Hello!
What is the minimal equipment for building a class-d low-bandwidth bass amplifier (maximal frequency 150 Hz)?
For example, i got to formulate one of the points more exactly: Often there are unwanted oscillations (i think, most of them in the MHz-range, and independent of the planned low switching frequency).
For analysis one needs a scope which can deal with that, so: What is the recommendable minimum?
FQR
What is the minimal equipment for building a class-d low-bandwidth bass amplifier (maximal frequency 150 Hz)?
For example, i got to formulate one of the points more exactly: Often there are unwanted oscillations (i think, most of them in the MHz-range, and independent of the planned low switching frequency).
For analysis one needs a scope which can deal with that, so: What is the recommendable minimum?
FQR
Last edited:
It's not that they don't want to but they may not know and/or they may view the fact that you have to ask implies you may not have the knowledge.
That being said there are a few things you will need:
The first is a scope as you mentioned. Most of the time 20-30Mhz analog should be enough to see what you need to see. I'd double that if digital. Either way, its not too difficult to get.
Now comes the difficult part. Because of the nature of the switching and the voltages involved, from my experience you NEED a differential probe. These can be VERY pricey. You can sometimes get by working on a switching device without one but when things go wrong they become essential.
The third is often times a quality LCR device (one with a kelvin connection and capable of more than one or two frequencies).
That being said there are a few things you will need:
The first is a scope as you mentioned. Most of the time 20-30Mhz analog should be enough to see what you need to see. I'd double that if digital. Either way, its not too difficult to get.
Now comes the difficult part. Because of the nature of the switching and the voltages involved, from my experience you NEED a differential probe. These can be VERY pricey. You can sometimes get by working on a switching device without one but when things go wrong they become essential.
The third is often times a quality LCR device (one with a kelvin connection and capable of more than one or two frequencies).
Last edited:
Very handy is a current probe. Look for a used TEK, like a 6042. You should be able to find one for @ $250-300 It is a 50 MHz unit. Very handy to see peak current, look for saturation in transformers and inductors, etc. I've had one for years and never regret buying it.
Of course a scope is a necessity. 50 MHz is a nice unit. Again used TEKs are great buys and generally rock solid investments.
paul
Of course a scope is a necessity. 50 MHz is a nice unit. Again used TEKs are great buys and generally rock solid investments.
paul
Thanks very much, guys! 
I don´t know nothing, but i have not enough net access to search all that i gotta know. Ok, so i´ll need a scope with the mentioned bandwidth, and those (or other - i want to be able to deal with more than 10 amperes) fitting probes. Yes, i also favor tektronix - it´s a terrific quality brand!
Until now my LCR-meter is not so good - it´s only a victor vc88, has only only one measuring frequency, and is a little unaccurate.
Only good thing. i already have, is a big amount of things related to 50Hz AC and some to DC, cause with line level things i´ve been working nearly every day of the least 20 years.
And fortunately this means really many meters, circuit breakers, fuses, relays, switches, heavy electrolytic caps, diodes, chokes and transformers, in all imaginable versions and power classes. Thats better than one isolation tranny of one size, it´s even more flexible than any laboratory linear supply. Because i can make thousands of combinations for test powering with the right voltage and limited as good as fused current.
And i got nearly 40 pcs. of IRFPC50A, and over 30 pcs. IRFP450, of which i want to build an adjustable electronic load for testing. Until now i don´t know exactly how i should build it the best way. But that would probably be a new topic here... ;-)
Again, many thanks to all!
I don´t know nothing, but i have not enough net access to search all that i gotta know. Ok, so i´ll need a scope with the mentioned bandwidth, and those (or other - i want to be able to deal with more than 10 amperes) fitting probes. Yes, i also favor tektronix - it´s a terrific quality brand!
Until now my LCR-meter is not so good - it´s only a victor vc88, has only only one measuring frequency, and is a little unaccurate.
Only good thing. i already have, is a big amount of things related to 50Hz AC and some to DC, cause with line level things i´ve been working nearly every day of the least 20 years.
And fortunately this means really many meters, circuit breakers, fuses, relays, switches, heavy electrolytic caps, diodes, chokes and transformers, in all imaginable versions and power classes. Thats better than one isolation tranny of one size, it´s even more flexible than any laboratory linear supply. Because i can make thousands of combinations for test powering with the right voltage and limited as good as fused current.
And i got nearly 40 pcs. of IRFPC50A, and over 30 pcs. IRFP450, of which i want to build an adjustable electronic load for testing. Until now i don´t know exactly how i should build it the best way. But that would probably be a new topic here... ;-)
Again, many thanks to all!
Last edited:
Thanks very much, guys!
I don´t know nothing, but i have not enough net access to search all that i gotta know. Ok, so i´ll need a scope with the mentioned bandwidth, and those (or other - i want to be able to deal with more than 10 amperes) fitting probes. Yes, i also favor tektronix - it´s a terrific quality brand!
Until now my LCR-meter is not so good - it´s only a victor vc88, has only only one measuring frequency, and is a little unaccurate.
Only good thing. i already have, is a big amount of things related to 50Hz AC and some to DC, cause with line level things i´ve been working nearly every day of the least 20 years.
And fortunately this means really many meters, circuit breakers, fuses, relays, switches, heavy electrolytic caps, diodes, chokes and transformers, in all imaginable versions and power classes. Thats better than one isolation tranny of one size, it´s even more flexible than any laboratory linear supply. Because i can make thousands of combinations for test powering with the right voltage and limited as good as fused current.
And i got nearly 40 pcs. of IRFPC50A, and over 30 pcs. IRFP450, of which i want to build an adjustable electronic load for testing. Until now i don´t know exactly how i should build it the best way. But that would probably be a new topic here... ;-)
Again, many thanks to all!
An adjustable electronic load is NOT easy at all. You have to fight oscillations, non-matching resistances, the effects of non-balanced loading on the transistors etc. it is dramatically harder than just a power supply. In its most basic form (one FET/BJT) then it can be easier but anything beyond that can become VERY difficult VERY quick.
Hello Pedro,
you´re surely right, there are many different kinds of electronic loads, and they can have many functions. Some of them are even programmable, you can determine a series of different conditions, changing in fixed ways, before starting. Building something like that is nearly impossible for me. But even building much simpler ones could bring trouble.
By thinking about my concrete demands i found a possible solution for fundamental testing of smps´s with symmetrical output: Eventually i could build a quasi-complementary amp, able to stand this outputs easily (what means possible high voltage and current). And feed it from a frequency-generator. Between the generator and the amp i could build a network of switches, potentiometers and resistors, to control the input and so likewise the output, what controls the power demand, too.
That would not be too difficult. I know, that´s not the same as a real electronic load, with functions like fixed current, fixed voltage, ... etc., but some testing would be possible with it. For the final loudspeaker load i could use loudspeakers and/or a network of resistors (and maybe chokes and capacitors), to simulate some fundamental conditions.
What do you all think?
Greets,
FQR
you´re surely right, there are many different kinds of electronic loads, and they can have many functions. Some of them are even programmable, you can determine a series of different conditions, changing in fixed ways, before starting. Building something like that is nearly impossible for me. But even building much simpler ones could bring trouble.
By thinking about my concrete demands i found a possible solution for fundamental testing of smps´s with symmetrical output: Eventually i could build a quasi-complementary amp, able to stand this outputs easily (what means possible high voltage and current). And feed it from a frequency-generator. Between the generator and the amp i could build a network of switches, potentiometers and resistors, to control the input and so likewise the output, what controls the power demand, too.
That would not be too difficult. I know, that´s not the same as a real electronic load, with functions like fixed current, fixed voltage, ... etc., but some testing would be possible with it. For the final loudspeaker load i could use loudspeakers and/or a network of resistors (and maybe chokes and capacitors), to simulate some fundamental conditions.
What do you all think?
Greets,
FQR
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.