Thanks Mooly. I too find the mechanical implementation of an amplifier build to be the most challenging that's why I prefer to use an existing case. It usually has all the proper cut-outs or holes for connectors etc and sometimes re-usable internal components. Maybe it takes recycling to a different level but I prefer to listen to my amplifiers than look at them 
First amplifier build!
Ok, I realise I'm a bit late to the party, but I've just completed my first proper amplifier build, a LJM Quad 405 Clone. I thought I'd just choose a cheap kit to gain some experience in soldering and diagnosing any issues, but it worked straight away and sounds rather nice! (to my ears at least)
Just wondering if anyone can recommend any upgrades to these boards, I see the LM301 op amp is a common part to swap out, but can I just replace this, or do I need to add additional capacitors, etc?
Thanks,
geoff.
Ok, I realise I'm a bit late to the party, but I've just completed my first proper amplifier build, a LJM Quad 405 Clone. I thought I'd just choose a cheap kit to gain some experience in soldering and diagnosing any issues, but it worked straight away and sounds rather nice! (to my ears at least)
Just wondering if anyone can recommend any upgrades to these boards, I see the LM301 op amp is a common part to swap out, but can I just replace this, or do I need to add additional capacitors, etc?
Thanks,
geoff.
Attachments
Hi Geoff,
You have come to the right page if you want to modify your LJM 405 clone. The good news is its built and its working. The bad news is that the majority of mods detailed on Keith Snook's web-page involves removing half those components you just fitted. But less is better right? Look back a few pages at my threads detailing these modifications or jump to Keith's very detailed description of his painstaking but thorough testing and modding to arrive at a much better sounding amplifier. At the moment, you are listening to a 1976 amplifier design with modern equivalent components. After the modification, you will bring that sound right up to the present day without spoiling the Quad 405 characteristic sound. Do all the mods suggested, in particular change input op-amp to a non-inverting configuration and bin that venerable LM301 for a LME49710 or similar. I removed the current limiting circuits, but kept the 91mOhm resistors - piece of shorting wire probably has same resistance so no point in replacing. Enjoy and let us know how you get on
You have come to the right page if you want to modify your LJM 405 clone
. The good news is its built and its working. The bad news is that the majority of mods detailed on Keith Snook's web-page involves removing half those components you just fitted. But less is better right? Look back a few pages at my threads detailing these modifications or jump to Keith's very detailed description of his painstaking but thorough testing and modding to arrive at a much better sounding amplifier. At the moment, you are listening to a 1976 amplifier design with modern equivalent components. After the modification, you will bring that sound right up to the present day without spoiling the Quad 405 characteristic sound. Do all the mods suggested, in particular change input op-amp to a non-inverting configuration and bin that venerable LM301 for a LME49710 or similar. I removed the current limiting circuits, but kept the 91mOhm resistors - piece of shorting wire probably has same resistance so no point in replacing. Enjoy and let us know how you get onThanks Roclite!
I've seen Keith Snook's pages on this, and also Bernd Ludwig has some useful info. There's a lot for a beginner to take in! I was hoping to start off by replacing the lm301 with either a opa134 or opa604, and then look at changing the input gain. That should keep me busy for a while
Is there anything else you'd recommend?
Thanks,
geoff
I've seen Keith Snook's pages on this, and also Bernd Ludwig has some useful info. There's a lot for a beginner to take in! I was hoping to start off by replacing the lm301 with either a opa134 or opa604, and then look at changing the input gain. That should keep me busy for a while
Is there anything else you'd recommend?
Thanks,
geoff
Yes, exactly NOT what you are suggesting. Rolling designer op-amps into the present inverting op-amp configuration will ...er...gain you very little, if anything. I'm afraid its no pain, no gain here! Buy yourself a mini craft knife and happy hacking. I'm getting pretty good with a Dremel for cutting tracks and drilling component holes (kinder on my fingers too), it doesn't have to look pretty as long as it works
. I'm getting pretty good with a Dremel for cutting tracks and drilling component holes (kinder on my fingers too), it doesn't have to look pretty as long as it worksOk, good to know! I'll do some more reading first I think, taking my dremel to the board seems like a very one-way road...
Do you have any more close-up picture of how your boards look now? I've seen the top pics in post #52, just curious what the underside looks like after the modifications
Thanks,
geoff
Do you have any more close-up picture of how your boards look now? I've seen the top pics in post #52, just curious what the underside looks like after the modifications
Thanks,
geoff
As it happens, I took pics in the unlikely event that I would ever have to do it again. The resolution should be good enough for you to relate all the mods to Keith Snook DCD Mod-3 sheet. Because it may (or may not) help you, I have attached my "crib-sheet" schematic to red-line the changes. I went further than Keith's and others recommended changes in certain places. For example, I added significantly more filtering to the op-amp +/- 15V supplies (22uF/0.1uF bulk decoupling + 4u7 25V surface mount cap soldered directly on the op-amp pins 7 and 4) but you don't have to. The benefit of the decoupling caps (apart from less noise/better filtering) is that you can use the same 3K3 dropper resistors as supplied with virtually any replacement op-amp. I also added 2 x 1N4007 speaker inductive fly-back protection diodes to the output. Pretty standard simple circuit. I also put 100uF 100V bulk decoupling on the underside (brown caps) - this makes an audible difference so recommended, especially if you have long cables from your PSU. The 1976 version didn't have these either
Make sure you get the 2SC2547 low noise transistor, its in the signal path so it matters, the current source can just use BC556B if you don't want to splash too much cash. I re-used as many of the original components as possible and just re-configured. I replaced the original Chinese 0.68uF orange blobs with a decent Euro MKT 63V. Sensitivity is set for 1.5V RMS by changing R4/C2. You can see what a really neat job I made with that Dremel. My top tip is to drill holes (0.6 - 0.7 mm drill) for new components near to where they need to be and use the component legs as the connectors to the new connection points. It saves on unsightly, messy post-wiring and it anchors the component to the board. Don't think I can give you much more help that this, but post if you really get stuck on something. BTW, I bought the TO3 transistor version but you have the plastic transistor version. They are almost exactly the same apart from the output stage and I think LJM put a proper power GROUND tag on the board. The TO3 version power GROUND was a bolt through the heatsink.
Make sure you get the 2SC2547 low noise transistor, its in the signal path so it matters, the current source can just use BC556B if you don't want to splash too much cash. I re-used as many of the original components as possible and just re-configured. I replaced the original Chinese 0.68uF orange blobs with a decent Euro MKT 63V. Sensitivity is set for 1.5V RMS by changing R4/C2. You can see what a really neat job I made with that Dremel
. My top tip is to drill holes (0.6 - 0.7 mm drill) for new components near to where they need to be and use the component legs as the connectors to the new connection points. It saves on unsightly, messy post-wiring and it anchors the component to the board. Don't think I can give you much more help that this, but post if you really get stuck on something. BTW, I bought the TO3 transistor version but you have the plastic transistor version. They are almost exactly the same apart from the output stage and I think LJM put a proper power GROUND tag on the board. The TO3 version power GROUND was a bolt through the heatsink.Attachments
No worries. Next time I fire my Quad 405 up, I am going to change the OPA604 op-amp for the much better performance LME47910. I found a really good non-china source that appear to be genuine parts - £7 including postage for 5 off!. At that price (normally £5+ each), I will take the risk. They certainly sound like the real deal, will make the original LM301 sound like a noise generator
Nice, I'd be interested how much difference the LME47910 will make.
I think I have a more fundamental problem to solve now, I'm definitely noticing some mains hum. I'll swap the input cables over for shielded ones, hopefully that will help. Did you have any hum/ground/earthing issues on your build?
I think I have a more fundamental problem to solve now, I'm definitely noticing some mains hum. I'll swap the input cables over for shielded ones, hopefully that will help. Did you have any hum/ground/earthing issues on your build?
Please re-think!
Just looked closely at your construction. There are one or two potential problem areas that might give you hum if it is not just a poor signal connection. For starters, you bring the secondary AC power across from the transformer to your capacitor/rectifier modules, past the power amp modules. Those cables are carrying huge AC signals (approx. 35 - 40 VAC) and must be radiating mains hum like mad. This is not good engineering practice. Look ay my power supply layout on my 405 chassis. All the AC power wires are kept to the minimum distance and all concentrated behind the smoothing capacitors. The wiring is zoned i.e. mains/transformer out/regulated DC - always keep them apart. You would be much better to install your PSU modules immediately next to the transformer using minimum length connections, and take the regulated smooth DC past the heatsink to the amplifier modules. That way you use your heatsink as a free screen. BTW, the LJM 405 clones had a Zobel network ground error - you can see my correction as the blue wire which goes beneath the smoothing caps to connect the 10R of the Zobel network to POWER GND not signal ground as it was on the original 1976 version. The input signal on 405 has a 10R ground lift resistor, this should prevent any hum at all.
You need to check your earthing, you don't appear to a have a star-point ground - that's more likely to be your problem.
Just looked closely at your construction. There are one or two potential problem areas that might give you hum if it is not just a poor signal connection. For starters, you bring the secondary AC power across from the transformer to your capacitor/rectifier modules, past the power amp modules. Those cables are carrying huge AC signals (approx. 35 - 40 VAC) and must be radiating mains hum like mad. This is not good engineering practice. Look ay my power supply layout on my 405 chassis. All the AC power wires are kept to the minimum distance and all concentrated behind the smoothing capacitors. The wiring is zoned i.e. mains/transformer out/regulated DC - always keep them apart. You would be much better to install your PSU modules immediately next to the transformer using minimum length connections, and take the regulated smooth DC past the heatsink to the amplifier modules. That way you use your heatsink as a free screen
. BTW, the LJM 405 clones had a Zobel network ground error - you can see my correction as the blue wire which goes beneath the smoothing caps to connect the 10R of the Zobel network to POWER GND not signal ground as it was on the original 1976 version. The input signal on 405 has a 10R ground lift resistor, this should prevent any hum at all.You need to check your earthing, you don't appear to a have a star-point ground - that's more likely to be your problem.
Hi Roclite, thanks again for the advice.
I had to take a few compromises with the layout for the amplifier chassis and components I was re-using, I realise it's not an ideal layout, but just wanted to get something up and running to gain some experience.
What should be connected to the star-point ground? I've connected the chassis to mains earth, but all of the other connections are from the power ground from the transformer.
I'll look into adding the blue wire to the input, I missed that on the build.
Sorry for all the dumb questions, still learning as you can tell
Thanks,
geoff
I had to take a few compromises with the layout for the amplifier chassis and components I was re-using, I realise it's not an ideal layout, but just wanted to get something up and running to gain some experience.
What should be connected to the star-point ground? I've connected the chassis to mains earth, but all of the other connections are from the power ground from the transformer.
I'll look into adding the blue wire to the input, I missed that on the build.
Sorry for all the dumb questions, still learning as you can tell
Thanks,
geoff
I understand you are learning. But you have the opportunity to make a great sounding amplifier with very little additional work and it will enhance your understanding for your next project. Part of my recommendation to re-site your PSU modules also takes better care of your star-grounding, with minimum length wiring. Always bear in mind Volts = current x resistance when you think of wiring - minimise the resistance to reduce your volt drops. This is particularly important for earthing and grounding. If you re-site your PSU modules, you can put your star-point stud or post (not connected to chassis) near your transformer and PSU. Since you have gone to the trouble of dual PSU's, you probably want to keep your star-point grounds separate, one for each channel. You can probably lose the PCB boards on the transformer and wire direct to the terminals - that will provide enough space for you to stack the PSU boards on top of each other with suitable spacers. Once again, go back to my previous chassis layout - you can clearly see the star-point brass bolt/nut mounted in plastic on top of a metal chassis bolt. I used a single star-point here. Connections to this point are:
1) PSU capacitors 0V
2) Transformer centre-tap
3) Speaker Return 0V
4) PSU HV Filter Caps 0V
5) Zobel Network GND
6) Heatsink Metalwork
For safety, you need to connect your chassis EARTH to the star-point using a GROUND BREAKER circuit (see Rod Elliot/ESP) for details. You can make one up using a standard 30A bridge rectifier/10R 5W/0.1uF 250V.
Its up to you to do what you feel safe and comfortable with. But any compromises you make during construction may have a significant impact on the final sound you achieve.
1) PSU capacitors 0V
2) Transformer centre-tap
3) Speaker Return 0V
4) PSU HV Filter Caps 0V
5) Zobel Network GND
6) Heatsink Metalwork
For safety, you need to connect your chassis EARTH to the star-point using a GROUND BREAKER circuit (see Rod Elliot/ESP) for details. You can make one up using a standard 30A bridge rectifier/10R 5W/0.1uF 250V.
Its up to you to do what you feel safe and comfortable with. But any compromises you make during construction may have a significant impact on the final sound you achieve.
Very confused...
Thought I should check, before blowing up amps or myself...
I'm now very confused where this wire should go to correct the zobel grounding error. According to the pcb diagram I received from the seller, there should be a link (labelled JUMP), which I've shown as yellow on the pcb. This seems to link VCC down to D4/R25/R28, which seems very wrong to me
I think the red line is what I need to go from R39 to power ground, and to either remove the leg or cut the pcb?
Thanks,
geoff
Thought I should check, before blowing up amps or myself...
I'm now very confused where this wire should go to correct the zobel grounding error. According to the pcb diagram I received from the seller, there should be a link (labelled JUMP), which I've shown as yellow on the pcb. This seems to link VCC down to D4/R25/R28, which seems very wrong to me
I think the red line is what I need to go from R39 to power ground, and to either remove the leg or cut the pcb?
Thanks,
geoff
Attachments
Electric shock!
Not sure what I did, but I put the boards back in the chassis without making any modifications
Powered on, and noticed a lot of distortion on the right channel, opened up the case and got a huge electric shock from the heatsink.
To say I'm a bit shaken up is an understatement!
Need to stop this until I can understand what is going on.
Not sure what I did, but I put the boards back in the chassis without making any modifications
Powered on, and noticed a lot of distortion on the right channel, opened up the case and got a huge electric shock from the heatsink.
To say I'm a bit shaken up is an understatement!
Need to stop this until I can understand what is going on.
Oh dear, looks like my advice about the safety GROUND BREAKER circuit came a bit too late. Good that your OK, if not a bit shaken. DONT power on again until you track down the fault. What did you last change/do? Take a DVM/multimeter and check no connection between mains Live/Neutral and your heatsink or any other part of your circuit. Your heatsink needs to be connected to chassis EARTH (safest) or PSU 0V with a ground-breaker circuit fitted. That means the transistors need insulating washers. Was it a DC shock or an AC shock? Has the insulation on your transistors failed? Try to re-build but use your meter to check for isolation where there should be isolation and connection where there should be. For the safety earth to work, there must be a low impedance path from every protected circuit back to EARTH pin on the mains input connector. I wouldn't bother with your other issues until you restore the amplifier back to a safe, protected condition.
Thanks roclite. I really can't explain what happened, I took the boards out of the case, changed my mind about the additional wire and put the boards back. My best guess is that there is some failure in the insulation between the transistors to the heatsink. Check with dvm for continuity, and all seems fine now
The heatsink is connected to mains earth, but indeed I have no connection between DC 0v and earth. The shock happened a bi fast, so not sure if it was ac or dc, mains or after transformer. If I had to pick, I'd say AC after transformer
Thanks for you advice, I'm learning from my mistakes - especially the painful ones!
The heatsink is connected to mains earth, but indeed I have no connection between DC 0v and earth. The shock happened a bi fast, so not sure if it was ac or dc, mains or after transformer. If I had to pick, I'd say AC after transformer
Thanks for you advice, I'm learning from my mistakes - especially the painful ones!
