWM on any of the input pins and connect Vcc2 to a 12V DC power supply, what would the current be at my input (A) and output (Y) pins?

Well since the load is essentially an inductor and a resistor... at t=0 (ie once EN is sent to the chip) the current would be 0A & it would increase, exponentially, until V/R = 12/5 = 2.4A

This is why I stated you need PWM to regulate the current to your desired 0.6A.