Thank you for the suggestion. I agree. I will construct a double integrator for the accelerometer. At this point I am satisfied that I have captured what I need with the accelerometer.
Next I will post some measurements of a smaller speaker in a closed box with voltage drive vs current drive. I will capture a microphone near field measurement at the resonance frequency of the speaker in the box.
Distortion measurements
Moving from acceleration to velocity measurements.
5 1/4" Audax HD13B driver (my favorite mid-woof) in a 1 cu ft -ish enclosure.
Measurements at speaker/box resonance at the drive level. Near-field microphone. Everything in the test is identical except switching out the power amp.
Current Drive
64.9 Hz, 10.075 VRMS, 224.4 mA RMS, 45.1 ohms, 7.7% HD
Voltage Drive
64.9 Hz, 10.074 VRMS, 231.1 mA RMS, 43.8 ohms, 8.8% HD
The drive is 2-ish watts.
With voltage drive, back emf current distortion is very noticeable.
My conclusion is that most of the HD is due to the non-linearities; BL, Cms, etc.
There is a measurable reduction in HD when driven by current.
Moving from acceleration to velocity measurements.
5 1/4" Audax HD13B driver (my favorite mid-woof) in a 1 cu ft -ish enclosure.
Measurements at speaker/box resonance at the drive level. Near-field microphone. Everything in the test is identical except switching out the power amp.
Current Drive
64.9 Hz, 10.075 VRMS, 224.4 mA RMS, 45.1 ohms, 7.7% HD
Voltage Drive
64.9 Hz, 10.074 VRMS, 231.1 mA RMS, 43.8 ohms, 8.8% HD
The drive is 2-ish watts.
With voltage drive, back emf current distortion is very noticeable.
My conclusion is that most of the HD is due to the non-linearities; BL, Cms, etc.
There is a measurable reduction in HD when driven by current.
Any opinion on this (supposedly current-drive) amp module?
TDA7294 2 x100W Current Feedback Audio Power Amplifier Board - Free Shipping - ThanksBuyer
TDA7294 2x100W Current Feedback Audio Power Amplifier Board IDE | eBay
TDA7294 2 x100W Current Feedback Audio Power Amplifier Board - Free Shipping - ThanksBuyer
TDA7294 2x100W Current Feedback Audio Power Amplifier Board IDE | eBay
for the first look its feedback path is mixed vfb+cfb,
the question is can we delete vfb circuitry making it only current amp?
the answer is: probably not because of the stability...
the question is can we delete vfb circuitry making it only current amp?
the answer is: probably not because of the stability...
Thanks Pawel
Too bad...
I wander why their is no offering for a current drive amp in the market. With digital active filtering becoming common practive among hifi enthousiasts current drive should become a natural choice in many situations.
Too bad...
I wander why their is no offering for a current drive amp in the market. With digital active filtering becoming common practive among hifi enthousiasts current drive should become a natural choice in many situations.
Current drive amplifiers may not stable into an open circuit. If you turn it on with no load it may not behave well.
If the load is highly capacitive, the amp may oscillate.
Vented box subwoofers will most likely have horrible frequency response.
Closed box or infinite baffle bass alignments will be best for a current amplifier.
I use a current amp to drive a speaker to high excursion at resonant frequency. It is very useful for this purpose.
Nelson has an article that might be of intrest.https://www.passdiy.com/project/articles/current-source-crossover-filters
I have made a lot of testing a few years ago (using a series resistor with a conventional amp) with closed box midbass drivers, as well as vented woofers boxes (with precise EQ...), and even compression drivers.
My conculsion was that I would only want to use current-drive on closed box midbass drivers, where the benefits are the greatests without apparent downside.
Are you planning on releasing a current-drive amp board in your SL Audio shop? That would be great!
It is my desire to either make one or to find an amplifier provider to make one for us. I believe that such an amplifier will be a great benefit for measuring loudspeakers. That such an amp would be of use for listening would be a side benefit.
I have purchased two different LM3886 designs that are supposed to force current. Neither of them perform to my liking.
I think you can get better performance from a discreet design that does not derive the current drive from feedback and an op-amp type circuit, but has an
intrinsic high output Z, and stability sans feedback.
I believe some of Nelson Pass's FirstWatt amps are examples.
I've had some luck with the attached, but haven't pursured it far.
Attachments
Which ones did you buy and what's wrong with them?
Was going to suggest some simple chip possibilities but thought you needed zillion volts & amps 😱
Current drive boards that i tried
Hello krglee
Here are the boards that I tried from ebay
Classic lines LM3886 dual-channel +LM318 AMP/ DIY
2 x 68W Watt LM3886 + NE5532 Audio Amplifier Board DC Serve Current Feedback
The first one is a knockoff of the MyRef. That design is an improved Howland current pump.
The second is a floating load current feedback topology.
Both of these boards had ringing/oscillation with a resistive dummy load. They would not pass a sine wave with any fidelity.
Since they did not come with any documentation I did not do any probing or attempt any trouble shooting. They were both a disappointment.
A 50 watt chip amp that is pure current drive would fill a need for testing small woofers and most mid-bass drivers. Even for huge woofers, 50 watts will reveal a lot of information at Fs.
Hello krglee
Here are the boards that I tried from ebay
Classic lines LM3886 dual-channel +LM318 AMP/ DIY
2 x 68W Watt LM3886 + NE5532 Audio Amplifier Board DC Serve Current Feedback
The first one is a knockoff of the MyRef. That design is an improved Howland current pump.
The second is a floating load current feedback topology.
Both of these boards had ringing/oscillation with a resistive dummy load. They would not pass a sine wave with any fidelity.
Since they did not come with any documentation I did not do any probing or attempt any trouble shooting. They were both a disappointment.
A 50 watt chip amp that is pure current drive would fill a need for testing small woofers and most mid-bass drivers. Even for huge woofers, 50 watts will reveal a lot of information at Fs.
I was offered the first board.
Exactly the same disappointing experience as yours.
It's comes out as a waste of time to put the slightest trust in this kind of products.
I ended by removing the LM318 and modifying the board to make an amp according to the standard application of the manufacturer of the LM3886.
I'm having trouble finding out what MyRef is. Any links?
I can't see anything on the eBay pages that suggest these are Current Sources.
Hi
as pointed out Mr. Esa Meriläinen in here: Amplifier topologies for current-drive | Current-Drive - The Natural Way of Loudspeaker Operation many modern chip amps only LM2030/40/50 did not have the oscillation in current amp aplication.
I think "MyRef" refers to Mauro Penasa's amp, for which there were a few threads, for example :
http://www.diyaudio.com/forums/chip-amps/134726-new-my-ref-rev-c-thread.html
My oscillating board coming from Ebay was a voltage output amp.
http://www.diyaudio.com/forums/chip-amps/134726-new-my-ref-rev-c-thread.html
My oscillating board coming from Ebay was a voltage output amp.
TDA2040 is my favourite small chip amp. It isn't 1ppzillion THD but is very well behaved otherwise. I've used it in several commercial active speakers.
I'm listening to Shostakovitch 10 on one of them now.
It's also by coincidence, the main device I did my Current Drive experiments on more than 30 yrs ago.
This was with simple feedback across 0R1 on the earth end of the speaker. This give good results but Woofer wants to use it for measurement which really needs one side of the speaker earthed.
___________________
MyRef is a normal Voltage source amp. The output stage is a Howland pump but the global feedback makes this irrelevant. http://www.diyaudio.com/forums/chip-amps/54571-my-audiophile-lm3886-approach.html
Anyone got a link to the 'final' schematic?
Woofer, is this what you used for your 'current drive' measurements?
___________________
padamiecki, can you post a scan of how Esa uses his chip amps for Current Drive?
___________________
Woofer, if we can get an accurate schematic for your eBay MyRef, we can extract the Howland pump and give you what you want.
You prepared to take instruction from a beach bum who claims to know what he's doing? 🙂
I'm listening to Shostakovitch 10 on one of them now.
It's also by coincidence, the main device I did my Current Drive experiments on more than 30 yrs ago.
This was with simple feedback across 0R1 on the earth end of the speaker. This give good results but Woofer wants to use it for measurement which really needs one side of the speaker earthed.
___________________
MyRef is a normal Voltage source amp. The output stage is a Howland pump but the global feedback makes this irrelevant. http://www.diyaudio.com/forums/chip-amps/54571-my-audiophile-lm3886-approach.html
Anyone got a link to the 'final' schematic?
Woofer, is this what you used for your 'current drive' measurements?
___________________
padamiecki, can you post a scan of how Esa uses his chip amps for Current Drive?
___________________
Woofer, if we can get an accurate schematic for your eBay MyRef, we can extract the Howland pump and give you what you want.
You prepared to take instruction from a beach bum who claims to know what he's doing? 🙂