Iraud200 - diyAudio
Go Back   Home > Forums > Amplifiers > Class D

Class D Switching Power Amplifiers and Power D/A conversion

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 28th July 2015, 10:38 PM   #1
diyAudio Member
 
Join Date: Nov 2012
Default Iraud200

I'm amazed no thread comes up under search for iraud200

I'm going to add an iraud200 vs iraud350 rant

Just ordered 2 of these, first one showed up, and it's bothering me a little. But before I get to that, let me put in a positive vote for the iraud200 design .. don't let the iraud350 boards fool you with the larger number (350)... the iraud350 units do not come with irfb4227 transistors, which are the real mosfet powerhouses. The output inductor of the iraud200 is about double the volume of the iraud350 The 200 has also about 15% more heat sink than the 350 .. The iraud200 is the real 2 ohm load driver .. the iraud350 ads often claim it, but but the boards don't really have the hardware installed for 2 ohm loads. I ordered the 350s thinking they were right but didn't know about the little differences that are big differences. It was a very disappointing experience to be fooled in that way. Don't get me wrong, I think there is a very small number of iraud350 boards out there with irfb4227 on board, but they are hard to find, and I still doubt that they have the output inductor installed to match the one on the iraud200. If you want the real deal, you want the iraud200.

the 4227 ohms out in on state at 1/20 of an ohm .. like it's hardly even there. Its rated for around 40 amps current at room temperatures. You want your class D boards to be sporting this transistor. It's the real deal.

The 200 has less capacitance on board by far than the iraud350 ... only plus side for the 350, but it may be less of a plus than you think Having huge capacitance on board means you can't protect the traces from exploding on transistor short circuit. In my opinion, its better to have the caps off board and fuses to protect your traces from major discharge in case of problems.

Both boards use a lot of surface mount components, which is probably good because I remember looking at a through hole board thinking ... how do those through hole components get along with a ground plane because the current flow can't get close enough to the board surface to capacitatively connect to the ground plane properly. However .. replacing surface mount components can be tricky if you have to do that.

The iraud200 appears to have no overcurrent protection. I can not find any hint of overcurrent protection on any spec sheets, though overtemp protection is mentioned. This does not surprise me because I have seen the irs2092 overcurrent protection utterly misbehave. Eliminating it entirely was my final solution .. Don't get me wrong, the irs2092 overcurrent protection logic probably works fine under circumstances .. usually 8 ohm load kinds of circumstance. This means you better have fuses installed for this board, or something better. Fortunately, the dang irfb4227 can handle so ***** much current that the only likely overcurrent event would be trying to drive a one ohm load or a short. This board should handle 21 amps of output current. Not kidding, but I wouldn't push my luck with that. I wouldn't ask it to deliver more than about 600 watts rms into 2 ohms. that's around 17.4 amps .. And that with extra heat sink .. the way I solved that is I just mounted flat surface of a larger heat sink to the top of their heat sink.

I thought the transistor looking thing on the left would be a fet driver .. NOPE, it's a thermostat, so the board does have over temperature protection (yay) and there's where my concerns start. I thought the part to the left of the heat sink was driving the high gate charge of the irfb4227, but no, it appears that the irs2092 is driving the irfb4227s directly as there is nothing on the board that looks like a gate driver (with no gate resistance .. this is a requirement for the high gate charge transistors) . Frankly, that is scary, and it tells me that this board may be designed to poop on you. The iraud350 version direct drives its transistors as well, but a: the transistors have about 30% less gate charge than the irfb4227 (and about double the on-resistance as well as almost double the thermal resistance) and b: the iraud350 boards I got had large heat sinks (over an inch tall) mounted to the irs2092 chip. and c:Those heat sinks got hot. In my opinion, the only way they could direct drive without a heat sink is to do something in the circuit to slow down the firing rate of the transistors .. i.e. slow down the feedback loop. This would a:help to relieve the gate charge problem and b:create a new problem of damage to the sound quality .. I may report on this later.

OK, so looking at the boards, I'm saying to myself I really like the big output inductor, That thing is quite convincing. but ...... I don't want to even charge up the boards without getting heat sinks mounted to the irs2092 chips. To me, that's just irresponsible .. please, anyone with understanding I am missing, let me know, but my exp. thus far says that the irs2092 is not designed for that. Now, just for fun, I found some suitable heat sinks on ebay .. 10 x 35 x 10 mm. They are tall, and deal with it because this chip needs help driving those transistors. And you will need heat sink glue to glue them on. Because I plan on driving 2 ohm loads, I am also going to supplement the regular heat sink and also heat sink the output inductor. .. and of course .. fuse the unit in case of accidental shorting of the output.


Now, just for extras ... I installed irfb4227 transistors on an iraud350, and it worked with no immediate disasters. However, I think the switching of the transistors was not optimal, and the switching losses were high. The heat sink got hot even with no signal, and I had to supplement it. After watching it drive my 2 ohm sub-woofer flawlessly for some months (though without really putting the pedal to the medal cause that's quite noisy you know) .. This driver is a 600 watt (rms) 15 inch subwoofer from MA audio.. dual 4 ohm voice coils wired in parallel (btw .. never just straight out believe your multimeter about the impedence of speakers because they have reactance, not just resistance) ... really big double stacked magnets and at least an inch of xmax in both directions, around 90-91 db efficiency is my best guess, mounted in a 12 foot long bass horn tuned for 20 hz quarter wave, so horn loaded, which means extra air impedance resisting the drivers movement. This board with the 4227 transistors kicked that driver so hard that it bottomed out and banged very audibly... The transistors just smiled at me and said ... fun ... do it again. :-) I can't imagine doing that with less than 20 amps of current. The power supply is 70 volts with 22000 uf/pole, so up to something more like 30 amps of current and 1800 watts would have been available for transients, which is probably not far from what happend. I'm going to guess 25 amps and 1250 watts transient. The pre-amp stage (not this board) was having issues that created transient jolts (solved since then). This board kicked the driver like that many times and showed no sign of weakness. Of course, its the IRFB4227 transistors that are doing the work. OMG those little transistors are amazing. That kind of jolt would instantly trash the outputs of many amplifiers. I saw it happen for example with a pioneer sx 1050 and only 4 ohm loads ... this was a 2 ohm load!!!

So, I was hoping for some prior experience with this board to show up, but no luck that I could find. anyone else willing to banter about this board, please drop a line, as I am going to be placing some small bets on this board. I am going to supplement an old pioneer sx-1250 monster receiver with two of them. It is suggested to have separate ground lines for this (to eliminate mutual interference), which I can't do on the 1250 even though it has separate supplies, so I have ordered some ferrite beads and will probably use some inductors as well to quell hf across the ground line. I'll probably report later on how well this "power supply decoupling" works. I remember I had two L-25D boards that just couldn't get along with each other without a struggle.

In any case, despite the heat sink issue, I am seriously rooting for the iraud200 as the cheapest, genuinely 2 ohm capable board for 60-80v power supply poles. ... Oh, and I'm seriously considering putting a really big heat sink on the 1250's transformer as well as maybe also some heat sink on the rectifier diodes, cause I'm going to need another 300 watts. :-)
  Reply With Quote
Old 28th July 2015, 10:55 PM   #2
diyAudio Member
 
nigelwright7557's Avatar
 
Join Date: Apr 2008
Location: Carlisle, England
I have designed a few irs2092 pcb's.
The layout needs to be as tight as possible with star grounding.
Decoupling is vital on 2092 and output mosfets.

I have also used 4227's but with transistor gate buffers. The 2092 gate driver is pretty weak.

I use a t106-2 core for the output filter. (barely gets warm)
I also noticed the output filter capacitors would blow if no speaker was attached and the circuit was powered. The output inductor rings and puts out a high voltage, so I use high voltage output filter capacitors.
__________________
PCBCAD51 pcb design software. http://murtonpikesystems.azurewebsites.net/
  Reply With Quote
Old 28th July 2015, 11:47 PM   #3
diyAudio Member
 
Join Date: Nov 2012
Default inductor

this inductor is labelled SAGAM I 220 YX I don't know what that means, but comparing it to the inductor on the iraud 350 boards makes the 350 boards look cheesy. It's 7/8 inch by 7/8 inch by about 13/16 inch I'm betting it will handle something close to the 24 or so amps required for its rated 900 watts into 2 ohms. Thanks for mentioning the bus pumping problem. I don't know if I will be facing that issue, but I better put my meter on the caps and watch what happens when there is no load. .. and stand ready to resolve any problem very quickly. Maybe like a 400 ohm power resistor could stabilize that without drawing too much power. OIC .. it already has a 1.5 kohm resistor looks like 5w from output to ground, so it's never totally loadless. I'm going to guess that that little cap popping problem has been dealt with.
  Reply With Quote
Old 29th July 2015, 12:43 AM   #4
diyAudio Member
 
Join Date: Nov 2012
Default no heat sink grease

OMG there is no heat sink grease on any of the heat sinked components on this board. Not one. Ladies and gents, do not power up this board until you grease all components. If the transistors are not glued to the mica, then you will want to grease both sides of the mica. Then, make sure there is a heat sink as described earlier glued with heat sink glue to the irs2092 chip. Do it or cry :-) heat sinks without heat sink grease all but don't exist. Obviously the manufacturer of the board has quality control problems. ... glad I haven't powered mine up yet. Also, get the transistors, diode, and thermostat properly greased to the heat sink before glueing on the heat sink to the irs2092.. otherwise the 2092 heat sink will get in the way of your screwdriver for the other components.

Last edited by mgshightech; 29th July 2015 at 12:56 AM. Reason: needs another detail
  Reply With Quote
Old 29th July 2015, 12:54 AM   #5
diyAudio Member
 
nigelwright7557's Avatar
 
Join Date: Apr 2008
Location: Carlisle, England
I found I didn't need grease on my 2092 amplifiers output mosfets, they just didn't get hot even on full power.

The 2092 doesn't need a heat sink under normal conditions.
Mine get warm but not too hot to touch.
I bought an infra red thermometer for testing.
__________________
PCBCAD51 pcb design software. http://murtonpikesystems.azurewebsites.net/
  Reply With Quote
Old 29th July 2015, 01:04 AM   #6
diyAudio Member
 
Join Date: Nov 2012
Default thermal runaway

I hope you're right, and the heat sink never heats up. that will mean I don't have to supplement it. However, my exp. with it so far is L25D ... heat sink got too hot ... Iraud350 heat sink only got warm with original transistors and no power movement, got too hot with power movement and also with the 4227 and no power movement .. even with grease, the thermal flow was not adequately managed. However, in all of these cases, I was running the boards with low impedance loads. If you look at the specs on the transistors, you will find that their internal impedance rises with temperature .. in other words, where the transistor alone might power a 16 ohm load if it is switched perfectly, the transistors will be looking for thermal runaway with low impedance loads and no heat sink. Rising temperature will cause more rise in temperature. I'm going to be driving 2 ohm loads ... that's why I got the board in the first place, cause I knew it could do that. Also, the irfb4227 is not normal conditions for the irs2092 15^2/20 = 11 watts and that's with no switching losses ... I'm getting the 15 from ten amps into the transistor and 5 amps bouncing back and bus pumping. More exact calculations could be made, but a transistor that size can not take 11 watts at all really, and definitely not without getting hot, which means that the internal impedance will double and quadruple.

Last edited by mgshightech; 29th July 2015 at 01:12 AM.
  Reply With Quote
Old 29th July 2015, 01:07 AM   #7
diyAudio Member
 
nigelwright7557's Avatar
 
Join Date: Apr 2008
Location: Carlisle, England
It depends a bit on how the mosfets gates are driven. If you are getting some cross conduction the mosfets can get very hot.
I used maximum dead time on my 2092 amplifiers and had no problems with heat.
Obviously depends on heat sink size too. If it is small then it is more likely to heat up.

You can run the amps and see how hot they get and decide from there.
An IR thermometer is a great tool for this type of problem.
Fingers are fine up to about 40-50 degrees C.
__________________
PCBCAD51 pcb design software. http://murtonpikesystems.azurewebsites.net/
  Reply With Quote
Old 22nd August 2015, 03:41 AM   #8
diyAudio Member
 
Join Date: Nov 2012
Default first test

Even with 5x more heat sink than it came with ... this thing gets so hot the temp protection shuts it off, even without signal, so you must be on to something there ... would adding heat sink grease to transistors or irs2032 cause "cross conduction"? or should I be looking for dead time calibration resistors? It must be dumping 100 wattsish to ambient .. It's useless as is. Unless the grease caused the problem ... it would have destroyed itself without adding it because of inadequate thermal connection between transistors and thermostat. Also ... am using 2 ferrite beads on the 0 volt line to kill rf interference ... that might have something to do with it. .. problem with this rf stuff is that it does weird things that are difficult to troubleshoot. Any suggestions would be appreciated. Also, it may be possible that I am dealing with a quality control problem, but I doubt that.

Last edited by mgshightech; 22nd August 2015 at 03:48 AM.
  Reply With Quote
Old 22nd August 2015, 03:46 AM   #9
diyAudio Member
 
Join Date: Nov 2012
I'm nervous about taking the ferrite beads off because I have had problems with amplifier circuits exciting each other to instability. .. and it's in a system with other boards .. though currently no other class D boards.
  Reply With Quote
Old 22nd August 2015, 04:38 AM   #10
diyAudio Member
 
Join Date: Nov 2012
Default second test

did a cursory check over two board(s) . The boards were ordered at separate times. two boards are behaving identically .. super fast heat up even without signal .. noticed that on both boards, two of the resistors are not what the board print calls out for ... one calls for 10k and has a 20k ... another calls for 15k and has a 20k ... at this point it could be interesting to swap them for the items the board demands.. only they are surface mount, and I don't know if they rely on the ground plane. .. so replacing them means an entire ordering cycle, so it might be better to think first.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



New To Site? Need Help?

All times are GMT. The time now is 12:20 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2015 DragonByte Technologies Ltd.
Copyright 1999-2015 diyAudio

Content Relevant URLs by vBSEO 3.3.2