Lt Spice |
Post Reply | Page <123> |
Author | ||
gen0me
Young Croc Joined: 20 February 2016 Location: UK Status: Offline Points: 999 |
Post Options
Thanks(0)
|
|
Ok lets go back to the single stage:
Output is reasonable but the input impedance is horrible. R6 is the test resistor. I check there amperage pulled from the source. So for 150mA and 1V input voltage the input impedance of this seems to be around 6,7ohm. It should be over 10k right? Need to read more.
|
||
I appreciate every like :)) https//www.facebook.com/genomesoundsystems
Mixes: https://www.mixcloud.com/gen-ome/ |
||
gen0me
Young Croc Joined: 20 February 2016 Location: UK Status: Offline Points: 999 |
Post Options
Thanks(0)
|
|
When I lower the C value the bandwidth cutoff on the lower end goes higher. I guess I dont need 10Hz. ~40hz should be ok.
Intended usage is for I guess something like home made studio speaker. It should work on phone voltage and give from it 50-100W.
Edited by gen0me - 01 January 2019 at 10:59am |
||
I appreciate every like :)) https//www.facebook.com/genomesoundsystems
Mixes: https://www.mixcloud.com/gen-ome/ |
||
Danielr
Registered User Joined: 30 May 2016 Status: Offline Points: 209 |
Post Options
Thanks(1)
|
|
Your input impedance really depends on your application.
~ 10 times greater than your source impedance. e.g for a 150 - 200 ohm microphone, aim for 1r5K - 2K for a line amplifier (pre-amp output impedance ~ 1k) aim for ~ 10K for guitar (with 20 - 40K pickups) aim for 200k - 400k input impedance. set it too low and you'll load your source and distort the signal. set it too high and you'll have issues with noise. Your input impedance is determined by the gain and emitter resistor. (R5 in your circuit.) You should start with the datasheet and work from there to determine your resistor values. use the tables on page 2 to plot the output characteristics then figure out what you want your load line to be to get a decent swing on the output avoiding cut off and saturation... pick a value for Vre. That'll let you figure out what you want to use for R3, and things will kind of fall into place after that.
|
||
Earplug
Old Croc Joined: 03 January 2012 Location: Europe Status: Offline Points: 7199 |
Post Options
Thanks(1)
|
|
A couple of points: 1) You can´t drive the 3055 directly from any (simple) line source. For a start, the hfe is far too low. You need a pre, or driver stage before. 2) For single rail operation, you need a cap on the output to remove the DC. And as noted above, there are plenty of Class A examples around. Worth doing some research first. For example: http://sound.whsites.net/project36.htm |
||
Earplugs Are For Wimps!
|
||
odc04r
Old Croc Joined: 12 July 2006 Location: Sarfampton Status: Offline Points: 5483 |
Post Options
Thanks(1)
|
|
Few more points, you won't drive an 8Ohm load properly from a common emitter stage which is typically aimed towards voltage gain. So don't even bother trying. You will need to buffer the output when it is at the required amplitude with a common collector or other similar stage with low voltage gain but high current.
You don't need R6, just plot the current supplied by AC1. C2 is extremely large for an AC coupling cap. Your values of R3 and R5 are silly, R5 might as well be a wire short for what its worth. I presume you have gone for so low values as to minimise power dissipation, but it won't work in the real world. R1, R2 and R5 set the quiescent DC emitter current flow so you can forget any kind of accuracy there with the values you have. As mentioned already, the current gain (beta or hfe) of a 3055 is very low. At 10A Ic it is 5. This gives you two problems. One is that under high current flow your voltage gain is tiny, the second is that your emitter resistance is going to be about 25mV*5 = ~0.1 Ohm. That appears in series with R5 giving you a total of about 0.11 Ohms Re total. The input resistance of a CE BJT amplifier is given by R1//R2//Re. In your case this is going to be dominated by Re and it will be tiny, so yes your circuit as-is will pull massive current from your source. It works in Spice because the source is ideal, in the real world your input signal level will collapse to virtually zero if you try and pull 150mA from a line level source, unless it is very well buffered. 1V/150mA is a load resistance of under 10 Ohms... Even a good headphone amplifier will likely struggle. The current through R2 should be ~10x your base current as a rule of thumb. Being generous lets assume 2A Ic, 50 hfe. Ib = 2/50 = 0.04A (40mA) so in R2 you'd want to see about 400mA for a stiff supply. Your actual flow in R2 (ignoring Ib) is about 55/10000 = 5.5mA. Likely no where near enough to drive the output properly. You have some gain in the simulation but the LTspice model is based on a mathematical extrapolation of the device transfer curves, you are in a weird part of the device operation. If you actually build this circuit I would be very surprised if the results matched reality. All of these issues stem from 2 things. The first is that you are using a screwdriver to pound in a nail. If you want voltage gain then use a small signal large hfe transistor designed to do it. If you want current drive for a power transistor then use a mid level driver transistor. The second is that while electronics simulators can be invaluable for testing ideas, like all models they are garbage in = garbage out. If you don't have an appreciation of how to DC bias the quiescent point of a BJT circuit you are just pissing in the wind by guessing with numbers. I'm all up for trying and learning but electronics doesn't reward guessing. (The odd exception being a well educated guess at a weird physical problem once the circuit is built). So go read up on BJT theory, and implement a circuit such as that shown here. Follow all the maths and work it out as you go, then verify that your simulation matches reality. After that you can try your own design case and build your circuit blocks one by one. One more thing to watch for, small signal frequency analysis pays no attention to the voltage rails of your circuit. Although it says you have 30V at your output it is very likely not a centre swing biased signal. Check it with a 0.1ssecond long 100Hz single tone time based analysis and your output signal is very likely going to be clipped on one side. Edited by odc04r - 01 January 2019 at 3:14pm |
||
odc04r
Old Croc Joined: 12 July 2006 Location: Sarfampton Status: Offline Points: 5483 |
Post Options
Thanks(0)
|
|
I just noticed this, do you have any idea how much heat you have to get rid of to run a 50w+ class A amplifier! |
||
Earplug
Old Croc Joined: 03 January 2012 Location: Europe Status: Offline Points: 7199 |
Post Options
Thanks(0)
|
|
Yup - the amp I linked to earlier is around 15W-20W - with a quiescent current of 1.5A! And running into 4ohm isn´t recommended. I think that maybe Class A is better suited for tubes than BJT´s! |
||
Earplugs Are For Wimps!
|
||
gen0me
Young Croc Joined: 20 February 2016 Location: UK Status: Offline Points: 999 |
Post Options
Thanks(0)
|
|
What is reasoneable choice for current preamp transistor?
Edited by gen0me - 01 January 2019 at 8:22pm |
||
I appreciate every like :)) https//www.facebook.com/genomesoundsystems
Mixes: https://www.mixcloud.com/gen-ome/ |
||
Danielr
Registered User Joined: 30 May 2016 Status: Offline Points: 209 |
Post Options
Thanks(1)
|
|
Are you trying to "learn how" for the fun of learning, or trying to actually build something to use? In some ways... I think the best advice is, Pick any, carry on as you are... For a learning exercise, it would be better if you pick the wrong one, and then as you are designing it you find out why... what Parameters you wish you had. Then you go to a data book and search out a transistor with the parameters that you want. By that time you should be pretty aufait with the design steps and be able to just crack on. If you actually want to build the circuit as a working amplifier... then still do that, because once you're done with the paper exercise designing it you'll know what transistor to pick. If you really just want a component number, then BC108, or BC109, they are old transistors (with detailed reference sheets that will make things really easy for you) also there are load of amplifier designs and tutorials kicking about that use these. No, they are not audio specific, but as a first step, they are less than a £1 each...
|
||
gen0me
Young Croc Joined: 20 February 2016 Location: UK Status: Offline Points: 999 |
Post Options
Thanks(0)
|
|
:) Idea was to learn on making gift. Its too late now but idea stayed. And I will have one more chance before Feb.
|
||
I appreciate every like :)) https//www.facebook.com/genomesoundsystems
Mixes: https://www.mixcloud.com/gen-ome/ |
||
odc04r
Old Croc Joined: 12 July 2006 Location: Sarfampton Status: Offline Points: 5483 |
Post Options
Thanks(0)
|
|
I think you'll be pushing it before Feb in honesty, next Christmas is more like it to do a decent job. But good luck!
|
||
odc04r
Old Croc Joined: 12 July 2006 Location: Sarfampton Status: Offline Points: 5483 |
Post Options
Thanks(0)
|
|
This is one of those things that you first need to derive a spec for, to further inform what is a reasonable choice! Take a look at Rod Eliotts P3A design, it uses a pretty classic design and contains small signal, driver, and power transistors. If you can understandall if its stages you'll be on your way to further enlightenment |
||
Post Reply | Page <123> |
Tweet |
Forum Jump | Forum Permissions You cannot post new topics in this forum You cannot reply to topics in this forum You cannot delete your posts in this forum You cannot edit your posts in this forum You cannot create polls in this forum You cannot vote in polls in this forum |