I have attached photos of my "Marantz Style" MA9S2 which I am pretty happy with. I do need to move it to a new case with a larger heatsink and a larger power supply board, more like the one in the photo of my in progress MX50X2.
I have also attached photos of the MX50X2 and L20.5. They are awaiting assembly in cases but they have been listened to outside of cases previously at lower voltages.
To be honest I am very impressed with all three. The LJM MX50X2 and L20.5 were pre-assembled and as far as I can tell they use good quality components.
The MA9S2 was a kit and the only problem I had was that one LED cracked and thus one of the strings of LEDs did not light up in the input bias section. But I don't think that is the fault of the kit at all.
I have also attached photos of the MX50X2 and L20.5. They are awaiting assembly in cases but they have been listened to outside of cases previously at lower voltages.
To be honest I am very impressed with all three. The LJM MX50X2 and L20.5 were pre-assembled and as far as I can tell they use good quality components.
The MA9S2 was a kit and the only problem I had was that one LED cracked and thus one of the strings of LEDs did not light up in the input bias section. But I don't think that is the fault of the kit at all.
Well it will be a couple of weeks until the parts arrive to finish assembling the MX50X2. (Then I can compare some more.)
But here is my first/early attempt to describe them a little bit sound-wise.
I will start with a bit of a caveat: I started (mainly) with kits that were fairly well established, refined and of good reputation. From my reading and research I gathered that the L20.5 and MX50X2 were two of LJM's best and most highly regarded kits. Thus the caveat is that I am having a hard time finding faults in these amplifiers. Instead I am finding that I am becoming much more aware of the source and recording quality as I try to compare them. While comparing them I have sometimes heard something I did not like and upon replaying that song I realize it is something in the recording that was not so obvious to me before. So it was not the amplifier or other equipment that I was noticing, rather it was the recording.
The L20.5 is what I call very precise and accurate. I think you can expect it to make a very good recording sound as the recording engineer intended. But the other side of this is that an amplifier like the L20.5 is perhaps the opposite of the type that "warms" the sound or adds a desirable amplifier characteristic or signature to the sound. Please don't misunderstand what I am trying to say. I think (so far) that the L20.5 is an outstanding design and outstandingly accurate. But others might call that "cold". Perhaps more knowledgeable forum members can comment but perhaps this is the amplifier type for powering recording studio monitors and for listeners who prefer that type of precision and accuracy. I can not really remark on any shortcomings yet.
I don't find it fatiguing to listen to but if I don't particularly like the recording quality I would be tempted to switch the CD out sooner. (If you follow what I am trying to say.)
Take this with a grain of salt since I would like to listen much more before making any strong statements.
I understand that many of LJM's designs have similarities but without a schematic for the L20.5 I can not comment much more. I know the output stage of the MX50X2 differs from the L20.5 however I would will need to listen to them much more to try to point out or describe a difference in the sound as a result. My initial impressions are that they are both really very nice kits.
With the MA9S2 I did not seem to notice the shortcomings of the source recording as often. I have listened to the MA9S2 the most and I think it is my current favorite however I will need to spend more time on the finished MX50x2 and L20.5 projects before I should reach a solid conclusion. That will take more time and then I plan to try building a second set of my chosen favorite kit so that one chassis (with two amps) can drive the left front speaker bi-amp and the second chassis (with two amps) can drive the right front speaker bi-amp. At least that is one of my possible next experiments that I am considering.
For these projects I tried to read and research quite a bit upfront so that I would only pick kits that others had very highly rated. So in a fashion I am not surprised that I am having difficulty finds faults to comment on.
I am hoping others can add their experiences of the L20.5, MX50x2 and especially the MA9S2.
But here is my first/early attempt to describe them a little bit sound-wise.
I will start with a bit of a caveat: I started (mainly) with kits that were fairly well established, refined and of good reputation. From my reading and research I gathered that the L20.5 and MX50X2 were two of LJM's best and most highly regarded kits. Thus the caveat is that I am having a hard time finding faults in these amplifiers. Instead I am finding that I am becoming much more aware of the source and recording quality as I try to compare them. While comparing them I have sometimes heard something I did not like and upon replaying that song I realize it is something in the recording that was not so obvious to me before. So it was not the amplifier or other equipment that I was noticing, rather it was the recording.
The L20.5 is what I call very precise and accurate. I think you can expect it to make a very good recording sound as the recording engineer intended. But the other side of this is that an amplifier like the L20.5 is perhaps the opposite of the type that "warms" the sound or adds a desirable amplifier characteristic or signature to the sound. Please don't misunderstand what I am trying to say. I think (so far) that the L20.5 is an outstanding design and outstandingly accurate. But others might call that "cold". Perhaps more knowledgeable forum members can comment but perhaps this is the amplifier type for powering recording studio monitors and for listeners who prefer that type of precision and accuracy. I can not really remark on any shortcomings yet.
I don't find it fatiguing to listen to but if I don't particularly like the recording quality I would be tempted to switch the CD out sooner. (If you follow what I am trying to say.)
Take this with a grain of salt since I would like to listen much more before making any strong statements.
I understand that many of LJM's designs have similarities but without a schematic for the L20.5 I can not comment much more. I know the output stage of the MX50X2 differs from the L20.5 however I would will need to listen to them much more to try to point out or describe a difference in the sound as a result. My initial impressions are that they are both really very nice kits.
With the MA9S2 I did not seem to notice the shortcomings of the source recording as often. I have listened to the MA9S2 the most and I think it is my current favorite however I will need to spend more time on the finished MX50x2 and L20.5 projects before I should reach a solid conclusion. That will take more time and then I plan to try building a second set of my chosen favorite kit so that one chassis (with two amps) can drive the left front speaker bi-amp and the second chassis (with two amps) can drive the right front speaker bi-amp. At least that is one of my possible next experiments that I am considering.
For these projects I tried to read and research quite a bit upfront so that I would only pick kits that others had very highly rated. So in a fashion I am not surprised that I am having difficulty finds faults to comment on.
I am hoping others can add their experiences of the L20.5, MX50x2 and especially the MA9S2.
For class AB I like the MX50SE with Sanken outputs. Like it better with high Iq at abt 0,3-0,4A. JLM L7 is not bad either, but is said to have reliability issues.
Best sound so far is from a ZeroZone class a JLH PNP (big boards with rectifier & caps). A lot better after some modifications (found in JLH thread). Sounds very nice, but low power and class a heat of course.
MX50x2 I did not like, same with L12-2.
Best sound so far is from a ZeroZone class a JLH PNP (big boards with rectifier & caps). A lot better after some modifications (found in JLH thread). Sounds very nice, but low power and class a heat of course.
MX50x2 I did not like, same with L12-2.
Last edited:
I have the MX50SE but I have not tested it. However it has K1047 outputs. What did you like better about it compared with the MX50X2 and L12?
Also, did you find a good source of the Sanken outputs that you recommend? (Or did you use the outputs that came with the kit?)
My L7 is not in a chassis right now so I guess I will be more careful about the heatsinking and supplies voltages in an attempt to avoid reliability issues. I don't think I push my amps terribly hard due to the combination of speaker efficiency, listening volume and room size.
So of the two photos I attached I assume it is the second one? Or is it a different kit? I assume not the first one which is just the amp.
I only saw the assembled amp for sale. So did you build a kit or start with the assembled board? Did you use the Sanken outputs that came with it? Do you think those Sanken outputs are good/genuine?
Attachments
-Sankens: the ones with the kit, measures better than any other outputs I tried (MX50 & JLH), and sounds a bit 'smoother' in the MX50
-L7, you can try higher Iq on this one too, I think it improved sound and measurements a bit
-Yes the big JLH boards, and I used the supplied transistors. Modifications are posted in the JLH tread if you can find it..: floating ground, and changed some component values on resistors and caps.
Using 2pcs 19,5V 7A rated HP laptop bricks for supply (dual mono is needed with floating GND). Those bricks didn't mind big caps on startup, and supply a pretty clean voltage. Traces can be cut and the capacitors supplied with the kit can be used for the floating ground setup (saves money and time).
Safety ground can't be connected in the bricks either.. (at your own risk)
-L7, you can try higher Iq on this one too, I think it improved sound and measurements a bit
-Yes the big JLH boards, and I used the supplied transistors. Modifications are posted in the JLH tread if you can find it..: floating ground, and changed some component values on resistors and caps.
Using 2pcs 19,5V 7A rated HP laptop bricks for supply (dual mono is needed with floating GND). Those bricks didn't mind big caps on startup, and supply a pretty clean voltage. Traces can be cut and the capacitors supplied with the kit can be used for the floating ground setup (saves money and time).
Safety ground can't be connected in the bricks either.. (at your own risk)
Are those regarded as the best JLH kits to try? I have seen them and they are on my list but I am trying to decide between those and a couple of other versions.
For example, there is the JLH 2003 with no output cap, four output transistors and two current sources. (Photo and schematic attached.)
Then I am also considering the two versions that have a regulator built-in and use the 5200 outputs. (Photos attached.)
Have you tried the other types? For this project I would like to get just the "better"/"recommended" JLH instead of ending up with an large assortment of JLH amplifiers like I have with the LJM kits.
Attachments
I have build the first one shown, from bare pcb, and with other output transistors 2sa1117/2sc2608, I'm really surprised how good they sound, but it demands at least 2 kilo heatsink and a clean dc supply.
I think it's called jlh 2003 version
I think it's called jlh 2003 version
I have 12 of each from an unused prototype amp that never made the market, from late 80s so I'm sure they are genuine
Unless you are committed to TO-3's there are many current BJT's that could be considered instead of the 1117/2608 that have superior parameters (Gain/Vce/ Ic / Price/ Reliability) etc: like this selection. I filtered out anything with gain lower than the originals, as I feel that is an important part of the selection of those trannies.
Attachments
If you want the extra 2amps Ic parameter, ON still makes the MJ21195/6 that will work but gain drops to 25 instead of 30 of the Sankens at 8amps
I tried few of them.
AIYIMA 100W 2SC5200+2SA1943
SENGTERBELLE Assembly LM3886 DC Servo 68W+68W
This amplifier DIY KIT LJM L6 has really good sound, I decided to make better PCB with ground plane. So I will have more fun with it when PCB will arrive. I just don't know why it has FERRITE coil? I plan to change it with air coil.
AIYIMA 100W 2SC5200+2SA1943
SENGTERBELLE Assembly LM3886 DC Servo 68W+68W
This amplifier DIY KIT LJM L6 has really good sound, I decided to make better PCB with ground plane. So I will have more fun with it when PCB will arrive. I just don't know why it has FERRITE coil? I plan to change it with air coil.
What was the difference you heard between the MX50SE and the MX50x2 (or L12-2)?
What do you attribute the difference to? The Sanken outputs versus the KEC KTD1047?
Do you think the MX50x2 with Sanken outputs might be good or is there something more to the differences?