EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: 460voltclub on September 10, 2020, 04:13:35 pm
-
I am building a subwoofer amplifier from a schematic. I have two 2n3773 transistors Pin #1 Base #2 Emitter #3 Collector. I have a +35V and a -35V rail. One 2n3773 transistor has the base connected to the emitter of a 2n6292. The 2n3773 emitter is connected to a 0.33 Ohm 5 Watt resistor at the speaker output. The 2n3773 collector is connected to +35 vdc.
My confusion is how do I connect +35 vdc to a 2n3773 collector mounted on a heat sink.
-
If you are mounting both 2N3773s on a common heat sink, then you need to insulate both from the heat sink, or insulate one transistor and insulate the entire heat sink.
There are TO-3 mounting packages available, with a thin insulating washer and insulated screw/nut/washer sets. You need to add a #6 solder lug on any insulated transistor to connect to the case, which is the collector of the 2N3773.
The TO-220 and similar plastic power packages that are somewhat compatible with the TO-3 metal package have a third lead for the collector, along with the mounting pad on the back of the transistor.
The Keystone mounting kit https://www.mouser.com/ProductDetail/Keystone-Electronics/4734?qs=fP5bVVCrK%2FcGQnu5Gb90jQ%3D%3D (https://www.mouser.com/ProductDetail/Keystone-Electronics/4734?qs=fP5bVVCrK%2FcGQnu5Gb90jQ%3D%3D) includes a socket that has a connection through the screws to the TO-3 case.
Be sure to put a thin coating of thermal grease on both sides of the washer to ensure good thermal contact to the TO-3 case and the heat sink.
-
Take care with the electrolytic capacitors C2 and C3. As far as i understand your schematic shows both of them in wrong polarity. An easy to correct mistake. C4 seems to be correct. C1 should be a foil capacitor.
This is a typical beginner project. There are many parts missing that a commercial product would include. In my opinion the main omission is an output current limiter (output short circuit protection). Another omission is a capacitive voltage divider in the feedback (parallel to R5/R6). It helps a lot against unwanted high frequency instability. On the output a commercial amplifier would have a 4.7 Ohm & 100 nF absorber, also to avoid instability. Anyway, such projects are a very good starting point, and building a good audio amplifier is an art that some people spend their whole life on. Others just wire up some kind of integrated "gain clone" and it may appear simple but it isn't.
Regards, Dieter
-
It feels so weird looking at a color bitmap rendition of what was high technology.....back in 1968.
Suffice it to say, the schematic faithfully reproduces that level of development, and hasn't been updated in any way to offer such desirable features as:
- Using cheap and available transistors
- Stable biasing network
- Wide frequency response
- Low offset voltage
- Class AB operation
- Low distortion
- Protected outputs (current limiting, thermal overload)
These can be added through various means, but rather than discussing them, it would be much easier to just shop around for a newer circuit.
A note, R14 isn't in the right place; I'm not sure if this oversight is historically correct or a transcription error, but it should be in the emitter circuit of Q6. That is, keep Q6 emitter tied with Q7 collector, and put the resistor between them and the output. This way, bias current creates a symmetrical voltage drop through R16 and R14, above and below the output voltage. These voltage drops correspond to D1-D2-(R9 || R10), which sets the bias current. I think between the rectifier diodes and the tiny resistance, this will be in class B operation, which saves on bias current and stability I suppose, but isn't well renowned for low distortion.
Tim
-
Thanks for the input. I build and fix electronics for the fun of it. I have a 12 x 12 shed that is my "Lab" and I have a Pioneer car stereo for music. I have a subwoofer and power transistors in my "maybe make something parts box" and I wanted a simple sub amp to build.
I repair truck PCM's and I have repaired 3 Tek O-scopes which I own. But for some reason I was having a brain fart and just couldn't get my head around this issue.
Once I read your posts I had a "Crap I remember now" epiphany.
Thank You
Steven
True to my username I have been jolted with 120v, 208v, 240v and 460v several times.
The last three phase 460 left three deep scars on my right arm.
Sometimes you can't fix stupid.
-
It'll certainly work, and it doesn't sound like you'd mind repairing it from time to time if it goes tits up... Maybe more inconvenient if a shorted transistor puts full DC in your speaker, extracting its magic smoke. I'd spend the few more components on a nicer design though. Can still use old transistors like 3773, they're slow and low gain, but that's just tweaking the compensation components, so long as you don't need the bandwidth. (Which, 10kHz BW would still be fine here, that gives plenty of loop gain down at subwoofer frequencies to help reduce distortion.)
Yeah... Lower voltages are shocking in varying degrees, but 480V is just outright angry... You're lucky!
Tim
-
I get a kick out of people taking older 1970's circuits and giving them a trendy label, keeping the old transistors that are long obsolete, and voila it's all over the Internet.
I would say C3 is too small for a "subwoofer amplifier" I think -3dB at 23.4Hz great back in the day when playing a warped record lol. R9 and R10 at 22R 1/2W seem way huge for the 11R 1W resistor they make, something is wrong with that. R3 is huge compared to R2.
So when I see a couple design copy-pasta mistakes, I kinda wonder what the original design is. It seems to be British with BC108's and not Elektor font (BC108 pic) but from another UK hobby magazine or maybe Maplin.
480V is fine but I did not like crawling around VFD's with hissing corona discharge, the DC bus is up there.
-
I would say C3 is too small for a "subwoofer amplifier" I think -3dB at 23.4Hz great back in the day when playing a warped record lol. R9 and R10 at 22R 1/2W seem way huge for the 11R 1W resistor they make, something is wrong with that. R3 is huge compared to R2.
The -3dB cut-off increases to something like 28Hz when you also take into account the effect of the input coupling cap C1 and input resistor R1.
-
It feels so weird looking at a color bitmap rendition of what was high technology.....back in 1968.
It is curious that the person drafting the schematic chose to colour code the transistors by type but kept the all resistors the same colour.
-
480V is fine but I did not like crawling around VFD's with hissing corona discharge, the DC bus is up there.
Says the guy from Canada while he is surrounded by 575V vfd’s with 900 volt rated dc buses :scared:
Back in the days before insulated/CAT rated screwdrivers and touch safe terminal blocks, got nailed by 480, tightening terminal blocks with power on. It wasn’t pleasant. Rated right up there with peeing on an electric fence :palm:
-
My confusion is how do I connect +35 vdc to a 2n3773 collector mounted on a heat sink.
Typically the TO-3 transistor and hardware are electrically isolated from the heat sink and a tab is attached to one of the mounting screws. There are also insulated TO-3 sockets made to go under the heat sink which make this easier.
So the collector connection is through one or both of the mounting bolts.