Somewhat off topic
Hi Tom,
I agree that line level and crossover component choices don't always translate, but in this project, time is tight so I will choose based on the C13 experiment. The crossover is an electrical 2nd order (acoustic 4th order LR). What I have chosen to do is use the Soni-cap Gen 1 with a bypass (TBD) for the bass driver and a 90%/10% combo of 940c/942c for the tweeter. The hope is that the 942c will add some dynamics to the 940c. I would have gone with all 942c but the largest 942c is 2.7 uF, so it's kind of small for crossover use.
Your project to build a line level passive crossover in front of the My Ref is very interesting. I agree with you that there aren't great options with the small value caps, especially at low cost. I am trying to figure out small caps of similar sizes for another project.
I have tried Amtrans AMCH, both on the FE and in another project and like them, however, it's hard to say how they would work in your application.
If you are willing to play on ebay, ERO KP1834 are available as 4.7 nF NOS from this reliable seller;
ERO KP1834 KP1838, ERO Film Capacitors items in 4.7 nF store on eBay!
Both Dario and I have ordered with success from this guy in the past. These old ERO caps are very nice film/foil caps that I have used both as signal caps and for bypassing crossover caps.
For your 33 nF, I have never tried it, but the Amtrans AMCO is a mylar cap that some people have used with success as a coupling cap.
I have recently purchased, but not yet tried, some NOS Siemens and EMZ film/foil polypropylene caps in 33 nF. I will have some extras, so I could send you a few if you would like. Send me a PM if interested.
Jac
Hi Tom,
I agree that line level and crossover component choices don't always translate, but in this project, time is tight so I will choose based on the C13 experiment. The crossover is an electrical 2nd order (acoustic 4th order LR). What I have chosen to do is use the Soni-cap Gen 1 with a bypass (TBD) for the bass driver and a 90%/10% combo of 940c/942c for the tweeter. The hope is that the 942c will add some dynamics to the 940c. I would have gone with all 942c but the largest 942c is 2.7 uF, so it's kind of small for crossover use.
Your project to build a line level passive crossover in front of the My Ref is very interesting. I agree with you that there aren't great options with the small value caps, especially at low cost. I am trying to figure out small caps of similar sizes for another project.
I have tried Amtrans AMCH, both on the FE and in another project and like them, however, it's hard to say how they would work in your application.
If you are willing to play on ebay, ERO KP1834 are available as 4.7 nF NOS from this reliable seller;
ERO KP1834 KP1838, ERO Film Capacitors items in 4.7 nF store on eBay!
Both Dario and I have ordered with success from this guy in the past. These old ERO caps are very nice film/foil caps that I have used both as signal caps and for bypassing crossover caps.
For your 33 nF, I have never tried it, but the Amtrans AMCO is a mylar cap that some people have used with success as a coupling cap.
I have recently purchased, but not yet tried, some NOS Siemens and EMZ film/foil polypropylene caps in 33 nF. I will have some extras, so I could send you a few if you would like. Send me a PM if interested.
Jac
problem
I long fight with a pernicious problem on a sheet of My Ref . While listening , after a series of "scratch " from the speaker , I does not hear anything , if I bypass the protections , the woofer streets sucked into the basket ( DC? ) . I'm now at the fourth replacement LM3886 . I repeated the welds but no result at all. The only thing on my mind is that during welding C34 perhaps has overheated . It could be a problem of self-oscillation ,
or something else?
Thank you.
I long fight with a pernicious problem on a sheet of My Ref . While listening , after a series of "scratch " from the speaker , I does not hear anything , if I bypass the protections , the woofer streets sucked into the basket ( DC? ) . I'm now at the fourth replacement LM3886 . I repeated the welds but no result at all. The only thing on my mind is that during welding C34 perhaps has overheated . It could be a problem of self-oscillation ,
or something else?
Thank you.
Long time ago had a similar problem with a defective opamp and outputing Dc. Use a multimeter do not destroy your speakers
Sent from my Nexus 5 using Tapatalk
Sent from my Nexus 5 using Tapatalk
Hi.
The FE BOM specifies lm318M. I'm seing lm318D at our local RS components website. It's actually more expensive than the "M" version. Any recommendation, which is better?
Thanks.
Yes, this is going to be a lot more expensive than my previous best effort myref. I'm going to try to stick with Dario's BOM as much as possible.
The FE BOM specifies lm318M. I'm seing lm318D at our local RS components website. It's actually more expensive than the "M" version. Any recommendation, which is better?
Thanks.
Yes, this is going to be a lot more expensive than my previous best effort myref. I'm going to try to stick with Dario's BOM as much as possible.
I can't find the post, but somewhere along the line, Dario answered that question. The LM318M is the part that Mauro originally used to optimize the MyRef design. Apparently, the 'M' has some characteristics that are both different from other LM318 builds and were useful to Mauro. Therefore, the 'M' has been preferred for all of the MyRef variants. I'm not familiar with the 'D' version, but you are safer with the 'M'.
Jac
Jac
As Jac already pointed out the recommended part is the Natsemi (now also TI) one (LM318M) vs the TI one (LM318D).
Mauro Penasa stated that the TI part is slightly different and since the entire circuit was designed around the original Natsemi he discouraged the use of the TI part.
At the time I've tried both and both works but the Natsemi part simply sounds better (fuller).
By the way, what is the recommended transformer size for a one channel( mono) setup?
is 20-0V, 20-0V 2.5amperes okay?
I remember reading in the AKSA thread that too big a transformer is not good. Does that apply to myref as well?
is 20-0V, 20-0V 2.5amperes okay?
I remember reading in the AKSA thread that too big a transformer is not good. Does that apply to myref as well?
Actually, that is lower in both voltage and current than recommended. For a toroid transformer and an 8 ohm speaker load, 24-0-24V is recommended. For a 4 ohm load, 22-0-22V. The VA rating is recommended to be 160 to 300. Dario and others have used R-core transformers which are more efficient, so they use 120 VA, if I remember correctly.
I'm guessing that the transformer you mentioned would work, but with a lower amplifier output power and less headroom.
As for the AKSA comment, I find that very interesting. I am always curious about the differences between amplifier designs.
As for the difference between 160 and 300 VA, Bob made a comparison back in the "fine tuning" days. He felt that the bigger transformer was an advantage giving it a more solid bass feel by my memory. Many others have used 160 - 180 VA transformers with good success, so using a bigger transformer is an advantage for this design, but probably fine tuning.
I hope that helps.
Jac
Last edited:
The general rule that works for domestic listening is to use a transformer VA rating that equals or upto double the maximum total output power of the connected amplifiers.
For a 30W maximum output power the amplifier will work with a VA from 30VA to 60VA.
BUT very small transformers have a very high transformer regulation.
Using such a small transformer as shown above (30 to 60VA) will result in a very high supply rail voltage when the amplifier is passing no output current.
Then the supply rails will sag badly when the load demands output current. This is audibly noticeable.
I recommend that for Power Amplifiers that you adopt 160VA, or higher, to minimise the sag under load.
A transformer that to too big cannot damage the amplifier.
But make sure you know what voltage and VA you need to order.
For a 30W maximum output power the amplifier will work with a VA from 30VA to 60VA.
BUT very small transformers have a very high transformer regulation.
Using such a small transformer as shown above (30 to 60VA) will result in a very high supply rail voltage when the amplifier is passing no output current.
Then the supply rails will sag badly when the load demands output current. This is audibly noticeable.
I recommend that for Power Amplifiers that you adopt 160VA, or higher, to minimise the sag under load.
A transformer that to too big cannot damage the amplifier.
But make sure you know what voltage and VA you need to order.
Andrew,
I will defer to Dario on this question. I have been a toroid and EI users so far, so I haven't researched this topic personally.
Jac
It is generally accepted that EI are less efficient than toroid.
But I suspect there are design decisions that can be taken that could make a highly efficient EI more efficient than a low efficiency toroid.
The same cannot be said for differences between Rcore and toroid. They are basically the same in that the core is a continuous strip without any joins.
Q.
could the longer core of an Rcore waste more energy and thus be less efficient?
But I suspect there are design decisions that can be taken that could make a highly efficient EI more efficient than a low efficiency toroid.
The same cannot be said for differences between Rcore and toroid. They are basically the same in that the core is a continuous strip without any joins.
Q.
could the longer core of an Rcore waste more energy and thus be less efficient?
Hi i just finished building this amp, I got one mono block working the other got a bad on semi diode and it started burning once turn one.
I was wondering is it recommended to have heat sink installed for all the diode and is it normal that the amp takes a long time to operate (likely 3 -5 minute)
I was wondering is it recommended to have heat sink installed for all the diode and is it normal that the amp takes a long time to operate (likely 3 -5 minute)
Not needed to heatsink the diodes. Not normal to take minutes to use the amp.
Sent from my Nexus 5 using Tapatalk
Sent from my Nexus 5 using Tapatalk
Thanks , for the reply
Anything I can do to trouble shoot the problem. The strangest thing is the the board behaves like that
circuit breaker sometimes switch on instantly (led lights up),
sometimes it takes a long time for the circuit breaker to switch on instantly (led lights up),
sometimes it turn on when i switched off power.
even if the led does not light up the LM3886 still becomes very hot.
I am beginning is that due to the DC speaker protection function at work ?
Last edited:
Hi sagitis,
You definitely have something very wrong. The relay and LED normally have a delay of about 1 second. You are correct, the LED and relay are part of the DC speaker protection circuit. If the relay and LED don't come on, it is usually due to DC at the output of the amp.
In normal operation, the LM3886 doesn't get very warm at all on even the simplest of heat sinks. Have you shorted the input and measured the voltage at the output? In a properly functioning FE, it should be a few mV.
Please take some close up photos of both sides of the boards and attach them to this thread. Maybe one of use will notice something.
Jac
You definitely have something very wrong. The relay and LED normally have a delay of about 1 second. You are correct, the LED and relay are part of the DC speaker protection circuit. If the relay and LED don't come on, it is usually due to DC at the output of the amp.
In normal operation, the LM3886 doesn't get very warm at all on even the simplest of heat sinks. Have you shorted the input and measured the voltage at the output? In a properly functioning FE, it should be a few mV.
Please take some close up photos of both sides of the boards and attach them to this thread. Maybe one of use will notice something.
Jac
You made the assertion that Rcore are more efficient.
Are you going to reply?