Perhaps go through the entire board and check that the resistor values are correct and are in the right locations.
Whether you bought it assembled or assembled it yourself perhaps it is work a double check?
I suggest that since some of the color rings look different from an Ebay photo of what looks like the same version of the L7. But it is really hard to see in the photos. The left photo is from Ebay and the right photo is yours. I don't know if there are any changes or updates in the values.
Edit: I just noticed that one has a copyright date of 2017 and the other 2018. So perhaps just different versions with different values.
Attachments
The specs certainly seem to be unrealistically optimistic. I agree, one pair of devices won't work for 56V rails and 4 ohm loads. As this drives the transistors very close to the SOA line, that does not allow for reactive load lines.
Peak current for 300W / 4ohms is 12.2A. Average current 3.9A per side, so input power is 437W. Assuming a resistive load line that leaves 137W of dissipation in the transistors.
That is just at peak output and maximum efficiency. Peak dissipation in the devices can be higher.
Another of my rules of thumb is that the transistor dissipation needs to be based on half-power square waves as a worst case scenario. That means pumping out 28V into 4 ohms, or 200W, and 200W of transistor dissipation. I would anticipate at least 4 devices in parallel aiming to dissipate 50W each(max) for such a design.
With 0.83C/W thermal resistance in the transistors, 100W (each) cannot easily be achieved. That's a big heatsink.
The circuit in the first post also has some errors I think. The PMOS transistor seems upside down and the bias to the gate taken to the wrong supply line?
Not only that but peak currents of 12 are pretty ferocious. It will need thick cables, and could cause problems if the capacitors aren't highly rated. Either from parasitic inductances (in the wiring - large currents generate larger voltages) as well as ripple current.
Peak current for 300W / 4ohms is 12.2A. Average current 3.9A per side, so input power is 437W. Assuming a resistive load line that leaves 137W of dissipation in the transistors.
That is just at peak output and maximum efficiency. Peak dissipation in the devices can be higher.
Another of my rules of thumb is that the transistor dissipation needs to be based on half-power square waves as a worst case scenario. That means pumping out 28V into 4 ohms, or 200W, and 200W of transistor dissipation. I would anticipate at least 4 devices in parallel aiming to dissipate 50W each(max) for such a design.
With 0.83C/W thermal resistance in the transistors, 100W (each) cannot easily be achieved. That's a big heatsink.
The circuit in the first post also has some errors I think. The PMOS transistor seems upside down and the bias to the gate taken to the wrong supply line?
Not only that but peak currents of 12 are pretty ferocious. It will need thick cables, and could cause problems if the capacitors aren't highly rated. Either from parasitic inductances (in the wiring - large currents generate larger voltages) as well as ripple current.
Last edited:
I just took the L 7 out and will not use that ****. Ordered other amp boards from Banggood. With 2 sets Mosfets.
It seems to me like the kits and assemblies being discussed are "knock-offs" - i.e. assemblies by other sellers/suppliers who make cheaper copies, with or without original PCBs. If you doubt this likelihood, I suggest that you follow LJM's comments regarding his other designs and competitors too. It's a big problem for him.
I have managed to buy several LJM original kits (and a few fakes) through the usual global platforms but if you want to buy LJM's kits and assemblies exclusively and so avoid the confusion and deception, you need to deal with LJM's own store at Taobao, the Chinese language selling platform. Unfortunately, you'll also need to translate to/from Simplified Chinese language plus organise shipping and taxes yourself. Good luck with that.
Otherwise, pick and choose your kits and assemblies very carefully. Saving say, 25% by buying from a cheap seller makes no economic sense if what you get is a bunch fake or poor substitute semis, under-rated resistors and diodes, relays etc. which is likely where the 25% saving comes from. Stupid really, because when you've replaced the components with known good ones, your cost just exceeded the genuine product - if you could be sure of sourcing it, that is.
In other words, if you prefer to buy than DIY, at least get the buying part right and don't assume that what think you see in pics is fact. Pics are not specifications; just visual indication or prompts of what we may be looking for. Trained professional buyers have been fooled many times in the electronics business, so what chance does an inexperienced amateur have?
I have managed to buy several LJM original kits (and a few fakes) through the usual global platforms but if you want to buy LJM's kits and assemblies exclusively and so avoid the confusion and deception, you need to deal with LJM's own store at Taobao, the Chinese language selling platform. Unfortunately, you'll also need to translate to/from Simplified Chinese language plus organise shipping and taxes yourself. Good luck with that.
Otherwise, pick and choose your kits and assemblies very carefully. Saving say, 25% by buying from a cheap seller makes no economic sense if what you get is a bunch fake or poor substitute semis, under-rated resistors and diodes, relays etc. which is likely where the 25% saving comes from. Stupid really, because when you've replaced the components with known good ones, your cost just exceeded the genuine product - if you could be sure of sourcing it, that is.
In other words, if you prefer to buy than DIY, at least get the buying part right and don't assume that what think you see in pics is fact. Pics are not specifications; just visual indication or prompts of what we may be looking for. Trained professional buyers have been fooled many times in the electronics business, so what chance does an inexperienced amateur have?
So without adjusting the bias to a higher level, the distortion of 0.3% is high.
Thank you
Hi I'm new here, thanks for accepting , i have a L7 amp board, would this work with +-65 Vdc on a 500w 4 ohm speaker ? ?What would be the rep[lacement for c1815 which cab stand 65V ?? Thx
