Hello rubah.ganteng
On my JLH test
I left it on for 2 hours so I could do the tests.
Power Transistor Temperature: 55ºC !!!
I put big heat sinks, and left them on the outside of the case.
I didn't use forced ventilation.
The heat sink has become warm.
Amplifier features:
JLH 1969 monoblock amplifier (10W) per channel.
Power supply (individual):
Transformer: 110 / 220V • 18V AC • 6A
Bridge rectifier: 8A
Capacitor: 22,000uF (63V)
PURE "CLASS A" Amplifier
Power: 18V DC
I Q. (BIAS) = 2A
Speaker = 6 ohms
Distortion-free maximum power (1kHz) = 7W rms (per channel)
Maximum undistorted signal input (1kHz) = 524mV (-3.5dB)
Maximum signal output without distortion (1kHz) = 6.45V
Temperature in the components:
Amplifier turned on for 2 hours:
Ambient temperature = 25°C
Q1 Transistor 2SC5200 = 55ºC
2SC5200 heatsink = 40°C
Q2 Transistor 2SC5200 = 55ºC
2SC5200 heatsink = 40°C
Bridge rectifier = 51°C
Rectifier bridge heatsink = 39°C
Transformer temperature = 40°C
Class A amplifier.
Circuit of Electronic Engineer John Linsley-Hood 1969.
In the video link I show the tests.
On my JLH test
I left it on for 2 hours so I could do the tests.
Power Transistor Temperature: 55ºC !!!
I put big heat sinks, and left them on the outside of the case.
I didn't use forced ventilation.
The heat sink has become warm.
Amplifier features:
JLH 1969 monoblock amplifier (10W) per channel.
Power supply (individual):
Transformer: 110 / 220V • 18V AC • 6A
Bridge rectifier: 8A
Capacitor: 22,000uF (63V)
PURE "CLASS A" Amplifier
Power: 18V DC
I Q. (BIAS) = 2A
Speaker = 6 ohms
Distortion-free maximum power (1kHz) = 7W rms (per channel)
Maximum undistorted signal input (1kHz) = 524mV (-3.5dB)
Maximum signal output without distortion (1kHz) = 6.45V
Temperature in the components:
Amplifier turned on for 2 hours:
Ambient temperature = 25°C
Q1 Transistor 2SC5200 = 55ºC
2SC5200 heatsink = 40°C
Q2 Transistor 2SC5200 = 55ºC
2SC5200 heatsink = 40°C
Bridge rectifier = 51°C
Rectifier bridge heatsink = 39°C
Transformer temperature = 40°C
Class A amplifier.
Circuit of Electronic Engineer John Linsley-Hood 1969.
In the video link I show the tests.
@sierratds u have explaint clearly 🙏
@amplidude my heatsink too small 😂🙏
I try 1A 🥵35 degrees
Wow Enough for me 😁 Happy burn test
Thanks u!🙏
@amplidude my heatsink too small 😂🙏
I try 1A 🥵35 degrees
Wow Enough for me 😁 Happy burn test
Thanks u!🙏
Hello everyone. I recently bought a jlh1969 to mosfet from aliexpress. Cheap and well made. The problem I have is that when I try to raise the quiescent current by more than 20mA, it goes crazy. The numbers on the multimeter vary a lot as if it had a self-oscillation. If I feed it from a smps source it works fine. Only with classical font does that happen to me. Can somebody help me? Thank you.
Attachments
Hi there,
Which kind of speakers are you mostly driving with the 10W version JLH amplifier? I've been pleased using the small version with tiny heatsinks for headphones or active monitors, but wondering now if I could use the bigger version to drive some passive 2-way speakers.. 10Watt seems little power, since the speakers I've been looking at are around 20-40Watt (ie. Braun L610). They never go below 20W I think. Is there a general use-guideline or forum links I am missing?
Thanks a lot, fellows!
Which kind of speakers are you mostly driving with the 10W version JLH amplifier? I've been pleased using the small version with tiny heatsinks for headphones or active monitors, but wondering now if I could use the bigger version to drive some passive 2-way speakers.. 10Watt seems little power, since the speakers I've been looking at are around 20-40Watt (ie. Braun L610). They never go below 20W I think. Is there a general use-guideline or forum links I am missing?
Thanks a lot, fellows!
Hello cucu70,
I don't know what's going on with your circuit,
mine I mounted it all on an island plate.
I did all the assembly by hand.
I liked the result.
See my final circuit.
I left it with the BIAS current at 2A!!
It's working great.
I made it monobloc.
One power supply for each audio card.
I have 2 now. The system is in stereo.
Others docs:
https://drive.google.com/drive/folders/1hqNqf1LpdyNb26nMPSwo5KOk3z-7M26V
My tests in my JLH 1969
https://www.youtube.com/playlist?list=PLceAdLTPocSHFkuiKAzEND4Blkb3-oGuQ
Hello Sierratds. I have built a few with bjt in the end and they all work great, but this one with a mosfet output is the only one that gives me problems. I'm going to try to feed them separately with linear sources to see what happens. Thanks for your answer. All the best.
Has anyone built the Chinese kit version that uses a long heat sink? I just ordered one that is said to be UK stocked. Be interesting to see how long it takes to arrive. Also quality of the bits.
I intend to use it to drive speakers under test and may try changing a couple of values to decrease the LF roll off. Bootstap and coupling capacitors.
I've wanted to build a speaker for rather a long time so a bit of a bucket list project.
I intend to use it to drive speakers under test and may try changing a couple of values to decrease the LF roll off. Bootstap and coupling capacitors.
I've wanted to build a speaker for rather a long time so a bit of a bucket list project.@cucu - a couple of points. Your MOSFET output devices have a high frequency response. The 220 ohm gate resistors should stop oscillations but it depends on having a good decoupling capacitor on the rails to ground on the PCB. I think the aliexpress units do have this but it is worth checking.
How are you feeding your amplifier? Another source of oscillation - and it does seem like oscillation - is for the input to have a high impedance. It needs a low impedance to avoid phase shifts in the input stages, and screened leads of course. If you feed it from a high impedance potentiometer for balance control or volume control it may be enough to cause trouble. You might try putting a 100pF capacitor across the input to ground if that is the case.
A TIP 42 is not a good driver for the circuit but it has a slow frequency response which should prevent oscillations (built in Miller capacitor).
Thirdly, is your meter showing a low frequency oscillation or high frequency? Can you tell? This is important because if your power supply - and you said that an SMPS does not cause trouble - then perhaps it is too poorly regulated and you are seeing low frequency oscillations. You should try to look at the signal on a scope.
And from the diagram, I'm not sure it is ideal for MOSFETs. They have a high gm-the source resistors will help limit this- but a high value potentiometer the collector of the driver is likely to be quite difficult to set. Better to have a lower value in series with perhaps selectable resistors to be able to get better control.
You may be better off modifying it for bipolar outputs.
How are you feeding your amplifier? Another source of oscillation - and it does seem like oscillation - is for the input to have a high impedance. It needs a low impedance to avoid phase shifts in the input stages, and screened leads of course. If you feed it from a high impedance potentiometer for balance control or volume control it may be enough to cause trouble. You might try putting a 100pF capacitor across the input to ground if that is the case.
A TIP 42 is not a good driver for the circuit but it has a slow frequency response which should prevent oscillations (built in Miller capacitor).
Thirdly, is your meter showing a low frequency oscillation or high frequency? Can you tell? This is important because if your power supply - and you said that an SMPS does not cause trouble - then perhaps it is too poorly regulated and you are seeing low frequency oscillations. You should try to look at the signal on a scope.
And from the diagram, I'm not sure it is ideal for MOSFETs. They have a high gm-the source resistors will help limit this- but a high value potentiometer the collector of the driver is likely to be quite difficult to set. Better to have a lower value in series with perhaps selectable resistors to be able to get better control.
You may be better off modifying it for bipolar outputs.
Hello John Ellis. Yes, the oscillation is low frequency because you can also see the membrane of the loudspeaker how it oscillates. A few hertz I would say. I have used a transformer + bridge rectifier + 15,000uF. When I switch to smps, everything is fine, even with 3A bias on each channel, for a short period, I have small heatsinks and fans. Even if I feed only one channel with classical source, it looks like it oscillates and I can't stabilize the quiescent current. I don't know what to say. At the moment I use it with smps and it works great, but I want to find out what is happening. Thanks for your input.
I have a few 1969s with bjt and they all go very well, I only had problems with this one. I know it's an oscillation, but I don't know where to play. I don't use a potentiometer and if I short the input or not, the same thing happens. I also have some old 2n1711, Korean originals and I plan to put them instead of tip41.
The low frequency oscillation may be caused by having an unstabilised supply, but it could still be caused by high frequency oscillation leading to an instability at LF.
I've never thought the bias arrangement on JLH's input to be optimum. I'm not sure what voltage you are using, but one option might be to pick a Zener diode a little below your nominal (loaded) voltage and use that on the input bias chain. You would need to reduce the resistor feeding it, and re-adjust the bias to obtain mid-rail but that would stabilise the centre rail voltage even if there were some undue ripple on the main supply.
I'm suspecting some interaction between the input filter on the bias chain and the PSU supply. Even if you have a large capacitor, if the current swing is large, you could be causing the PSU voltage to drop, which would upset the bias and reduce the current, leading to a rise in the voltage and so on, a problem once known as "motorboating".
A stabilised supply would be the best solution, as your SMPS shows, but you could try stabilisiing the input at lower cost.
I've never thought the bias arrangement on JLH's input to be optimum. I'm not sure what voltage you are using, but one option might be to pick a Zener diode a little below your nominal (loaded) voltage and use that on the input bias chain. You would need to reduce the resistor feeding it, and re-adjust the bias to obtain mid-rail but that would stabilise the centre rail voltage even if there were some undue ripple on the main supply.
I'm suspecting some interaction between the input filter on the bias chain and the PSU supply. Even if you have a large capacitor, if the current swing is large, you could be causing the PSU voltage to drop, which would upset the bias and reduce the current, leading to a rise in the voltage and so on, a problem once known as "motorboating".
A stabilised supply would be the best solution, as your SMPS shows, but you could try stabilisiing the input at lower cost.
The input and output are the same phase - maybe some 50 or 100hz coupling ot ripple when a transformer etc is used to power it????
He published again about 20years after the initial article. Searching hood class a can find it. From that
The original circuit
The adjustment change a month or so later
And a new design seems it needs a hefty heatsink for 15w 8 ohm
He published again about 20years after the initial article. Searching hood class a can find it. From that
The original circuit
The adjustment change a month or so later
And a new design seems it needs a hefty heatsink for 15w 8 ohm
The MOSFET amplifier has a circuit similar to 1969. But it works differently, in the AB class. The quiescent current is small (recommendation 20-90mA). The operating point is highly dependent on the supply voltage, so it is recommended to use an SMPS or a linear power regulator.
Additional efforts may be required to correct the amplifier.
From another forum:
Additional efforts may be required to correct the amplifier.
From another forum:
Last edited:
Yes, this is well known as the "JLH 96".
The poster is trying to solve a problem with the original circuit. Of course different transistors have to be used - different types have been mentioned throughout this thread. MOSFETs have been used by some including Nelson Pass, and appear to have become popularised by Chinese PCB companies.
"Input and output are in phase" applies to the vast majority of audio amplifiers.
Builders may or may not be aware that output leads should be twisted pairs to minimise inductive loops and input leads need to be screened.
PSU leads should also be twisted pairs for the same reason.
These precautions ought to be standard practice but newbies may not be aware of these principles.
From the problem the OP raised low frequency oscillation can arise because of HF coupling as I mentioned. If speaker cone movement can be seen it suggests rather lower than 50Hz, which is why I suggested it may be due to poor PSU regulation despite a large capacitor.
However, with MOSFETs, the transconductance increases rapidly once the threshold is reached, and if there is HF coupling due to poor wiring then it would give rise to oscillations once the quiescent current is increased beyond about 10mA- which indicates threshold has been reached.
The source resistors should limit the transconductance to a manageable value and provide some thermal stability. In JLH's original circuit, thermal stability was achieved through older BJT's having a lower gain at high temperatures (RCA 2N3055 original) and gain droop (original RCA and more recent transistors) at higher currents. MOSFETs only reach transconductance stability with temperature at a relatively high current, so ballast resistors are more important.
Other layout issues include proper earthing of any heatsinks and wiring the earth/ground leads of PSU, speaker and input to a single common earth point. Unearthed heatsinks can provide a modest capacitive coupling to any other wire nearby, another possible cause of HF oscillation.
LF oscillation could be caused by ground loops between power stages and preamps, or CD players for example, but these, like poor power supply filtering tend to be 50 or 100Hz.
We really need an oscilloscope on the output to understand what is going on with the poster's MOSFET circuit. Until more clues are available, suggestions are to try to pin the root cause down. In simulations, there is a slight overshoot on square wave signals and definite ringing if the input impedance from the preamp is high - a sign that all may not be well.
The poster is trying to solve a problem with the original circuit. Of course different transistors have to be used - different types have been mentioned throughout this thread. MOSFETs have been used by some including Nelson Pass, and appear to have become popularised by Chinese PCB companies.
"Input and output are in phase" applies to the vast majority of audio amplifiers.
Builders may or may not be aware that output leads should be twisted pairs to minimise inductive loops and input leads need to be screened.
PSU leads should also be twisted pairs for the same reason.
These precautions ought to be standard practice but newbies may not be aware of these principles.
From the problem the OP raised low frequency oscillation can arise because of HF coupling as I mentioned. If speaker cone movement can be seen it suggests rather lower than 50Hz, which is why I suggested it may be due to poor PSU regulation despite a large capacitor.
However, with MOSFETs, the transconductance increases rapidly once the threshold is reached, and if there is HF coupling due to poor wiring then it would give rise to oscillations once the quiescent current is increased beyond about 10mA- which indicates threshold has been reached.
The source resistors should limit the transconductance to a manageable value and provide some thermal stability. In JLH's original circuit, thermal stability was achieved through older BJT's having a lower gain at high temperatures (RCA 2N3055 original) and gain droop (original RCA and more recent transistors) at higher currents. MOSFETs only reach transconductance stability with temperature at a relatively high current, so ballast resistors are more important.
Other layout issues include proper earthing of any heatsinks and wiring the earth/ground leads of PSU, speaker and input to a single common earth point. Unearthed heatsinks can provide a modest capacitive coupling to any other wire nearby, another possible cause of HF oscillation.
LF oscillation could be caused by ground loops between power stages and preamps, or CD players for example, but these, like poor power supply filtering tend to be 50 or 100Hz.
We really need an oscilloscope on the output to understand what is going on with the poster's MOSFET circuit. Until more clues are available, suggestions are to try to pin the root cause down. In simulations, there is a slight overshoot on square wave signals and definite ringing if the input impedance from the preamp is high - a sign that all may not be well
.I'm not sure about the thermal behavior of IRFP150. Maybe it is stable enough for this to work. But the main problem with all "JLH" amps is that the upper transistor comes back on as the output swings negative. To fix this, you need a voltage offset between the VAS and the M1 gate. The bootstrap will provide the extra positive swing needed to drive positive peaks. And it seems few have realized that it's important that R8 and R9 are the same or similar because the VAS needs current to drive R9 and we have no ~Baxandal diode. It's all a "terrible hack", but it can work with a lot of tweaking for specific parts. I keep thinking of ways to make it work, but the results are never better than a classic EF circuit. If you want class-A, just use large Re's and bias it higher than class-AB. Large Re's make it thermally stable and are inefficient, but nothing like class-A. I think the appeal of JLH amps is not the lack of distortion, but the presents of significant amount of 2nd order distortion.
Hello Chumingo,
I hope you don't mind me answering your question with a link to an explanation, briefly speaker Watts are meaningless what you need to find out is their db rating. For the full story and the simple maths please go to this page https://jlh1969classa.blogspot.com/p/watts.html it's too much to copy and paste here.
You will find that you are listening to micro Watts!
Cheers -J