Operational Transconductance Amplifier - OTA LM13700 - Simply Put

You can join me on Discord as well! -- / discord

[@simplyput2796]

I screwed up a bit on the buffer part at the end: It's not a voltage follower, it's a darlington emitter follower (they're very similar but not the same), and the emitter resistor (the one near the output) is supposed to be connected to the negative supply, not to circuit ground. The explanation is actually mostly right though! I will do a video explaining the darlington emitter follower, but for now just kind of skip past the buffer explanation here.

[@plazmotech5969]

This video was fantastically explained, thank you!

[@aaronjayden370]

i dont mean to be so offtopic but does anyone know a method to log back into an Instagram account?? I was stupid lost the login password. I love any tricks you can give me!

[@jaxsonariel2151]

@Aaron Jayden instablaster =)

[@aaronjayden370]

@Jaxson Ariel Thanks for your reply. I got to the site through google and I'm waiting for the hacking stuff atm. I see it takes a while so I will get back to you later with my results.

[@aaronjayden370]

@Jaxson Ariel it worked and I finally got access to my account again. Im so happy! Thank you so much, you saved my account :D

[@jakescustomshop]

Incredible. 4 Years later it is still impossible to find tutorials on these things. Thanks!

[@DrScientistSounds]

You have quickly become my favourite electronics channel on KZbin, thank you very much for these videos!

[@georgisorev3858]

+1

[@knowledgehunter4871]

+1 😘

[@OlayonChingon]

+1🤓

[@paguerchanel]

I wish the University where I study allowed me to put youtube videos in the references so I cite this, is so well expleained! Greetings from Mexico!

[@andyvictoria9534]

Very nice explanation, been confused about OTA vs Op Amp, you got it so clear, thanks!

[@1973jdmc]

You are an ANGEL- I have an exam coming up and until your explanation I was totally lost- YOU ARE OFFICIALLY A SUPER- HERO

[@arthurharrison1345]

Thanks for this great video! I love using the dual OTAs for VCFs and VCAs. The fact that there are two OTAs in the LM13700, and that they are inherently matched in parameters and thermally coupled makes then extremely versatile for precision applications. Many years ago, my colleagues and I submitted a paper to Texas Instruments which describes applications that capitalize on the fact that each chip has a matched pair, but by that time, the part was not so popular, and they didn't add these applications to their data sheet.

[@ozerik]

Your explanation was NOT terrible! Thanks! I've designed a few circuits with this chip, and next one I do will involve less mystery and guesswork!

[@thenagus]

Great explanation! I've bounced off OTA explanations a bunch of times before while trying to understand synthesizer VCA circuits, but now it all makes complete sense.

[@DeusExAurum]

Thank you so much! OTAs look really complicated (especially when first looking through the datasheet) but your explanations are incredibly clear!

[@OPA111AM]

Thank you so much for putting time and effort into these videos, they are so helpful. Your explanation skills are excellent. I learn a lot of electronics through KZbin and do not understand why KZbin recommended you to me only now. Greatings from Germany!

[@garydirkse9900]

Terrific work explaining OTA's. They are so different from opamps. I need to watch your video more than once. I am so old that I remember the conductance unit was the mho.

[@sreeharibalachandran]

This video gave me a quick insight on OTA..Thankyou Very Much.

[@nmidu]

Thank dude!! I was study the datasheet of the LM13700 but i found this video and save my life. Big hug from argentina.

[@JacksonAudio]

You are an absolutely brilliant educator. I was studying OTA's and when I saw that you did a video on it I yelled, "YES!!!!!". Thank you so much for your investment in this industry!

[@tugbars4690]

This was extremely easy to understand. You are a really good teacher.

[@iNDREI_Ro]

Finally, I good electronics oriented channel. Thanks.

[@i.k.ishaan]

Thank you for this video. Keep it up. My concepts are more clear now!!

[@MoritzKlein0]

great explanation!

[@Spunkyinternall]

thank you, 4 years ago I scratched my head over it. This makes it way more clear how to handle/nurture the pins. (lost my focus every ones in a while during the video, I understand the jump cuts, (to make it as short as possible, but give some moments to breath/relax, (could be just me)). I think you would make a great teacher :).

[@mrquint15]

finally some understandable videos on OTAs. Thank you!

[@ptreth]

We used OTA's a lot in analog video signal processing, mostly as a "back porch clamp," but also anywhere you want to intentionally introduce slew rate limiting. We would forgo the output emitter followers and the linearizing diodes completely, and often feed the current output to a capacitor (referenced to ground). Bring the capacitor voltage to the inverting input and you get a cheap but effective sample and hold. Bias current on for sample; bias current off for hold. Once used CA3080's but the LM13700A is better. Regard from Paul, formerly of Grass Valley Group.

[@s.myklebost1991]

This is so great! The OTA definetly seems less scary now.

[@arianprins8347]

Thanks for the great explanation!

[@wuukaa9079]

Thank you so much omg! I really got what this lm13700 is about now :)

[@edk7825]

Your of the OTA tutorial is "on the spot". I like the way you use the whiteboard with the pre-drawn cirquit symbols. Very clever idea to keep scribbling to a minimum and maximise the quality of the semi-drawn cirquit.

[@parapos]

great explanation,as simple as usual,thank you!

[@alicewyan]

Thank you so much for this video!

[@fer_fdi]

Excellent, thank you!

[@linusog]

Thank you so much for the great explanation!

[@georgisorev3858]

Great before an exam session

[@yaboop]

4:30 what is transconductance? 8:43 OTA symbol explanation 10:16 what are the inputs? 12:25 iabc explanation 13:41 buffering the output 16:56 connecting the chip 18:22 gain control resistor 20:57 linearising diodes, inputs 26:05 quick summary 27:54 output, internal buffers 33:05 final summary

[@NitishKumar-hu4hh]

Thank you so much for this amazing video....

[@lodwar]

Wonderful Thank you!

[@nicolasgreccovairo]

Thanks man! very helpfull! kind regards from argentina!

[@codewizard58]

The application notes in the TI datasheet are very useful. For synth use VCO, VCF, VCA are all there.

[@hubercats]

This is a very helpful video. I'm trying to figure out how to interface with a hydrophone and an OTA seems like it might be just the ticket. Thank you for taking the time to make and share the video! - Jim

[@hubercats]

Hello “Simply Put” - I’m hoping you can help get me started with interfacing to a Murata 7BB-20-6LO piezoelectric element. The data sheet for this device lists a capacitance of 10 nF at 1 kHz and a resonant frequency of 6.3 kHz. Does this information suggest a particularly appropriate preamp approach? - Thanks again - Jim

[@ekremkorkmaz1]

i mean damn. well done brother. i dont like electronic but you make me love electronics and you

[@knowledgehunter4871]

Same here 🤗❤️

[@stelialala305]

Loved the video!

[@alfredomamani2269]

Thank you a lot for this explication. Definitely helped me :,)

[@jackwillis5446]

Thanks! Building a "Korg MS-20" filter from a diagram I found on the web. Calls for a LM13600 and a single 9-volt supply (idk). Had never seen this part before.

[@mariomionskowski6223]

your exponation was not terrible. Very good explaned, thank you.

[@hageir]

YOU RULE. Subscribed.

[@ivani3rr]

Great! Thanks dude!

[@mohamedaboelenein7727]

You man are awesome! ty!

[@naderhumood1199]

Terrific keep it up. Great vedio Sir.Thank you v much indeed.

[@g.d.8065]

One of the main reasons you might want to use an OTA is basically to be used as a voltage-controlled resistor (with the control voltage being applied to Ibias). For instance, the voltage driving Ibias can be varied to change the output gain, allowing it to effectively act as an amplifier with a voltage controlled volume knob, or as a (two-quadrant) voltage multiplier (same thing), though to have linear multiplication, you'll need an extra circuit to convert a larger voltage to work within the range of Ibias. In a similar way, it can be used to make a voltage controlled filter (controlling the output current gain being equivalent to changing the R in an RC filter). It's also used in some synthesizer VCO circuits (as they usually work by having an integrator that generates a ramp that resets at a certain threshold, with the charging speed of the ramp (and thus the frequency) being controlled by the current into the integrator. A triangle wave can also be generated by changing the direction of the current rather than discharging the capacitor all at once). There's also a pretty interesting component called a Norton op-amp (the LM3900 being an example), which reacts to the difference in current at its inputs rather than voltage. Examples: Thomas Henry's VCF-1 Electric Druid's page on multi-mode filters Achim Gratz (Stromeko) also has a good (work-in-progress) paper on OTAs that goes into more detail. The LM13700 datasheet also has a lot of interesting applications.

[@g.d.8065]

​@@Ormaaj You try driving a potentiometer at 5khz with a motor. ;p That's acceptable for the kinds of applications listed here anyway (filters and VCAs). You could also make a vactrol (photoresistor and LED pointing at each other in a sealed case), but the response time is relatively slow.

[@g.d.8065]

​@@Ormaaj sound-au also has a few articles on VCAs (one that compares different types for compression/limiting, and a couple others that include a Op-Amp+JFET-based limiter and a preamplifier with integrated VCA-based multi-channel volume control). w2aew also has a video on the Gilbert cell analog multiplier: kzbin.info/www/bejne/bZ_QnpVmpdaHi5I

[@ashwatson2142]

Thanks so much for this. I've been trying to figure out what exactly the voltage across that Iabc resistor ought to be

[@allinthefamiliy3122]

perfect thanks for explaining. Im looking for a cheap PGA and this might do the work

[@michaelalldritt7649]

Brilliant really appreciate your explanation. i am using the LM13700N from a very old circuit diagram. could not understand how the analogue signal was managed. thank you

[@Reverend11dMEOW]

BMC015 has me all a-gaggle

[@guillemdc]

Hi mate, thx for the awesome video. Would it be possible to control the amp bias (pins 1 and 16) with PWM? It'd be cool for synths and phasers.

[@superpie0000]

the ota changes gain based on load impedence, the current stays fixed based on inputs, but with fixed current comes diffrent voltage at diffrent load impedence. its Ohms law, so fire

[@jeremyjedynak]

Great video! Could an OTA be used to mimic the behavior of a thermistor where you want to observe overall circuit behavior for different programmatically input "thermistor" resistances rather than having to apply a thermal input to an actual thermistor?

[@TomLeg]

Siemens is the unit for conductance, which is the inverse of resistance. So S = 1/R. Since E = IR, i.e. R = E/I, then S = I/E, aka S = Amps / Volts

[@drOffset]

Great explanation. I'm hung up on the current idea. What is the voltage of the output current, is it the difference of the inputs but at a varying current? I need to watch this a few more times and compare with popular synth circuits. Thanks for the video!

[@simplyput2796]

It might make more sense to you when I go through an actual voltage amplifier using an OTA, but it's actually much simpler than you think. Current out = differential voltage in (times) transconductance. Transconductance is sort of like gain, except it's amps per volt instead of volts per volt, and your transconductance is set by the current going into the "Iabc" or "Ibias" pin (the one drawn going into the two circles). So you're going to have Iout=gm*Vin as the current going across the resistor connected directly to the OTA output. No matter the resistance, you'll have that current. So if you change the resistor, you change the voltage swing, because V=I*R. So for example, if your maximum current swing across that resistor is +/- 1 mA, and your resistor is 1k ohm, then your voltage will swing +/- 1 volt. If you change it to a 2k ohm resistor, your voltage will swing +/- 0.5 volt. So you set your maximum voltage range by the resistor value based on your maximum current range.

[@CandyGramForMongo_]

Buried reply to just talk to you. I really appreciate your videos! You are speaking right where I need to be. Do you have a bio somewhere? Do you have a public email address?

[@youssefdirani]

17:54 I thought of a Zener for clamping like an assurance of voltage not exceeding a value of say 40 mV but the voltage is too low. Anyway a circuitry of overvoltage protection maybe is good to think of. Like using a voltage comparator that will short the OTA inputs in case of voltage higher than 40 mV. Or hey, why don't we hook up an op amp inputs in parallel with OTA inputs so that the op amp regulates the input difference for us... IDK what I'm saying.

[@noobulon4334]

You can always put the zener in the voltage divider

[@aaroncollins1999]

where can I get a pair of those sweet headphones?

[@ranbymonkeys2384]

Can an op-amp be compared to the Heisenberg Uncertainty Principal seeing how there is a always a give take relationship?

[@simplyput2796]

That's clever, but it's more of a stretch than Mister Fantastic.

[@poetlorryit]

Any relation to the Comic Book Guy?

[@youssefdirani]

31:24 do you think the buffer BJTs are in amplification mode or in saturation mode ? And if they were in amplification mode would this impose a restriction on the base resistor to be low enough so that base current becomes low enough not to enter saturation mode ? I hope I'm clear.

[@youssefdirani]

AFAIK when in amplification mode the relation Ic = gain x Ib is preserved. While in saturation, not necessarily. When you say the the positive rail of the collector feeds the load I feel it's saturation. From another hand, in saturation, I think Vce is 0, which is not our case. Anyway...

[@youssefdirani]

I strongly believe we are in amplification mode, where base current is linked by the transistor gain to the collector current, and both currents are determined mostly by the source side and in a low degree by the load side. And Vce is last parameter that will be determined. And we might probably be limited by a max Vce for normal operation, and by consequence the transistor supply voltage must not be far away from the signal voltage.

[@simplyput2796]

Being a novice, I'm not 100% sure I understand all of what you've said, but if I understand the way the OTA and the Darlington emitter-follower buffer works, the idea is that the base current through the buffer transistors should indeed be as minimal as possible, so that it doesn't divert significant current away from the resistor between the OTA output and ground, thus not messing up the current->voltage conversion that resistor does. So yes, if I do understand, then we are in amplification mode and the resistor has to be chosen correctly to not let the voltage go too near the positive supply when the OTA is at maximum current output.

[@youssefdirani]

@@simplyput2796 thank you... You're great and I'm your student. One more question, do you think the resistor right below the emitter is better be significantly higher than the load resistance, like 5 or 10 times higher ? Or this is not really a constraint. I feel it's better be higher but it's not really a constraint since it's only about how much power will be consumed.

[@EricFontaineJazz]

I don't think you are quite correct at time 24:48 when you say it says that the voltage at the linearizing diode pin is going to be one diode drop above circuit ground, which you later say is halfway between positive and negative supply. Actually when TI's datasheet on page 11 says "As the input voltage has a DC level of 0 V, the Diode Bias input pins are 1 diode drop above 0 V, which is +0.7 V", the datasheet is referring to their typical application schematic (Figure 17) which has the + and - inputs connected to a 1k potentiometer where the center washer of that potentiometer is connected to an external ground. Note: the IC itself doesn't have a circuit ground. Looking closely at Figure 16, it seems the Diode Bias more specifically is going to tend to be one diode drop higher than the common voltage of the - and + input voltages. It could be that the - and + input voltages are centered around another voltage level other than 0V. If for instance the - and + input voltages were centered around +2V, then the Diode Bias input would be at +2.7V, I believe.

[@liamgilmour3941]

Would I be able to swap a CA3080E for an LM13700 for audio purposes?

[@simplyput2796]

I don't have enough experience to determine that. I would recommend comparing their datasheets side-by-side relative to your particular application.

[@liamgilmour3941]

@@simplyput2796 No worries dude thanks anyway, good explanation btw

[@annonymousname2.0]

I watched the entire video hoping we would see if he has a hole in the other side of his headphones.

[@garystone9177]

Nice job! That really cleared it up.... Now if you could do one for the LM3900, my life would be complete....

[@simplyput2796]

According to the TI datasheet, it's a norton operational amplifier, which apparently is transresistance? Fascinating. I shall do research. Time for another Mouser order, I guess!

[@garystone9177]

@@simplyput2796 - I am fascinated by that chip. I build modular music synthesizers for my hobby (ok, addiction) and found that one of the most interesting forefathers of synthesis (Serge Tcherpnin - en.wikipedia.org/wiki/Serge_Tcherepnin ) used the LM3900 EVERYWHERE in his designs. Most of the original designs were cloned (with Serge's permission) by a fellow named Ken Stone, and are currently being sold by Elby Designs. The schematics are available on Elby's website, and may prove interesting for you... www.elby-designs.com/contents/en-us/d77_Serge_Modules.html. The chip itself has proven to be very hard for me to understand, but it seems to be a real swiss-army knife.

[@garystone9177]

BTW -- Take a look at this schematic - probably Serge's crowning achievement. The universal slope generator. A single module that be used in more than a dozen ways. www.elby-designs.com/webtek/cgs/serge/cgs114/cgs114_dusg.html

[@simplyput2796]

In principle, something that is controlled by current should be incredibly flexible, because effectively that means it's controlled by current OR voltage, meaning you control it by varying the resistance or by varying the voltage, whereas something controlled by voltage you can only really vary the voltage. OTA + buffer = OpAmp, but you can't take the buffer back off the OpAmp to make an OTA. This sounds like quite a rabbit hole to dive down.

[@youssefdirani]

@@simplyput2796 op amp + FET = OTA Not sure but a guess

[@pyrotek45]

i wish i could find some better designs for vca's and vcf's without using ota's since they're sorta getting discontinued and harder to get. tired of seeing the lm13700 in every diy schematic tbh.

[@gkdresden]

I still don't know why I should use an OTA.

[@ff-qf1th]

4:25

[@ff-qf1th]

24:30

[@TrebleWing]

Or......... stay with me here......... I could just keep using op amps...... lol j/k I loved the explanation, but wow, I can see why these are not popular at all

[@Hungry_Hunter]

I think the use of the word Transconductance is incorrect since this is not true transconductance as defined as the ratio of change in anode current with respect to grid voltage in a vacuum tube. Maybe we need another word to describe this effect.

[@denebvegaaltair1146]

You spend way too much time explaining things that should be basic prerequisite knowledge

[@DrScientistSounds]

You have quickly become my favourite electronics channel on KZbin, thank you very much for these videos!