Sure-Electronics.com class d amps

Status
Not open for further replies.
MikeHunt79 said:

Also, I pulled the heatsink today and put some thermal goo under there. Seems strange that they would ship it without goo... 😕


Not really. This amp dissipates very little heat so thermal interface material is not really needed; the cooling is sufficient without. I ran 2 channels on mine hard (just below where ditortion becomes noticeable) at 26V for about 30 minutes and the it was only slightly warm. Actually my speaker's dust caps felt warmer than the amp (probably not a good thing).

Also, about the output ripple, the voltage seems consistent with what you measured, but it's at 440kHz (see pic).
 

Attachments

  • sure_d_amp.png
    sure_d_amp.png
    23.9 KB · Views: 892
Just wanted to metion I have smoked all my zobels too. It happened when I grounded the input oddly enough.

Also I wanted to say I am using it as a subwoofer amp, all channels are paralled togeather without resistors. It works well. (No thats not how the zobels got smoked, it was running each channel into a separate VC when that happened 😉.

I have yet to do any power measurements, but it is sufficient to get my little tang band W8 730C pretty loud.

I am running it off of a 19V laptop PSU. Although I have a 24V supply it can only deliver 2A 🙁...(well at least thats what its rated for)

Any luck running these at higher voltages? Or any Idea how to get more power out of them into a 4R load? Can I parallel all channels and use a transformer to up the voltage to the sub? Is there a way to wind it with a 1 ohm impedance on the primary and a 4 ohm impedance on the secondary...I think that would get the most out of it as a sub amp.
 
Yeah, you could use an impedance matching transformer. Although it's very likely you'd have to wind your own. Probably more work than it's worth, but I'll let you decide on that.

What I'd do is put two dual-vc drivers into one sub enclosure and run one channel to each vc.

If you're crazy enough you could also pack 4 drivers in there.😀
 
TheMG said:



Not really. This amp dissipates very little heat so thermal interface material is not really needed; the cooling is sufficient without. I ran 2 channels on mine hard (just below where ditortion becomes noticeable) at 26V for about 30 minutes and the it was only slightly warm. Actually my speaker's dust caps felt warmer than the amp (probably not a good thing).

Also, about the output ripple, the voltage seems consistent with what you measured, but it's at 440kHz (see pic).
Good to see you're getting similar results to me. 🙂 I see what you mean about the heatsink, it's amazing how little heat this thing makes when churning out 200W+. 🙂

I think I need to replace my zobels you see, as I've had smoke come from the board (I'm assuming zobel resistor), and it still works fine, so I'm going to replace them all to be on the safe side...

10R 5w is what fredos used, so I assume this would be best? Or would 3W or 1W resistors work just as well? Any particular type? Metal film? I know nothing about zobel resistors you see (apart from they help filter out supersonics)...

EDIT: I'm guessing 26v would be the limit when bypassing the diode?
 
Hmm I just bypassed the diode and I'm using 24V and its working good.

All the impedance matching trafos I found won't help me too much.

I'm thinking the multiple driver idea is not a bad one.

Too bad I can't justify the cost.

Oh well even 50Wrms is shaking stuff...plenty of SPL for music.

And it is suprising how cool these do run.

Do we really need zobels...I swear they get hotter than the amp
Better them burning than the speaker I guess
 
fredos said:
Run it for real with 4X 4 ohms load on it, you will see that this thing come really hot! I've try it in 2 way stage wedge last week-end on a rock band...They heat a lot in real pro audio application..But, it did'nt blow!

Fredos

I would hope it has thermal overload protection....

Put some thermal goo under the heatsink, that should help things also...
 
Hum or hiss?

I'm having no problem with hum at all, even on a cheap wall wart with lots of ripple and using my laptop as a source. If you're getting hum, you might want to check your source and make sure you don't have a ground loop going on.

The hiss (white noise) however, is always there and impossible to get rid of. It's not much of a problem if you're at some distance from the speaker, and any noise in the room (such as computer fans) drown it out. But in a very quiet listening situation, it is noticeable.
 
24v using 2x12v car batterys ?

Have just ordered one off ebay. Has anyone tried powering it on two X 12v car batteries wired in series? I think I read earlier that there is a risk that the voltage could go over the 26v max. I guess I could check that with a multimeter first? I want to use it on camping trips 😉
 
Re: 24v using 2x12v car batterys ?

col said:
Have just ordered one off ebay. Has anyone tried powering it on two X 12v car batteries wired in series? I think I read earlier that there is a risk that the voltage could go over the 26v max. I guess I could check that with a multimeter first? I want to use it on camping trips 😉
TheMG said:
Fully charged 12V lead-acid (car) batteries actually measure around 13.5V, so you'd be getting 27V with fully charged batteries, which is over the limit.
This is not a huge problem, simple add a diode or two to drop the voltage. This is what I would do if i wanted to power this from SLA's.

Does anyone know if this amp has overvoltage protection?

Also, the zobel resistors... They're the ones near the speaker terminals aren't they? I'm just about to replace them with some 7w resistors. They are a bit bigger than I expected! :bigeyes:
 
Yeah, I thought of using diodes to drop the voltage some time after posting. That would definitely work, just got to put enough in series to drop the voltage.

I have no idea if they have over-voltage protection, it's not stated in the datasheet, and I'm definitely not going to test it.

Are you sure you need 7W for the zobels? Seems a little overkill I think. Some 1W or 2W would probably have been fine. I haven't had any problems with the zobels on mine, although I can't say I'm driving it hard, the surround on my very old speaker's woofers started to rip apart so I had to turn it down, damned foam surrounds that dry out and become brittle as hell.
 
It will be more flexible with the 24v PSU and I have a 600w inverter so will follow that path.

I have a pair of JBL SB-2 Control subs with dual voice coils rated at 120WRMS 8ohms each voice coil. They have a pass through that you can connect satellites to which works really well. I am going to try running one of them with both voice coils in parallel which should be 240w 4ohm on 2 of the amp channels paralleled and will see how it copes. If it doesn't completely overheat I will try the other SB-2 on the other 2 channels, that should max it out 😉

Also, I have a pair of JBL GTO937 6x9" car speakers which are 2ohm 94db 100WRMS. I will try them too as the 2ohm load might extract some more power.

OK, will now wait for it to arrive.
 
The inverter will work fine, just remember you will probably get les runtime compared to the diode route... Also, I've found this amp to work really well with 2 and 1 ohm loads - just make sure you parallel everything that should be paralleled. 🙂
TheMG said:
Yeah, I thought of using diodes to drop the voltage some time after posting. That would definitely work, just got to put enough in series to drop the voltage.

I have no idea if they have over-voltage protection, it's not stated in the datasheet, and I'm definitely not going to test it.

Are you sure you need 7W for the zobels? Seems a little overkill I think. Some 1W or 2W would probably have been fine. I haven't had any problems with the zobels on mine, although I can't say I'm driving it hard, the surround on my very old speaker's woofers started to rip apart so I had to turn it down, damned foam surrounds that dry out and become brittle as hell.
I think someone in this thread said he had this amp running at 27v or so... I think it may have shut itself down on overload, but don't quote me on that. It's an amazing amp for the money, that's for sure...

The 7W resistors were the only ones I had that were more than 1/4W, so I managed to (messily) squeeze them in...

I've also leaned over two of the inductors, as I need to squeeze this in a car amp chassis... This shouldn't negatively affect things tho should it? I also need to move the 4 large caps to somewhere else, so I'll have to have them on extended cables from the PCB.... Again, will this cause any problems?
 
Generally speaking you want to leave leads on capacitors as short as possible. Long leads add inductance and resistance.

You could probably solder the caps to the other side of the board. Even get some new caps (longer leads) so you can tilt them 90 degrees, flat against the board.

I'm assuming what you're trying to do is get enough clearance to mount the IC directly onto the chassis as a heatsink?
 
col said:


Any idea what diodes I would need? Good to have this option for longer run time on batteries.
Any standard diode that can handle the current/power dissipation will be fine. There ain't no free lunch though, and these diodes will burn a fairly significant amount of power, so take that into account. They need to be thermally up to the job. Schottky types will have too low Vf to be useful (though they can help spread the power dissipation if you're dealing with lots of current).

Let's be conservative and estimate 66% efficiency for the class-D module, and say we're driving 100W/ch at most. 200W x 1.5 = 300W load, and that we want to run on 24V supplies.

300/24 = I = 12.5A current draw

Diodes in this current range have Vf about 0.9V. If we estimate max battery voltage at 28V and aim for 25V into the amp, we need 4 diodes. Each diode has power dissipation of 12.5*0.9 = 11W (!). This is too much for most packages without a heat sink. TO-220 or stud diodes with heat sinks capable of it would probably be sufficient, or bolted to the car's chassis (with insulators of course!). Get at least 15A rated diodes.
 
Status
Not open for further replies.