SSLR distortion is about 1 ppm at 20 kHz. If you do a search on the forum you will see an AP plot that confirms this. LTspice sims show about the same level.
A photovoltaic coupler is about $1.50 from mouser IIRC. I use two in parallel for fast switching.
Re on a fully turned on Trench MOSFET in the Vds range of 100 - 150 Volts is well under 10 mOhms. If you want low distortion, you need low Rdson.
Suitable mosfets are about $3-4 each from Mouser.
Your SSLR should not cost you more than about $ 8-10.
A photovoltaic coupler is about $1.50 from mouser IIRC. I use two in parallel for fast switching.
Re on a fully turned on Trench MOSFET in the Vds range of 100 - 150 Volts is well under 10 mOhms. If you want low distortion, you need low Rdson.
Suitable mosfets are about $3-4 each from Mouser.
Your SSLR should not cost you more than about $ 8-10.
I got a Mouser order recently.
It was declared by Mouser "all fees paid".
But expect to pay for V.A.T. at time of delivery.
We have an exemption for small import values. Mouser stated the value of the goods @ £12??
then added on almost the same again for post and handling and fees.
The importers declared the small package as "tax exempt", handwritten on the delivery label.
BUT the delivery firm insisted on a £11 payment at the door as their fee, in addition to what Mouser paid them for the international delivery.
I ended up paying ~£35 for £12 ($18) of smd components.
It was declared by Mouser "all fees paid".
But expect to pay for V.A.T. at time of delivery.
We have an exemption for small import values. Mouser stated the value of the goods @ £12??
then added on almost the same again for post and handling and fees.
The importers declared the small package as "tax exempt", handwritten on the delivery label.
BUT the delivery firm insisted on a £11 payment at the door as their fee, in addition to what Mouser paid them for the international delivery.
I ended up paying ~£35 for £12 ($18) of smd components.
Last edited:
I pay a flat $20 for all Mouser orders. I go onto the Taiwan website where I am currently resident ( I select English language). All import duties and taxes are taken care of.
Mouser is one up on Digikey in my view.
When I lived in the UK I used RS. But, RS outside of the UK stinks and Farnell is ok, but they seem a bit lost wrt catalog and how they handle orders.
For EU, you may want to look at one of the German catalog houses.
Someone may have mentioned this already but Hafler amps had a clever solution for output protection. The output fuse was placed inside the feedback loop. That way any impedance caused by the fuse would be compensated for by the feedback, and the damping factor and THD would be nearly the same as if the fuse were not there (mathematically existent, but negligible).
The same result could be achieved with any other protection device - relay, MOSFET switch, SS relay etc. - if the feedback connection is moved to the speaker terminal after the protection device. When the device opens up, the amplifier will clipp to the rails, but it will not blow up because there will no longer be any load attached.
UREI amplifiers and speakers took this one step further and actually ran the feedback loop connections out to the speaker terminals (inside the speaker enclosure) through coax cable to null out the speaker wire impedance too.
The same result could be achieved with any other protection device - relay, MOSFET switch, SS relay etc. - if the feedback connection is moved to the speaker terminal after the protection device. When the device opens up, the amplifier will clipp to the rails, but it will not blow up because there will no longer be any load attached.
UREI amplifiers and speakers took this one step further and actually ran the feedback loop connections out to the speaker terminals (inside the speaker enclosure) through coax cable to null out the speaker wire impedance too.
Which part number? The TLP191 in your PDF is $4 at mouser plus about $60 min shipping. Most of the DIP versions are $9 each though there are a couple at $5. I would absolutely build your design in a heartbeat if that style of optocoupler was available easily here.
In AU, eBay is often the cheapest source of small quantities. I can get basic optocouplers for about $1-$2 each, or high speed stuff like FOD3180 (2A MOSFET gate driver) for $3-$5; they just require that I have an external floating supply for the gate drive.
Last edited:
I've just looked on ebay - TLP191 1$ in small qty's.
I am really surprised at your 60$ shipping charges - that's a rip off.
If you use the photovoltaics you shouldn't require a floating gate drive - they generate the drive.
One other trick might be to try to see if you operate a standard 5 pin opto in photo diode mode. The detector transistor base connection becomes the - output and the collector the plus output. I'll need to try it.
If you want to get rid of the optos because of cost, another option here might be to use a charge pump - but by using the amplifier output to charge a cap to above the amp supply rails.
I am really surprised at your 60$ shipping charges - that's a rip off.
If you use the photovoltaics you shouldn't require a floating gate drive - they generate the drive.
One other trick might be to try to see if you operate a standard 5 pin opto in photo diode mode. The detector transistor base connection becomes the - output and the collector the plus output. I'll need to try it.
If you want to get rid of the optos because of cost, another option here might be to use a charge pump - but by using the amplifier output to charge a cap to above the amp supply rails.
Last edited:
Not if you want 'em shipped to AU, though please (pretty please!) prove me wrong on that one.
Yep, and yep.
Don't think that'd work at low signal levels, or at startup. 555 seems easier and sonically cleaner (and stabler) than loading up the output with a capacitor+diode.
The bootstrap action is only required on signal peaks where the output approaches the rails.
If a low power bootstrap has a sonic signature I'd expect the amplifier OPS is pretty sub optimal. Boot strapping is quite an accepted and well tried technique in audio amplifiers.
With a 555 you've got very fast rise/fall times that I expect would require greater attention to layout etc.
If a low power bootstrap has a sonic signature I'd expect the amplifier OPS is pretty sub optimal. Boot strapping is quite an accepted and well tried technique in audio amplifiers.
With a 555 you've got very fast rise/fall times that I expect would require greater attention to layout etc.
Last edited:
I just looked on ebay as well, in europe, and only one seller for 3.49euros, plus shipping.
The thing is, this part is smd and I'm looking for through hole type, dip would be best.
What would you suggest for through hole part?
Not for me unfortunately. I would've done that without any problems some 5 years ago, but no longer. Not only my eyesight has gone down a bunch (even with glasses), my dexterity is also mostly gone, so I hope to still be able to solder through hole stuff, but smd is totally out of my current abilities now.
That's why I was looking for other options such as that PVI1050N, is it a worthy candidate?
I don't use any smds in my designs because I won't be able to deal with them (onset of aging).
And btw: I am also looking at making ssr for the rails. Those would be simpler, with DC to handle.
I really like your e-amp, and I would love to build one, but smds would totally ruin the fun for me.
A few years ago I had to replace a small fan in my mac laptop. It had a tiny 4 or 5 pins surface mounted connector and I ended up breaking it because I couldn't see it well enough and my hands shake too much. Even with a big magnifier to help me see, my hands still don't let me do anything too precise any more. I hate this, but there is nothing I can do against aging.
Member
Joined 2009
Paid Member
Not all SMD is 'too small'. I found the ideal combination to be through-hole and medium sized SMD parts. It allows for a more compact design and less solder flux to clean up.
A good example here : http://www.diyaudio.com/forums/solid-state/245619-tgm8-amplifier-based-rod-elliot-p3a.html
I used a solid state relay for speaker protection and it works just fine. The turn-off time is limited by the discharge time of the FET gate capacitance. A resistor provides leakage path for this capacitance to discharge. Having two opto's allows use of a smaller resistor for faster turn-off but in my opinion this is unnecessary because the turn-off time with a single opto is still much faster than a relay, still much faster than the time it takes for the circuit to decide that there is a dc-fault and not just a big bass note. I built the speaker relay into the amplifier pcb in this project.
A good example here : http://www.diyaudio.com/forums/solid-state/245619-tgm8-amplifier-based-rod-elliot-p3a.html
I used a solid state relay for speaker protection and it works just fine. The turn-off time is limited by the discharge time of the FET gate capacitance. A resistor provides leakage path for this capacitance to discharge. Having two opto's allows use of a smaller resistor for faster turn-off but in my opinion this is unnecessary because the turn-off time with a single opto is still much faster than a relay, still much faster than the time it takes for the circuit to decide that there is a dc-fault and not just a big bass note. I built the speaker relay into the amplifier pcb in this project.
How small is small though?
Looking at that pcb picture of your post #6 there, for example that small smd resistor (r33) is far too small for me to safely solder. There is no way I could properly put it in place, hold it there and solder it. I would probably break it or burn it. I would've done this without any problems years ago, but no more.
That's exactly what I have in mind. I want to put the ssr for the output and rails right on the amp's pcb, with everything else too, psu, protections... I hate too many wires, so I'll do anything to avoid as many as possible.
Member
Joined 2009
Paid Member
Of course it does depend on the constructor. In my case I use 1206 sized resistors and SOD 123 sized diodes. I find two terminal SMD parts are relatively easy and for three-terminal parts I tend to stick with through-hole.
What I do is put a little solder on one of the pads on the pcb first. Then I pick up the resistor with a pair of tweezers (in the past I used fine nosed pliers but I find tweezers much easier to use) and place it on the pcb. At this point I use the iron to warm up the pad/resistor which melts the solder and forms the connection to one end of the resistor whilst also now holding it in place on the board. Then I take the iron to the other end of the resistor and with a tiny bit of solder finish the other joint.
Well, if it's me, then I have to do without smds, as I just won't be able to solder them, the result would be much less than acceptable and I wouldn't have any fun doing that, as this would strain my eyes, even with magnifying glasses, and my hand just isn't steady enough to do this any more.
So that is why I am looking for everything through hole.
As I can see, this is 3.2mm x 1.6mm metric size, and that's definitely much too small for me to deal with.
You must be young!
