DC speaker protection for Sure class-D amps?.. - Page 2 - 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 13th November 2009, 01:45 PM   #11
diyAudio Member
 
Join Date: Jul 2005
Location: Devon
Sorry, I'm having a "stupid day" today...

The signal gen in the sim software works fine - The voltage dial sets the "peak-to-peak" voltage (ref. to GND), so the voltage across the pos and neg pins are of course at exactly HALF of the p2p voltage.

Also, the circuit does NOT work properly without using a non-polar cap (or two electrolytics back-to-back) directly BEFORE the diode bridge. This forms a proper RC circuit.

With only one cap placed after the diode bridge, the circuit responds more like a peak detector, (so the relay will chatter all the time when playing at higher volumes.)

So, the circuit above actually works quite well with a non-polar cap of around 16uF placed BEFORE the diode bridge.

With 6K input resistors and a 16uF cap, the relay cuts out at 1Hz if there is a 2.6V pk-to-pk signal ACROSS the speaker wires.

(The "bridged class-D" stuff and the simulation is confusing, because there really is only 2.6V ACROSS the speaker wires at the peaks and not 5.2V as it says on the dial).

I can't test it at higher voltages because the sig gen in the software doesn't go higher than 6V. But, you can visualize how the graph of "voltage vs frequency" would pan out.

I'm not good with calculating or plotting this stuff on graphs, so if anyone has any pointers?

If you use a similar circuit on the Sure amp, you should be able to find the best response time by adjusting the corner frequency.

Also, remembering that your input cap acts as a HPF, the relay shouldn't cut out during normal music playback (unless the amp's output frequency drops below your input HPF cutoff frequency.)

In other words - at or below 3.3Hz, the relay will still cut out when there 6V peaks across the speaker wires. To stop this happening during normal music playback you'll need to adjust the cap in the circuit depending on the HPF formed by the combination of your input capacitance and the amp's input resistor (100k I think?).

If the amp then fails for some reason and suddenly presents 2.6V DC (or more) across the speaker output (in either polarity), the circuit SHOULD switch the relay off within around 400ms (with the current circuit). Please don't quote me on this though!

If you want to use this for tweeter protection (active setups), then I think you could make the circuit's RC corner frequency higher so the relay cuts out when the amp's output frequency drops too far below your tweeter's crossover freq?

I hope this makes sense. I never got on with AC theory at college, so this stuff is tough for me to work out!


OzOnE.
  Reply With Quote
Old 16th November 2009, 12:34 PM   #12
diyAudio Member
 
Join Date: Jul 2005
Location: Devon
Right, I'm going to try to keep the typing down a bit and show something a bit more "practical" now.

I've made a PCB layout for the DC protection circuit. I chose to use surface-mount components, as it doesn't use the very smallest parts so is still fairly easy to solder (I personally quite like soldering SMT stuff in most cases.)

The use of SMT components means that the PCB is not much larger than the relay itself! This should be good for confined spaces / multiple channels.

I've actually used a double-pole relay for each speaker as these were the type of relays I had left over from another project. I can easily change it for other layouts if anyone is interested?

The way it works is that it only switches the "positive" speaker output, but you still need to connect a wire to "negative" amp output for the circuit to detect the voltage across the amp output (We could call this the "sense" wire.)

It will also need a voltage supply for the relay, so you can choose this depending on the relay type.

The circuit should work for both bridged or standard push-pull / half-bridge amps (ie. common "negative" output). btw, the circuit should still work the same even if the input / output polarity is swapped.

A word of warning though, I haven't built this specific design yet, so they will need some basic testing (and tweaking of values) on first use, but should in theory work OK.

Attached is a picture of the PCB layout to give you the basic idea. I still need to order some components and maybe some PCBs to build these though. (I could make some DIY PCBs, but the quality usually isn't great.)

Would this type of module be of any interest to anyone?

OzOnE.
Attached Images
File Type: jpg DC Protect SMT PCB.jpg (113.4 KB, 504 views)
  Reply With Quote
Old 16th November 2009, 12:43 PM   #13
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Borbely tells us to swap one of the double pole relay contacts so that current flows in the parallel contacts in opposite directions. Can't recall if he gave a reason for reversing.
__________________
regards Andrew T.
  Reply With Quote
Old 16th November 2009, 12:44 PM   #14
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
will the relay be capable of breaking the DC current in the event of a major output stage failure?

Will an input mute help to reduce the DC current that must be interrupted?

Will supply rails fuses help?

Have you considered supply rail crow bar protection?
__________________
regards Andrew T.

Last edited by AndrewT; 16th November 2009 at 12:48 PM.
  Reply With Quote
Old 16th November 2009, 12:57 PM   #15
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Quote:
Originally Posted by OzOnE_2k3 View Post
The circuit only needs 2.2V of DC to turn off the relay...........
...................Could 2V of DC across the speaker cause much damage though? This is still around half an Amp on a 4 ohm speaker? (assuming a failed amp will still supply that much current).

This is why I hate it when people "tap-out" the speaker phase using a 1.5V battery. It must be doing some damage if it's held on for too long?
I would expect a long term 1.5Vdc (>=100ms) to trip the DC detect circuit.i.e. a tap with a 1.5V battery should trigger the DC detect circuit.
I would also expect a 10Hz 20Vrms to not trip the DC detect circuit
I would expect a 20Hz 28Vrms to not trip the DC detect circuit fitted to a 100W into 8r0 amplifier.

1.5Vdc, from a battery to a 4ohm speaker (Re~3r0) will send ~750mW to the speaker voice coil. If your speaker cannot survive that for a half second or so then there is something wrong with the speaker you have selected to connect to a 100W amplifier.
2.2Vdc increases the voice coil power to ~1.6W
__________________
regards Andrew T.
  Reply With Quote
Old 16th November 2009, 01:08 PM   #16
diyAudio Member
 
Join Date: Jul 2005
Location: Devon
Quote:
Originally Posted by AndrewT View Post
will the relay be capable of breaking the DC current in the event of a major output stage failure?

Will an input mute help to reduce the DC current that must be interrupted?

Will supply rails fuses help?

Have you considered supply rail crow bar protection?
tbh, I haven't really looked into anything more advanced because I wanted something that would give the minimum basic speaker protection for amps like the Sure amp before hooking up to my speakers.

I have heard about the "diode" effects when connecting up relay contacts - I suppose it would be simple to change, but this type of thing gets a bit heavy for me. I mean, they use standard relays and fuses in most commercial gear (at the lower price range at least), so I could actually use a dual-pole relay for two channels, then add the DC detect circuitry to the same board.

This circuit is not meant to have ultimate protection for high-end amps, but it's way better than nothing on the Sure amp considering there have been a few "meltdowns" already.

I think it's always a good idea to calculate your average current consumption (and maximum amp output wattage) during music playback, then add fuses on the power rail(s) to protect against abnormal situations. I know many audiophiles don't like fuses or relays though, but we could be getting into snake oil territory there.

It would be nice to have all the extra protections possible, but this is all a bit beyond my skill level and budget tbh.

I've attached a rather amusing drawing of how this module would connect to the Sure amp (for anyone wondering). You can tell I'm no artist!

btw, the "/PROTECT" output is an active-low signal which could be used to tell a microcontroller which channel went into protect mode. This could also be used to shut off the relays on all the other channels at the same time (just in case the amp fails big time after the first channel pops.)

The /PROTECT line is connected to the relay supply via the 10K resistor and also via the transistors though, so I would probably use a zener on this line to limit the voltage before connecting to a CPU board.

Please keep in mind that I'm very much open to suggestions for this circuit, but my knowledge of this stuff only goes so far.

OzOnE.
Attached Images
File Type: jpg DC Protect Sure Hookup.jpg (81.6 KB, 441 views)
  Reply With Quote
Old 16th November 2009, 01:14 PM   #17
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
I do like the idea of the SMD based detector circuit lying under the relay footprint. Neat.
__________________
regards Andrew T.
  Reply With Quote
Old 16th November 2009, 01:38 PM   #18
diyAudio Member
 
Join Date: Jul 2005
Location: Devon
Hi, Andrew,

I just wonder why you don't often see DC protect circuits on the web which specify that they also work on bridged amps? This is very surprising to me.

Would you say that around 1.5V is the minimum DC voltage that you would want for the circuit to trigger? This should be fairly easy to achieve by playing with the cap / resistor values.

The problem is stopping the relay from dropping out with higher voltage sine wave / music. Now that I think of it, perhaps a zener diode could be placed after the input resistors to limit the maximum voltage across the cap? This might make the circuit less prone to trigger at higher voltages?

Using a zener is likely what I'd use if I had a PIC chip to monitor the speaker voltage. I would think that using a PIC should make the detection period the same for any voltage (above the diode bridge voltage drop and the PIC's input threshold of course)?

It would actually be quite simple for me to add a small 8-pin PIC chip to the relay board, but I'm always a bit paranoid to rely on a CPU for protection (even though this is the way of the modern World it seems. )

Looking at the PIC design on Rod Elliott's site, he has a table with some example LPF component values for different frequency / delay / voltage response times. So, in either case (simple LPF or LPF + PIC), it's still fairly tricky to select suitable parameters.

I'm wondering if it would be enough to instead just measure the length of the DC pulses from the diode bridge rather than rely on a LPF at all?

OzOnE.
  Reply With Quote
Old 3rd March 2010, 12:31 PM   #19
diyAudio Member
 
Join Date: Jul 2005
Location: Devon
Hi, all,

Sorry to drag this thread up again, but I've just started looking at my amp project again because I recently bought a nice Tracopower 24V PSU for my 4-channel Sure amp...

(I've still never heard the Sure amp or my Sympatico amps on my 4-ohm speakers because I don't want to risk it without having basic DC protection.)

I'm still having major problems trying to figure out the best type of circuit for this amp...

Yesterday, I soldered four DPST relays onto a bit of veroboard (with their coil transistors and back-emf diodes), but I'm now wondering if something similar to the DC protect circuit I drew up (post 9) will be any good?

Does anyone know of the simplest way to implement DC protection for this type of class-D amp? The problem is that most protection circuits on the web are only for standard dual-supply / push-pull amps where one of the speaker output pins is almost always referenced (or connected) to ground.

With these single-supply "H-bridge" Sure amps, and a 24V supply, both speaker outputs sit at around 12V with no input / idle...

Does anyone know of a simple method of measuring the voltage "between" the two outputs, but still have the DC protect circuit referenced to ground?

@AndrewT - I've tried simulating a PIC chip protection circuit, but it kept falsely triggering the protection at certain input frequencies. Did you use the PIC's ADC to get a value for each speaker output voltage, then take the "difference" between the two values. Then, set a time delay if the difference stayed at DC for too long?

I'm having trouble visualizing how the PIC circuit would work for Class-D?

If anyone has any pointers, I would be very grateful.

I have a nice ES9008 8-channel DAC with 7.1 decoding and some new amps which have been sitting on a shelf looking very sad for the past few months. Please help!

OzOnE.
  Reply With Quote
Old 3rd March 2010, 02:35 PM   #20
sendler is offline sendler  United States
diyAudio Member
 
Join Date: Aug 2004
Location: ny
Default off the shelf

I'm going to try an off the shelf speaker protection module this weekend to see if it works. It should still work just referencing the + and - to each other. If dc is low between the legs as normal, it will connect. If one leg of the output went open it would have no reference and the other leg would be at full rail dc but the speaker would quit playing and wouldn't care. If one leg failed with a short the the dc sensor would sense full dc and go into protection.
__________________
Scott
  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


Similar Threads
Thread Thread Starter Forum Replies Last Post
For Sale: Tube Amps, Class-A Amps, BG Ribbons, Chassis and Other Goodies! opc Swap Meet 23 2nd November 2009 05:04 PM
How about a round-up of Class A kit power amps, or collectable vintage class A? Brisso57 Solid State 4 14th February 2007 10:30 AM
speaker protection (OR) overload protection myanmar Solid State 7 13th July 2006 08:21 AM
What speaker cable for Class-T amps, Cat5 MrDodo Class D 12 8th December 2005 10:44 PM


New To Site? Need Help?

All times are GMT. The time now is 08:55 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2