Ok so I built a job 1969 amp and the problem is that the bass fizzles even at moderate volume. So I set out to investigate... The toroid is 2*0-12V (3 Amp ) toroid. The AC voltage across the secondary shows 13 volts. Voltage across the rectifier and capacitor bank = 18 volts ( disconnected from the amp )
When the amp is connected. The voltage across Rectifier+ Cap bank = 9 volts. Any idea why is this happening? Why does the voltage drop when the amp is connected. By connected I mean the amp is just connected to the power supply witb no input or speaker output.I assume that for such a low watt amp shouldn't a 0-12 V 3 amo transformer be enough?
When the amp is connected. The voltage across Rectifier+ Cap bank = 9 volts. Any idea why is this happening? Why does the voltage drop when the amp is connected. By connected I mean the amp is just connected to the power supply witb no input or speaker output.I assume that for such a low watt amp shouldn't a 0-12 V 3 amo transformer be enough?
How do I measure that? Also can you please tell me the transformer rating that I need to get to make this work.
What voltage and what ampere? I have a guy who makes custom toroids.
Thanks
I have a 74 VA transformer and 8000 uf per rail. I understand that the transformer might be under rated but I guess 8000uf per rail should be good enough.
You just connect an ammeter in series with the DC supply to the amp. Make sure the meter can handle the current (typically you want the 10A DC range).
The recommended DC voltage for 8 ohm operation is around 27 volts DC, so that requires a 20 volt AC winding. Your 12 volt winding will only give around 16 volts DC, which is not enough.
VA rating needs to be around 100va in order to not stress the transformer (assuming running a stereo pair of amps).
Each channel should draw around 1.2 A from the supply. You may need to tweak the bias setting resistors as modern semiconductors will give different results to the original.
The recommended DC voltage for 8 ohm operation is around 27 volts DC, so that requires a 20 volt AC winding. Your 12 volt winding will only give around 16 volts DC, which is not enough.
VA rating needs to be around 100va in order to not stress the transformer (assuming running a stereo pair of amps).
Each channel should draw around 1.2 A from the supply. You may need to tweak the bias setting resistors as modern semiconductors will give different results to the original.
If you want to make a pair of monoblocks (two physical separate amps) then a dual 0-20 and 0-20 transformer (two separate windings) would let you do that. Each winding would need to be around 50va.
If you are running two channels from one transformer winding then you need a 0-20 winding of 100va. Note that it is 20 volts, not 25.
The reason the transformer va rating seems high is because transformers are rated to supply their current into a resistive load, a bridge rectifier and reservoir capacitor draws a high peak current pulse each half cycle and so the transformer has to be derated to account for that.
The JLH is a low power amp but because it is Class A operation it draws a very high current all the time. This is the big downside to this type of design.
If you are running two channels from one transformer winding then you need a 0-20 winding of 100va. Note that it is 20 volts, not 25.
The reason the transformer va rating seems high is because transformers are rated to supply their current into a resistive load, a bridge rectifier and reservoir capacitor draws a high peak current pulse each half cycle and so the transformer has to be derated to account for that.
The JLH is a low power amp but because it is Class A operation it draws a very high current all the time. This is the big downside to this type of design.
As Mooly said, your transformer needs to be closer to 20vac per winding. (20*1.414 - losses). The 25v transformer will give you in the region of 34vdc.
What impedance load will you be driving? (8ohm? 4 ohm?)
The selection of R1, R2, C1 and C2 are dependent on the load you wish to drive. This also determines your power supply voltage and current requirements.
The lower the voltage supply, the higher the current demand from the transformer - 3ohm load wants 2A at 17v, 8ohm wants 1.2A at 27v.
You would also generally want to use a supply whose capacity is a good margin above the power requirements.
I found these notes pretty helpful:
http://diagramas.diagramasde.com/audio/Linsley Hood Class.pdf
What impedance load will you be driving? (8ohm? 4 ohm?)
The selection of R1, R2, C1 and C2 are dependent on the load you wish to drive. This also determines your power supply voltage and current requirements.
The lower the voltage supply, the higher the current demand from the transformer - 3ohm load wants 2A at 17v, 8ohm wants 1.2A at 27v.
You would also generally want to use a supply whose capacity is a good margin above the power requirements.
I found these notes pretty helpful:
http://diagramas.diagramasde.com/audio/Linsley Hood Class.pdf
A 10W Class A amp will need about 25W DC power (and 25W heatsinks). To get 25W DC from a normal PSU you need a 50VA transformer. Stereo needs twice as much, then add a bit of contingency, so you end up with 120VA transformer for a 10W per channel Class A amp. If it were Class B you would be OK with your 70VA transformer.
10-15 ohms, if correct, is about 100 times too much resistance. A choke used in this type of power supply would need to be similar in size to the mains transformer; I'm going to guess that your chokes are quite tiny and intended for other uses.
Have you considered Ohm's Law? It is very helpful when designing a PSU.
Have you considered Ohm's Law? It is very helpful when designing a PSU.
Jlh 1969 Amp transistors overheating
The output transistors on one of the channel are overheating. I guess this must be due to over bias. I haven't set the quiescent current so I guess thats why transistors are overheating. In this scenario what should me the ideal qs current. What value should I set it to ?
What I read from the manual was that the ammeter has to be in series with power supply and amp and then adjust the trimmer to set quiescent current. Does this look correct?
Here is the link to the product. Chinese translated English is hard to comprehend...
10-15W JLH1969 Class A Amplifier AMP Board Left Channel PCB Assembled MOT/2N3055 Sale sold out - Banggood Mobile
The output transistors on one of the channel are overheating. I guess this must be due to over bias. I haven't set the quiescent current so I guess thats why transistors are overheating. In this scenario what should me the ideal qs current. What value should I set it to ?
What I read from the manual was that the ammeter has to be in series with power supply and amp and then adjust the trimmer to set quiescent current. Does this look correct?
Here is the link to the product. Chinese translated English is hard to comprehend...
10-15W JLH1969 Class A Amplifier AMP Board Left Channel PCB Assembled MOT/2N3055 Sale sold out - Banggood Mobile
The heatsink brackets shown here seem to be quite thin - perhaps less than even 3mm and I doubt they will be able to conduct the heat of the power transistors well enough. The amplifier will probably work OK but the transistors may run very hot - even if the fan cooling is good. Heat generation is normal for class A but too much in hot weather can damage the PCB and components. Keep the heatsink near the transistors below 65C in all circumstances from say, 0-40C ambient temperature. Note that this is burning hot to touch and is baking the PCB and nearby components continuously, whilst the amplifier is powered.
I have only assembled a few kits of JLH'69 these but they definitely need more substantial heatsink brackets than most sales illustrations show. With small brackets such as you have, I think a minimum 4mm thickness is necessary for 2 x 2N3055 @ specified 1.2A bias current. 5-6mm thickness will be better.
How to know how much is enough? Measure the temperature of the transistor case and compare with the actual heatsink temperature. If these are way different after allowing temperatures to stabilise, you don't have good thermal coupling. I would aim for no more than 20C difference between transistor case and the heatsink fan tunnel.
I have only assembled a few kits of JLH'69 these but they definitely need more substantial heatsink brackets than most sales illustrations show. With small brackets such as you have, I think a minimum 4mm thickness is necessary for 2 x 2N3055 @ specified 1.2A bias current. 5-6mm thickness will be better.
How to know how much is enough? Measure the temperature of the transistor case and compare with the actual heatsink temperature. If these are way different after allowing temperatures to stabilise, you don't have good thermal coupling. I would aim for no more than 20C difference between transistor case and the heatsink fan tunnel.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.