Author Topic: Voltage boosting an opamp  (Read 20411 times)

0 Members and 1 Guest are viewing this topic.

Offline YansiTopic starter

  • Super Contributor
  • ***
  • Posts: 3930
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Voltage boosting an opamp
« Reply #50 on: November 10, 2019, 04:47:01 pm »
It is not your issue to question what is or isn't viable for others. That kind of questioning is what indeed pisses me off. Either stay on topic and discuss the amplifier topology, or go away please.

This circuit is part of a much more complex project, where shaving $5 out of a BOM really counts.  Even if only 10 pieces of it would be made (I guess probably more will), $50 spared just on one amplifier is not an insignificant amount for a homegamer.

PCB area is a stupid argument anyway, as if space would be the constraint, I would probably not be going this route in the first place.

I have a lot more things to do, than just this amplifier that currently will not earn me any $, so please forgive me I still work on this project for more than 2 years.
« Last Edit: November 10, 2019, 05:19:10 pm by Yansi »
 

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15797
  • Country: fr
Re: Voltage boosting an opamp
« Reply #51 on: November 10, 2019, 05:23:08 pm »
Jeez, just relax. If you don't want other people's opinions, don't use a public forum. This forum is not your playground where you set your own rules. When you open a thread, all kinds of people are going to give various opinions, and that's actually what makes it interesting and fruitful. If you're not interested in one opinion, you can just ignore it. And again *I* did not revive this thread, it got active again, and I reacted to it (was not even on the forum when it first started.) It's every user's right, as long as they are not insulting anyone. I don't think I did. But you did with your "stupid" words.

Threads in a public forum are not just a private discussion. If you wanted it to be private, just use private messages. Many other people looking for a similar solution can be interested in reading various options and opinions. Without counter-arguments, some people after reading it may conclude that using HV opamps is stupid whatever the context - which is obviously silly. Threads are shared discussions, not just a private and free tech support tool.

As to the time it took, I don't doubt (or care, it's your business) you had many other things to do, but it just shows designing a discrete version was definitely not as simple as it first looked, so considering other options was definitely not stupid engineering-wise. So I consider my point (which was short enough not to require hours to read) to be relevant. Again, take it or leave it, but it could help others deciding what's the best option for their own project.
 

Offline YansiTopic starter

  • Super Contributor
  • ***
  • Posts: 3930
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Voltage boosting an opamp
« Reply #52 on: November 10, 2019, 05:42:50 pm »
Your solution is the same type, as if you would just say "buy a complete ready test instrument for the purpose", why making one. No thanks.

Now I politely ask you to leave this thread, as you did not contribute anything useful, rather just being provocative and nitpicking about what is or isn't expensive, which is not the topic this thread is about. Neither are you entitled to evaluate how much work something required to make. This is my thread and I make the rules.

Thank you for making mess out of this thread, no-one will be willing to read. Please leave. Bye.
 

Online magic

  • Super Contributor
  • ***
  • Posts: 7453
  • Country: pl
Re: Voltage boosting an opamp
« Reply #53 on: November 10, 2019, 07:46:28 pm »
Without counter-arguments, some people after reading it may conclude that using HV opamps is stupid whatever the context - which is obviously silly.
HV ICs is boutique stuff, only available at big distributors and pricey. I feel like people use them either because they are space/power constrained or because the knowledge how to whack together a simple analog circuit has largely been forgotten in the mainstream, due to the huge success of low voltage ICs which are indeed highly competitive both on performance and cost. I think I have seen examples of the latter.

Discrete still lives in some commercial power analog applications like audio. Here's one of the highest performance speaker amps out there. I can see a few ICs but they are in a sea of transistors, I think it's safe bet most of the high voltage signal path is discrete. Go tell those guys they should be using opamps :P
 

Online magic

  • Super Contributor
  • ***
  • Posts: 7453
  • Country: pl
Re: Voltage boosting an opamp
« Reply #54 on: November 11, 2019, 07:07:24 am »
And speaking of the actual circuit...

Just wondering, how could I get more stable biasing. Huh.  ???  What would be the dis/advantages of using 4 diodes vs. Vbe multiplier?
I'm not sure what bias instability you have, but if it decreases under heat, which is what it probably does, you should perhaps replace one pair of diodes with 680Ω resistors to match the tempco of the parallel branch.
 

Offline YansiTopic starter

  • Super Contributor
  • ***
  • Posts: 3930
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Voltage boosting an opamp
« Reply #55 on: November 11, 2019, 01:09:47 pm »
Not saying it was unstable, but it just got me thinking how or if it could be improved right away.

Your suggestion does make sense, that way the tempco of those two Vbe get compensated. However what about those two voltage controlled current sources? There are still those -2mV/K, i.e.  0.002V/100ohm = 20uA/K drift. Which if I understand correctly will make a change of 1mA in bias @50K temperature difference, which is not that nice.

Hopefully the calculation can be done that way - taking the tempco of the PN junction divided by the Re of the current source - considering constant voltage at the base. (decreased Vbe means increased Re voltage, hence increased current).
 

Online magic

  • Super Contributor
  • ***
  • Posts: 7453
  • Country: pl
Re: Voltage boosting an opamp
« Reply #56 on: November 11, 2019, 07:44:20 pm »
Almost correct, although in principle you should also account for intrinsic emitter resistance. That's the inverse of transconductance: how many mV ΔVbe per 1mA ΔIc. In a VCCS circuit, rE adds to your emitter resistor, for relatively obvious reasons.
That being said, transcondutance at collector current Ic is approx Ic/30mV, depending on die temperature. So your rE is a few ohms, quite insignificant.
 

Offline macboy

  • Super Contributor
  • ***
  • Posts: 2309
  • Country: ca
Re: Voltage boosting an opamp
« Reply #57 on: November 12, 2019, 01:29:34 am »
Your solution is the same type, as if you would just say "buy a complete ready test instrument for the purpose", why making one. No thanks.

Now I politely ask you to leave this thread, as you did not contribute anything useful, rather just being provocative and nitpicking about what is or isn't expensive, which is not the topic this thread is about. Neither are you entitled to evaluate how much work something required to make. This is my thread and I make the rules.

Thank you for making mess out of this thread, no-one will be willing to read. Please leave. Bye.
It most certainly is not your thread just because you started it, and you do not make any rules in this forum.
 
The following users thanked this post: SiliconWizard, InductorbackEMF

Offline Simon

  • Global Moderator
  • *****
  • Posts: 18118
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: Voltage boosting an opamp
« Reply #58 on: November 12, 2019, 07:50:55 am »
Your solution is the same type, as if you would just say "buy a complete ready test instrument for the purpose", why making one. No thanks.

Now I politely ask you to leave this thread, as you did not contribute anything useful, rather just being provocative and nitpicking about what is or isn't expensive, which is not the topic this thread is about. Neither are you entitled to evaluate how much work something required to make. This is my thread and I make the rules.

Thank you for making mess out of this thread, no-one will be willing to read. Please leave. Bye.

Now I will politely ask you to leave this forum if this is the way you intend to conduct yourself! you don't own the thread oh powerful one so I have locked it, SO THERE!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf