Thanks Clave. Can I get an idea of the price difference? OR, is there a list somewhere (tried to find one) which has the BOM for the Myref so I can price it locally (or internationally) on parts? If I know the price difference I'd know. I have a budget, but it isn't really set since there isn't a definite deadline on this project - could be two weeks, could be three months if I justify it as worth it.
In terms of audio performance - probably the SureElectronics TK2020 will blow my socks off, but I want to build something "simpler" than Class D, and from almost scratch - buy some PCB's, buy the parts, build, build into enclosure etc.
In terms of audio performance - probably the SureElectronics TK2020 will blow my socks off, but I want to build something "simpler" than Class D, and from almost scratch - buy some PCB's, buy the parts, build, build into enclosure etc.
It really depends on parts quality (some audiograde parts are insanely priced) and on project complexity.
A Gainclone has a low part counts and can cost from 70$ and up (you can go cheaper on eBay but it's a different design).
A My_Ref cost more (more parts) can be built from 120$ and up
A My_Ref Fremen Edition (my variation) cost start from 200$ and up
This is a basic comparison for amplifier modules, you should add transformers, connectors, chassis.
If you consider the amp as a whole the price difference is much lower.
A My_Ref can easily blow a TK2020.... IMHO
And it's more fun to build too.
I'll probably be drummed out of the "MyRef Society" , but there is one non-LM3886 kit that could be a great match for a laptop video editing desk. It drastically surpasses a "gainclone" and can hold it's own against some of the early MyRefs. When you see the price you'll know why I took the risk of being blackballed by my fellow LM3886ers
TDA2794
There are several eBay vendors that sell the identical kit. I consider it the value leader with an amazing price performance ratio.
No doubt a MR V1.3 would be the top choice, but it may be an overshoot for what you describe.
, but there is one non-LM3886 kit that could be a great match for a laptop video editing desk. It drastically surpasses a "gainclone" and can hold it's own against some of the early MyRefs. When you see the price you'll know why I took the risk of being blackballed by my fellow LM3886ers TDA2794
There are several eBay vendors that sell the identical kit. I consider it the value leader with an amazing price performance ratio.
No doubt a MR V1.3 would be the top choice, but it may be an overshoot for what you describe.
When you see the price you'll know why I took the risk of being blackballed by my fellow LM3886ers
Bob,
to do a fair comparison you should add the price of a PS, like this one.
The total price is similar to the stereo chipamp.com Gainclone kit.
Bcmbob....just when I thought I could make a decision A short list of three options is always a good idea, I'll into them although the chip sounds familiar from the last couple of days searching and staring and planning.
Maybe I should build a cheap desktop and start building a RefC for a cool living room setup.....STOP! Now I'm getting ahead of myself again.
Thanks again. I'll now build the amps on paper to get and idea of costing. :thumbup:
A short list of three options is always a good idea, I'll into them although the chip sounds familiar from the last couple of days searching and staring and planning. Maybe I should build a cheap desktop and start building a RefC for a cool living room setup.....STOP! Now I'm getting ahead of myself again.
Thanks again. I'll now build the amps on paper to get and idea of costing. :thumbup:
Inp,
whether it's a chipamp from any manufacturer does not matter.
The amplifier will take about 10% of your Completed Amplifier budget.
The case and heatsink and transformer and PSU and hardware and wiring will take up the other 90% and ALL of this can be used with any chipamp.
Build a simple single channel first. Get it working, listen for a while in your near field duty and decide from there what you need for your full two channel amplifier.
You are now armed with most of the information to decide whether spending another 10% for a second matching simple chipamp, or spend double that for each of two MyRef (a total of an extra 40%) or ten times that for each fully upgraded FE (a total of an extra 200%).
The budget goes absolutely haywire when you go fully upgraded. If you then add on the cost of expensive components you want to try swapping and discard, then budget no longer exists, it's very deep pockets.
whether it's a chipamp from any manufacturer does not matter.
The amplifier will take about 10% of your Completed Amplifier budget.
The case and heatsink and transformer and PSU and hardware and wiring will take up the other 90% and ALL of this can be used with any chipamp.
Build a simple single channel first. Get it working, listen for a while in your near field duty and decide from there what you need for your full two channel amplifier.
You are now armed with most of the information to decide whether spending another 10% for a second matching simple chipamp, or spend double that for each of two MyRef (a total of an extra 40%) or ten times that for each fully upgraded FE (a total of an extra 200%).
The budget goes absolutely haywire when you go fully upgraded. If you then add on the cost of expensive components you want to try swapping and discard, then budget no longer exists, it's very deep pockets.
I'm 8 years late to the party, but I must commend the OP on their effort.
I just realised yesterday that the circuit I'm hashing out on paper has some similarities to this circuit. I'm working on a Howland current pump employing the LM1875 and buffered feedback. I'm just trying to see where it's taking me. It works "on paper" but a prototype is in order. My circuit also includes simple latch up protection and DC nulling.
I also thought of a potential improvement to this circuit. Since the whole idea of employing an op amp input is to benefit from a more linear input stage, how about using the input stage of Doug Self's "blameless" amplifier instead of an op amp? Linearity of the input stage is crucial, because the input stage is an error amplifier; and any errors the input stage makes will translate into distortion. I'm interested in what the community thinks of this idea.
I just realised yesterday that the circuit I'm hashing out on paper has some similarities to this circuit. I'm working on a Howland current pump employing the LM1875 and buffered feedback. I'm just trying to see where it's taking me. It works "on paper" but a prototype is in order. My circuit also includes simple latch up protection and DC nulling.
I also thought of a potential improvement to this circuit. Since the whole idea of employing an op amp input is to benefit from a more linear input stage, how about using the input stage of Doug Self's "blameless" amplifier instead of an op amp? Linearity of the input stage is crucial, because the input stage is an error amplifier; and any errors the input stage makes will translate into distortion. I'm interested in what the community thinks of this idea.
A wide variety of discrete opamp stages will work, so you can probably use an LTP + current mirror stage, along with a VAS, instead of a monolithic opamp. In most cases, compensation has to be tweaked to get optimal stability and sonics. I have used a folded Kaneda-style Class-A discrete opamp stage, and it is stable and works well as a plug-in replacement for the monolithic LM318 *without* any compensation changes. Some optimization tweaks are most probably necessary for sonics, and this tends to be a bit subjective.
You just described Mr. Self's input stage. If you're not familiar with his work, I refer you to this.
Distortion In Power Amplifiers
I am referring specifically to the discussion of figure 8. It is very clever and actually quite simple. When I first read it, I had a facepalm moment. It's so obvious in retrospect that I slapped my forehead for not coming up with it myself.
I've got my unit with talema 25V 300VA and everything works fine and no noise. however it's just annoying that my house mcb (rated @1300VA) trips once i turn on the unit, although it does not happen all the time. using softstart module will not fit on my casing either.
Would it make so much different on power output / audio quality using 160VA? I assume that it should not trip my mcb anymore
Regards
Would it make so much different on power output / audio quality using 160VA? I assume that it should not trip my mcb anymore
Regards
It depends on the inrush current - it can happen even with a 160 VA trafo, though less frequently. Toroids and R-cores also tend to have higher inrush current than EI-cores of similar rating (probably due to better coupling between the primary and secondary). You could try using a low-value NTC thermistor on the primary side to cap the inrush current, though there's a school of thought that says that it hurts sonics.
200 to 240 VA is generally sufficient for a stereo MyRef setup, and 160 VA will also work in a pinch.
Edit: I noticed that you've omitted the speaker protection relay and circuit, from your pic in the chipamp photo gallery. That greatly increases the risk of blowing up your woofers, and in this case it's probably another cause of your MCB tripping. Try disconnecting your speakers and powering up the MyRef - if it stops tripping the MCB, then it's the surge current through the speakers at startup that's tripping it.
Last edited:
never thought of this, I've tried to unplugged speakers and no mcb trip for several times. it seems that i should assemble the relay circuit
how to increase delay time such as 3-5 sec? which value that need to be changed?
Regards
Last edited:
Not sure, the default values generally work well. There are two caps in the protection relay driver circuit, 22 uF and 100 uF, one of which (maybe 100 uF) that sets the start-up delay. The third 220 uF cap is the protection-circuit supply filter/bypass, which can be left untouched.