Transistor equivalent

[KeepItSimpleStupid]

I don't get it.
==
The OP is counting the beans in this thread. When stuff gets designed for mass production, sometimes a critical part gets left out or substituted and sometime unknowingly. One of has an entirely different character when I was doing it for work. 1) I never wanted to see it again for repair 2) I wanted it to be re-useable/repourposed/modular I had stuff in use for 20+ years at work.
3) If you use DIN rail stuff nearly all the parts become totally reusable (Think Erector set) except the wires and the case.

I built a bunch of temperature control boxes: Same 120, Some 240. Configurable power fail drop out, Load disconnect via relay and indicator (had to work on exposed wiring in a vacuum chamber), Some had independent over temperature probes. All had twist lock power inlets that could be used for 240 or 120 and prominently labeled. The cord could be either - no protection.
All had a 10 pin terminal block connector that could do whatever it needed like analog out, analog in, RS232, RS485, alarms; whatever. It meant a 120/240 to 24 VAC or 120 VAC control transformer inside. Wiring changes were not necessarily straightforward. Two front panel fuses were installed even though one might be populated. In the US we have to fuse both legs of the split phase.

One thing I regret. I purchased a 24 channel relay board and I didn't use a suspended chassis construction. Replacing a relay would have been a real chore. Almost like rebuilding it. Fortunately, I could easily move to a spare. One of these https://www.google.com/url?sa=t&rct...al.pdf&usg=AFQjCNFbYPsZCBpMvsahMh0r9GrarOfAqQ in the bottom of a 2RU case.

The deal; was to have 24 label able lighted (current state) push buttons on the front panel with a small computer/manual toggle switch and the "computer's state" displayed at the same time.



==

 

[spec]

I don't get it.
==
The OP is counting the beans in this thread. When stuff gets designed for mass production, sometimes a critical part gets left out or substituted and sometime unknowingly. One of has an entirely different character when I was doing it for work. 1) I never wanted to see it again for repair 2) I wanted it to be re-useable/repourposed/modular I had stuff in use for 20+ years at work.
3) If you use DIN rail stuff nearly all the parts become totally reusable (Think Erector set) except the wires and the case.

I built a bunch of temperature control boxes: Same 120, Some 240. Configurable power fail drop out, Load disconnect via relay and indicator (had to work on exposed wiring in a vacuum chamber), Some had independent over temperature probes. All had twist lock power inlets that could be used for 240 or 120 and prominently labeled. The cord could be either - no protection.
All had a 10 pin terminal block connector that could do whatever it needed like analog out, analog in, RS232, RS485, alarms; whatever. It meant a 120/240 to 24 VAC or 120 VAC control transformer inside. Wiring changes were not necessarily straightforward. Two front panel fuses were installed even though one might be populated. In the US we have to fuse both legs of the split phase.

One thing I regret. I purchased a 24 channel relay board and I didn't use a suspended chassis construction. Replacing a relay would have been a real chore. Almost like rebuilding it. Fortunately, I could easily move to a spare. One of these https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&cad=rja&uact=8&ved=0ahUKEwiDyeb44a_JAhWF2B4KHeOqAgsQFghKMAY&url=https://www.artisantg.com/info/keithley_erb24_manual.pdf&usg=AFQjCNFbYPsZCBpMvsahMh0r9GrarOfAqQ in the bottom of a 2RU case.

The deal; was to have 24 label able lighted (current state) push buttons on the front panel with a small computer/manual toggle switch and the "computer's state" displayed at the same time.
==

Understand now.

==
The OP is counting the beans in this thread. When stuff gets designed for mass production, sometimes a critical part gets left out or substituted and sometime unknowingly. ==

It used to drive me up the wall when the management, or someone, meddled with a design, then blamed you when the circuit wouldn't work properly.

One time, when I went to a new department, this chap kept on about a module that I designed. In the end, he even built it to prove it wouldn't work. Sure enough, when he tested it, it didn't work properly. He was very pleased with himself and started telling everyone that my designs were junk.

In the end, I went to see the circuit he had built and test it out for myself. It was a complete rat's nest, but the circuit was right. The other thing I noticed was that the lab power supply he was using was only 1A and the circuit really needed more than that. It was intended to run off a car battery.

I got him to turn it on to demonstrate the problem. Sure enough, it didn't work properly. An inverter, instead of singing like a bird, was starting up and immediately shutting down, so that it made a chirping sound like a bird in distress. The power supply was tripping and when the inverter stopped working it was recovering... and so on.

I got a nice juicy low-tech 10A supply and we tried again- same chirping sound from the inverter but much louder, and a lot of current from the power supply. I found that the inverter trans were taking huge gulps of current and then giving up. I was baffled and then spotted the problem - my friend had dropped the transformer core and cracked it. Not only that, but be hadn't taped the two halves of the core tightly together. A touch of super glue and some tape and the inverter sang like a bird- he never apologised though, and continued sniping behind my back. I never understood what his problem was- insecurity, I sppose- you should have seen his designs. He looked the part though: smart suit and impressive at meetings.

It must be great to have something you have designed in the field. I never did- just one user group, normally military.

We did have two departments, small boat radar and scopes, that designed for production though.

The company invented the brain scanner- did some work on that, but only documentation and reviews.

 

[MrAl]

Besides the above, any improvement in current capability is good. For one Z isn't constant and 8 is a 'number". Both damping factor and the "ability" to force current will provide some ammunition against the opponent whose motto is "The current through an inductor cannot change instantaneously".

Hi,

I am not sure if this is what you were talking about but, i learned a long time ago when studying state variable feedback that if we measure the right quantities and go by a certain theory in Control Theory (a controllability theory) that we can control almost anything almost instantaneously to an extremely close tolerance with very little error. For this discussion that would be the speaker cone movement, measuring various things about the speaker and maybe the sound it produced. But then i found that the system limits creep in. If the power supply voltage isnt high enough, we still reach a limit of speed of response, because trying to force 1 amp through an 8 ohm speaker with only an 8 volt power supply may not be possible to do really really fast because of the series inductance, which would demand a much higher voltage than 8 volts at least for the time it took to get the speaker mass to overcome the inertia and start to actually move in the required direction. "System saturation" is a good way to describe this. With a feed forward control only though (most systems today) that doesnt happen only because it doesnt have the right feedback to begin with.
I've dreamt of a speaker driver that measures the movement of the speaker cone, and applies the appropriate signal in order to make it fully and quickly compliant. Did very little about it so far though. The idea, in theory, would turn almost any speaker into a full range speaker, provided it was not driven too hard for the normal range of sounds.

 

[Nikolai Petrenko]

My wife and I were wondering about your origins. Going by your name, we figured Russia.

I am Vietnamese but small part in me is Russian. My name come from complex reasons. I always call is my "secondary name"
I can't buy LM350 but LM317 and LM337 are available, IRF530 and IRF9530 more compact than 2N3055 when use in supply circuit.
I start to feel how serious the amplifier we are designing, 80W out but 500W in . I know this idea will make trouble to you, spec, but what is your opinion of use class AB instead of class A in subwoofer amp?
(my brother said me: "What a big amp! If I am you I will use class AB in satellite amp and class B in subwoofer amp!")

 

[spec]

I am Vietnamese but small part in me is Russian. My name come from complex reasons. I always call is my "secondary name"

I think it is a great name- like a movie star.

I can't buy LM350 but LM317 and LM337 are available,

OK

IRF530 and IRF9530 more compact than 2N3055 when use in supply circuit.

Will investigate

I start to feel how serious the amplifier we are designing, 80W out but 500W in . I know this idea will make trouble to you, spec, but what is your opinion of use class AB instead of class A in subwoofer amp?
(my brother said me: "What a big amp! If I am you I will use class AB in satellite amp and class B in subwoofer amp!")

It would be a beast- big and beautiful.

Class A or Class AB is not a problem. The amp I have in mind will work in both. The only difference is that the heatsinking will be less for the AB configuration. Of course the quesent disipation in class AB will be much less- about 21W total for all three amps and power supplies.

(1) Each amp disipation: 60V x 100mA= 6W

(2) Each power supply dissipation: (80V-60V ) * 100ma= 1W

Why not go for class AB for all three amps? Shall I work on that?

 

[Nikolai Petrenko]

Can you compare sound quality of class A vs class AB? (Hmm, I always ask like this)
A good speaker that can handle 20w may has diameter 15-25cm? How large is a 40w subwoofer should have?

 

[Nikolai Petrenko]

here is my version of supply circuit

 

[Nikolai Petrenko]

here is a website that sell components,spec can you go to this link and read all the transistor can be use, I don't know a lot of them. there are 3 page, roll down, you can see button to jump to page 2 or 3. I apologize because let you to involve this problem, but I am busy now, learn to pass the first important exam of this year. The next year I will have even more important exam in June so wait for my amp in the next year summer holiday
https://mualinhkien.vn/danh-muc-san-pham/185_tran-cam.html

 

[audioguru]

Most hi-fi audio amplifiers are class-AB. Each year more amplifiers are class-D.
A properly designed or IC class-AB hi-fi amplifier sounds perfect. Some class-AB discrete amplifiers have an adjustment to set the bias current in the output transistors to eliminate crossover distortion but an IC has the bias set automatically.

Your voltage regulators are not needed for an audio power amplifier. They will not work properly:
1) The Mosfets are source-followers that have a voltage loss of 3V to 6V that changes when their temperature changes and changes when the load current changes. Therefore the output voltage is too low and is not regulated.
2) The output voltages from the Mosfets are not regulated because the Mosfets are not included in the IC circuit, instead the Mosfets are simply hanging on their outputs.

 

[Nikolai Petrenko]

Sorry, I real don't know anything about mosfet, I just thing it like bjt, I will learn.
Here is the fix version, use bjt TIP41-42. Main power in my local very unstable if a high power device is switch on, can make "pop" sound in the amp.

 

[spec]

 

[audioguru]

Audio power amplifiers usually do not need voltage regulators.

Your new power supply but with bipolar transistor emitter-followers hanging on the output also has poor voltage regulation.
The transistor is usually included in the circuit for excellent voltage regulation like this:

 

[spec]

 

[KeepItSimpleStupid]

At least one person, OK two, know the importance of "star grounds". Nice!

 

[spec]

 

[spec]

At least one person, OK two, know the importance of "star grounds". Nice!

much experience of oscillating and distorting amps!

 

[Tony Stewart]

5ch x 450W 0.5%THD @120Vac in
wt: 116 lbs (52kg)
price: $6,395
class: A
DF=400
Thermaltrak devices have tracking diodes in the same package as the transistors so bias is optimized in real-time.
https://www.soundandvision.com/content/ati-at6005-amplifier-specs
https://www.wwsp.com/ati/powerampspecs.htm


But you may need a variac or two separate 20A outlets to get 120V input to meet THD specs at full load from input line loading.

 

[spec]

I've dreamt of a speaker driver that measures the movement of the speaker cone, and applies the appropriate signal in order to make it fully and quickly compliant. Did very little about it so far though. The idea, in theory, would turn almost any speaker into a full range speaker, provided it was not driven too hard for the normal range of sounds.

Hi MrAl,

Same here- measuring the actual cone movement or even the sound in the room would complete the feedback loop- more or less nose to tail. It has been tried but, as far as I know, never really worked out. You can put a transducer on the stiff cone of a base driver, the old KEF B139, for example, where the cone moves like a piston up to say 200Hz. The end result would be to flatten the frequency response. With other drivers the question is what do you measure because the cone doesn't necessarily move like an ideal piston so the problem is what do you use for the error.

But the real killer in any feedback system is loop stability. In general, the more accurate, implying high open loop gain, the more stability problems you have.

Using feedback from a microphone would seem to be the ultimate, but I think even more problematic because the room alters the sound.

 

[spec]

5ch x 450W 0.5%THD @120Vac in
wt: 116 lbs (52kg)
price: $6,395
class: A
DF=400
Thermaltrak devices have tracking diodes in the same package as the transistors so bias is optimized in real-time.
https://www.soundandvision.com/content/ati-at6005-amplifier-specs
https://www.wwsp.com/ati/powerampspecs.htm

But you may need a variac or two separate 20A outlets to get 120V input to meet THD specs at full load from input line loading.

Xmas present perhaps- have 3kW Variac

 

[spec]

A good speaker that can handle 20w may has diameter 15-25cm? How large is a 40w subwoofer should have?

15-25cm= 6" to 10" in English. The range of base drive units is much greater than that, about 4" to 20", and over for some high power types.

You really ask the wrong question:

In a typical audio signal, most of the power is in the low frequencies, and also cone extension is inversely proportional to frequency (I think that is right). The net result is that the sub chassis has to handle a lot of power and push a lot of air. That's why they tend to be huge, hefty, and heavy. The other thing is that, cone extension, for a given power, is inversely proportional to cone area: the bigger the area the smaller the extension.

In general, speaker distortion is proportional to cone extension which is inversely proportional to cone area. So you may think, why not always have a large cone? The reason is size and cost. But there is more to it than that: the larger the cone, the more difficult it is to keep it moving as an ideal piston.

That is the foundations, but the next subject is frequency response. Ideally, you want a frequency response flat from DC to 200Hz say- it varies. Physics makes this impossible and the speaker will resonate, typically at 10Hz for a very large unit up to 150Hz for a small one. The resonant frequency is proportional to compliance/mass, where mass is the weight of the moving part of the cone and compliance is the overall stiffness of the suspension. Base drivers universally have a two-part suspension with a spider at the centre and a surround at the periphery of the cone where it joins to the speaker frame.

You might ask, why not make the cone heavy and the suspension really compliant to get a resonant frequency as low as you like. As usual in life, you never get something for nothing.

The heavier the cone the less efficient the speaker becomes and the more difficult it is for the suspension to keep the cone moving without wobbling around, generating distortion, and even rubbing the voice coil on the magnet pole.

Those are the basic constraints when designing a sub speaker chassis and there are two broad approaches for realising a good bass response:

(1) Large heavy cone and relatively stiff suspension giving a small excursion: inefficient.

(2) Small cone and very compliant suspension giving a large extension: efficient.

There are advantages and disadvantages with each approach

In general, the current trend for home HiFi sub chassis, in the UK anyway, is category (1) towards the middle: 8" to 10" inch cone and medium compliant suspension and a resonant frequency of around 20Hz. On the other hand, car sub chassis are normally in category (2) especially the very high power types.

In the valve (tube) amp days, speakers were quite efficient, so 10W to 20W amps were enough to get a decent volume.

In the 70s and 80s the trend, in the UK, was for very inefficient speakers and, as a result, power amp power went up accordingly: 30W to 100W even 1KW.

Guitar speakers are altogether different and are very efficient by comparison: light paper cones and no power-absorbing cross-over networks. Also, guitar amps don't need to handle very low bass. The frequecy range of an electric lead guitar, is is 82 Hz to 660 Hz (bass guitar: 41Hz to 198Hz). That's why a 30W, Vox AC30 combo, for example, sounds so loud. Guitar loudspeaker units also have characteristic distortion. If you play an electric guitar through a HiFi system it sounds clinical and uninteresting- not so miked acoustic guitars.

The next constraint is the voice coil: speakers, in general, are dreadfully inefficient, typically 3% (I think), so if you fed a speaker chassis with a 100W continuous tone, 97W would be dissipated in the voice coil resistance. There is more to it than that, but sufficiently accurate for explanation. To cope with this kind of dissipation special techniques are used: ceramic voice coil former, high temperature insulating varnish on the coil wire, ferro-fluid cooling, and so on.

The two main reasons for the low efficiency are:

(1) The voice coil resistance causes a total loss and simply converts the input power into heat (I↑2*R). Ideally the voice coil resistance should be zero Ohms, but this is not practicable.

(2) The coupling between the speaker cone and free air is very bad at low frequencies. That's why massive exponential horns are used to match the radiation resistance of the cone to the resistance of free air. At the other extreme, a tweeter with a normal shaped cone, tends to produce a disconcerting beam of sound, so dome cones are used to disperse the sound.

Finally, when a bass speaker chassis is fitted into a cabinet, its characteristics are radically altered, but that is a whole subject in itself.

To answer your question about which size speakers are best for a bass sub- it depends on the many factors listed above. In general, in your situation, the bigger the larger the diameter of speaker cone the better, but the funamental design of the speaker chassis will define what you can do with it. The cabinet that you use- closed box, ported, transmission line, etc will greatly effect the outcome too.

One thing you can try, but you stand a chance of wrecking the speaker chassis, is to modify the cone (see post 223). Ultimately, the allowable voice coil dissipation will limit you. You could make an excellent base sub if you bought a bass speaker chassis designed for the job. Have you considered casting an alluminium cabinet?

https://www.ejjordan.co.uk/

https://www.stereophile.com/content/epos-es-14-loudspeaker-john-atkinson-review#XgY7QxYbfAt0PRIE.97