I have got one TTC5200 / TTC1943 board like one shown here
Transistor circuit diagram of 2sc5200 and 2sa1943 - Electronics Help Care
I am using a 35-0-35v transformer which gives about 47-0-47v DC. I don't think the sound is very clean like the LM3886 or LM1875 amplifiers I made before. For filtering the DC voltage I have used the same 4700uF 50v capacitor used for the LM3886 amplifier.
Is the amplifier circuit good or am I missing something here? Thanks.
Transistor circuit diagram of 2sc5200 and 2sa1943 - Electronics Help Care
I am using a 35-0-35v transformer which gives about 47-0-47v DC. I don't think the sound is very clean like the LM3886 or LM1875 amplifiers I made before. For filtering the DC voltage I have used the same 4700uF 50v capacitor used for the LM3886 amplifier.
Is the amplifier circuit good or am I missing something here? Thanks.
Well, After listening to it again I think it is alright at low volume. At high volume distortions are coming.
I don't see any circuit diagram, just the crude copper etch pattern. It doesn't look promising as a project but I guess that's down to your power supply, actual parts quality and how you go about the assembly and set-up adjustments.
Amplifiers like this one require bias setting. Bias is a small DC current of about 35mA that flows through each output transistor, all the time. It needs to be measured and adjusted correctly to minimise crossover distortion, which is what it is for. This is already preset in chipamps but not in discrete parts designs like this one. Did any instructions point this requirement out?
Amplifiers like this one require bias setting. Bias is a small DC current of about 35mA that flows through each output transistor, all the time. It needs to be measured and adjusted correctly to minimise crossover distortion, which is what it is for. This is already preset in chipamps but not in discrete parts designs like this one. Did any instructions point this requirement out?
No instructions, nothing. I bought the fully assembled board only without the 5200 and 1943 for about 1.4 USD. There are no variable resistor to adjust on the board. I have only added the 5200/1943 with heatsink and I am running it.
Oh, my gosh - $1.40 US? Even the Chinese output transistors alone would cost more than $1 each in large quantities.
Draw yourself a schematic diagram by whatever means you have, such that you have a conventional circuit diagram that we can all follow. Assuming you have a multimeter that can read very low voltages, resistance and currents with some accuracy, power the amplifier without speakers connected and measure the voltage across each 0.27 ohm resistor. From the voltage across the resistors, you can calculate the bias current in each power transistor and they should all be within 5% of each other. For example, 30 mA current in each resistor would show a voltage across the resistor leads of V=I*R or 0.03*0.27 = 8.1 mV
Just note that these are quite small measurements and very cheap meters may not have the precision to give meaningful results.
Draw yourself a schematic diagram by whatever means you have, such that you have a conventional circuit diagram that we can all follow. Assuming you have a multimeter that can read very low voltages, resistance and currents with some accuracy, power the amplifier without speakers connected and measure the voltage across each 0.27 ohm resistor. From the voltage across the resistors, you can calculate the bias current in each power transistor and they should all be within 5% of each other. For example, 30 mA current in each resistor would show a voltage across the resistor leads of V=I*R or 0.03*0.27 = 8.1 mV
Just note that these are quite small measurements and very cheap meters may not have the precision to give meaningful results.
Ok, I will try to make a schematic diagram out of it. I have made 5 other amplifiers, LM3886 x3 and LM1875 x2. All of them worked very well without much of adjustment. In the LM3886 board I designed, I only had to remove the Thiele network coil because it was picking up noise. All of them sound excellent now.
This is the 6th amplifier for the subwoofer. I will make a subwoofer filter (Rod Elliot Project 48 Rev A) and try with this amplifier. I have a 12 inch subwoofer, I hope two+two 5200/1943 will be powerful enough for it. I have no idea but I guess for subwoofer very clean sound will not be required. If it works fine then I will leave it like it is, otherwise I will try again. Thanks for your help.
This is the 6th amplifier for the subwoofer. I will make a subwoofer filter (Rod Elliot Project 48 Rev A) and try with this amplifier. I have a 12 inch subwoofer, I hope two+two 5200/1943 will be powerful enough for it. I have no idea but I guess for subwoofer very clean sound will not be required. If it works fine then I will leave it like it is, otherwise I will try again. Thanks for your help.
Oh btw, when I went to Maplin in the UK for a DC jack, I was really shocked to see the price was £1.50. In our country a DC jack costs £0.05. The Maplin one was of good quality but can never be more than £0.15 here.
Please, I have a similar problem and I need your help, I'm building an amplifier using 2sa1943=16 N 2sc5200=16 the DC voltage I want to use is +/- 120 volts which is to be supply to output transistors, but for driver circuit will be supply by +/- 60 volts. Please I want to know if this +/- 120v will either over heat the 16 pairs output transistors or not. Thank you
That circuit is very poorly implemented. TO-220 Power transistors in the diff pair, and you wonder why it doesn’t sound very good? I would *expect* a 3886 to sound better. It’s using transistor overkill everywhere - probably so that it will work with fake transistors on sufficiently low voltage without blowing straight up in the first 10 seconds. For a buck 40, you get what you pay for.
This ain’t a “similar problem”. Not even close. First off, running output stage off a MUCH higher voltage than the driver circuit is not a good idea for a typical unity gain output stage. You can’t make use of the full voltage - unless the output stage has voltage gain. Newbies shouldn’t try this unless they are following a proven circuit - there are MANY pitfalls.
+/-120 volt rails is excessive for ANY output transistors, no matter how many you put in parallel. To use these voltages, you either need a class G or H switched rail voltage design, or run the outputs in a series-parallel configuration to lower the Vce. They can’t handle anywhere near full power at 120v, they are useable at 60. Two banks of 8 in series *is* the correct number if you were going to do this. It will take a BIG heat sink and forced air. And a lot more experience before trying this. Physical construction is going to be challenging, and little bitty stability problems can become big fireballs. There are a few class H designs floating around on this forum - like the Apex A900 and some derivatives of it. Proven schematics and PCB artwork exist. You might want to dig these up. You could also clone QSC, Crown, or Crest pro amps. Plenty of those running on +/-120 or even higher (and they work), but none of those just run one big bank of transistors running off a fixed voltage. There’s a reason for that.
Those examples are all "chipamps" where the amplifier comes complete on a single chip die and only requires a power supply and connection to a signal input and speaker plus any optional extras like more comprehensive speaker protection with muting, volume control or perhaps a separate preamplifier.
It's not hard for a complete novice to get chipamps working but a discrete component amplifier requires a lot more thought, planning and usually adjustments to circuit values in order to get stable, good quality output. Often builders destroy their semis in the process of setting up their amp. because they didn't know what they were doing before tweaking things, either by adjusting bias or substituting components without a clue what that actually does.
Ruleworld: That link does not describe a circuit diagram. As others said above.
Nor is the description accurate. "One transistor can handle a maximum of 1.2A"
That does not apply to a 2SC5200 (15A - but it would be unusual for anyone to push it that far). Nor is the calculation "four transistors mean 4.8A" anywhere near right. The four transistors are in two pairs, so each pair could only handle 2.4A on the first incorrect premise. The author of the article does not seem to know what he is talking about.
Nor is the description accurate. "One transistor can handle a maximum of 1.2A"
That does not apply to a 2SC5200 (15A - but it would be unusual for anyone to push it that far). Nor is the calculation "four transistors mean 4.8A" anywhere near right. The four transistors are in two pairs, so each pair could only handle 2.4A on the first incorrect premise. The author of the article does not seem to know what he is talking about.