Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Chip Amps Amplifiers based on integrated circuits

LM1875 in parallel configuration and used in a composite amplifier.
LM1875 in parallel configuration and used in a composite amplifier.
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 14th February 2020, 10:15 AM   #431
FauxFrench is online now FauxFrench  France
diyAudio Member
 
Join Date: Dec 2017
Location: The mountains, calm and quiet.
Quote:
Originally Posted by papasteack View Post
Oh...you talk about the remote sensing resistor maybe ? It's to compensate wire resistance. It s a 4 wire output amp ^^
Many thanks for your explanation.
  Reply With Quote
Old 14th February 2020, 12:02 PM   #432
kokoriantz is offline kokoriantz  Lebanon
diyAudio Member
 
Join Date: Dec 2015
Location: south east asia
The current amps don't have any voltage gain and they are very stable. They even don't have any ground reference to know at what potential they are flying. Yes the inputs reach the Vee/Vss limits , this why I adjusted the current sense resistors to 0.5 instead of 0.1 so that the master clips before the inputs reach to their limits . With 8 ohms load per IC the output can clip at 22.4v while the current slaves still pouring the same current as the master.
  Reply With Quote
Old 14th February 2020, 01:37 PM   #433
FauxFrench is online now FauxFrench  France
diyAudio Member
 
Join Date: Dec 2017
Location: The mountains, calm and quiet.
Quote:
Originally Posted by Turbowatch2 View Post
If you build a real amp at the end, it will be a great advantage to select the chips for offset and then arrange them into groups with lowest differences. This reduces the resistor value needed at each chips output and the amount of correcting voltage applied.
If you have a larger number of chips, maybe 15 while 12 needed, from the same production run and non fakes, you should be able to eliminate a large portion of the problem. Also, you get a qualified idea how much correction is needed, instead of estimating based on data sheets.

This is one way to reduce the need for adjustment. This was what I did in part to reduce the risk of mismatch in my triple-LM1875. If there is a possibility that the LM1875/TDA2050 are fake, it is almost a necessity as the production lots perform quite differently.

The slight disadvantage, in particular when screening the ICs, is the need for a test-circuit and some test gear. This may be a problem for some DIYers to whom only the most basic tools are available. I understand the wish to have a version where the use of genuine ICs means the possibility to trim the circuit to good function "as is".
For sure, my suggestions for adjustment take wider tolerances than experienced in practice into account.
  Reply With Quote
Old 14th February 2020, 01:46 PM   #434
Turbowatch2 is offline Turbowatch2  Germany
diyAudio Member
 
Join Date: Aug 2013
Location: Northern germany
Maybe one thought about the output offset, which will always be some mV, resulting from the input that hangs at a small voltage. External DC is successfully blocked by a high quality capacitor.
It is no problem for an amp or loudspeaker to have a little voltage at the output. Something, even like like 100 mV is nothing to be concerned on that side.
The problem arises if parallel connected amps have different offsets and each chip wants to bring the output on itīs level.
If we select the chips for identical offset, this problem is minimized, if we presuppose that identical produced objects, mounted on one heat sink will have an identical thermal drift. So we come to matching chips.

*This selecting of parts for some parameter is often made in the voodoo corner of hifi, as it brings some individuals in the favorable situation to virtually sell air for big money.*

Here it has real advantages, as you can minimize the value of the resistor that is supposed to flatten out this problem. One advantage of paralleling chips or transistors, what it is we really do, is the lower output resistance. If we need to increase a resistor to reduce an others resistance, this gets questionable.

Even as individual, active circuits can steer the inputs to zero volt automatically, I would prefer the simpler approach to select chips and then adjust them to 0 volt with a pot.

PS To match the chips a simple eBay amp, upgraded with a Textool socket should be fine. Not even a heat sink needed, if voltage is keep reasonable.

Last edited by Turbowatch2; 14th February 2020 at 01:49 PM.
  Reply With Quote
Old 14th February 2020, 02:52 PM   #435
palstanturhin is offline palstanturhin  Finland
diyAudio Member
 
palstanturhin's Avatar
 
Join Date: Jun 2011
All circuit suggestions most wellcome!
If one can prototype and show results, even better.

I will try put the booster transistors to my proto this weekend and see what happens.
There should be no need to adjust or match anything with that topology at least.
Should be very diyer friendly...
  Reply With Quote
Old 14th February 2020, 03:14 PM   #436
kokoriantz is offline kokoriantz  Lebanon
diyAudio Member
 
Join Date: Dec 2015
Location: south east asia
Quote:
Originally Posted by palstanturhin View Post
All circuit suggestions most wellcome!
If one can prototype and show results, even better.

I will try put the booster transistors to my proto this weekend and see what happens.
There should be no need to adjust or match anything with that topology at least.
Should be very diyer friendly...
You only need to adjust the series resistor to decide up to what power the IC will work in solo . The circuit bellow, the two last pairs start working above 2A for example. I simulated with the composite you used , I saw no any anomalies when the transistors activate, although I chose slowest ones . Of course c5200/a1943 are still much faster ones if you detect any wrong doings .
Attached Images
File Type: png classB.png (250.6 KB, 66 views)
  Reply With Quote
Old 14th February 2020, 03:15 PM   #437
FauxFrench is online now FauxFrench  France
diyAudio Member
 
Join Date: Dec 2017
Location: The mountains, calm and quiet.
Quote:
Originally Posted by Preamp View Post
.................................................. .........
All those problems we're having with LM's not performing to specs, being counterfeits or simply mislabelled TDA's made me think about going back to the first schematic that was posted in this thread by kokoriantz. This circuit is probably not the best performing, but it should be the most forgiving one, while still being better than a plain vanilla 1875. Here's a schematic of a prototype I'm planning to build:

Some of the advantages:

The LM (or TDA or whatever really) is running with 20+ dB inside its specified range (R2/R3, R9/R10). This should work with any part out there, even the cheapest ones without any working protection circuits that are not stable with negative gains.

Output offset is negligible when using capacitors C4 and C5 to AC-couple the feedback path. This makes trimming the whole thing a lot easier. For a more High-End approach, one could easily add a servo circuit at the non-inverting inputs of each LM.
Only the gains of the LM blocks have to be trimmed via R3 or R10, much like in the Linn circuit that was posted above.

C1 and C2 are needed to make the whole thing stable. The values above work with a simple simulation model. I will see if they hold up in the real world. It would be great to find a one-size-fits-all version that works with genuine and counterfeit parts alike.

R7/R8 set the overall gain of the amplifier. Values below 20dB are easily possible. For extremely low gains like 1x, 2x, 3x etc. the capacitors C1 and C2 may need some further adjustment for stability.

C7 removes the input offset of U3. Again, a servo might be added to remove the even the last mV from the output. I consider this to be overkill though.

Now on to the veroboard and soldering iron !
Hi Lasse,
A nice circuit for test!
It is close to the circuit Chris tested but with a controlling OP-AMP loop.
Using the LM1875 with a gain of 10 should make almost any LM1875/TDA2050 work, even the fake ones. With increased LM1875 gain, you will get a little more noise but that is not serious as I see it.
With 10 times LM1875 gain, you push the OP-AMP amplitude working point towards higher THD but also that is not a serious problem. You can eventually adjust your R1/R5/R6/R12 for a lower effective gain.
Your conclusion on the off-set reducing effect of C4 and C5 seems right but at the same time they may cause some problems if any. My fear would be that these capacitors, if triggered by a low frequency signal, could cause a momentary output voltage difference at the two amplifier outputs because the return to balance is slow due to the size of these capacitors. That could be what happened for Chris. Due to the higher gain (10x), matching of gains is more important and due to the capacitors you have to match the gain with an AC signal which is more cumbersome and perhaps less precise. You only need to adjust one gain to match the other as the exact gain is of no importance. I am curious to see if these capacitors give you problems.

For the rest, I have no comments. A good year ago I made some composite amplifier tests with single LM1875. For these tests, adding C2 would only make the whole thing oscillate.

I fully agree with your "one-size-fits-all" approach and at least for now keep it simple and avoid DC-servo's if not really mandatory. There I fully support Turbowatch2's view that bothering about a few Milli-volts is exaggerated.

I very much look forward to hear about your experiences.

NB: You mention a "Linn circuit". Which one is that?

Last edited by FauxFrench; 14th February 2020 at 03:17 PM.
  Reply With Quote
Old 14th February 2020, 03:53 PM   #438
FauxFrench is online now FauxFrench  France
diyAudio Member
 
Join Date: Dec 2017
Location: The mountains, calm and quiet.
Quote:
Originally Posted by palstanturhin View Post
All circuit suggestions most wellcome!
If one can prototype and show results, even better.
.................................................. .......



It is not that parallel LM1875 is necessarily the best choice. Actually, I assume that there is better. We started with LM1875 out of convenience and I believe we have gained very useful information about its behavior and limits. I will continue that branch because I am already well on my way with that concept.

Anybody else who feels like constructing a power part that preferably can operate from a symmetrical supply is very appreciated. Please provide a schematic and test results so we have a selection to choose from depending on skills acquired, power needed and budget.
The OP-AMP part seems more predictable as long as the power part is reliable. Remember that the power part (amplifier) has to be DC-coupled for the best results.

Yesterday, I studied all chip-amps I could think of to find out why the LM3886 is so popular. My conclusion was that it has by far the highest output current limit, good supply range, it is reasonable in price and as a further advantage has a low THD though not a primary concern.

Last edited by FauxFrench; 14th February 2020 at 03:57 PM.
  Reply With Quote
Old 14th February 2020, 03:53 PM   #439
maxhifi is offline maxhifi  Canada
diyAudio Member
 
Join Date: Oct 2005
Maybe you missed my post #398, where I showed how Linn paralleled two TDA1514A in the Majik, and discussed this.
  Reply With Quote
Old 14th February 2020, 04:00 PM   #440
FauxFrench is online now FauxFrench  France
diyAudio Member
 
Join Date: Dec 2017
Location: The mountains, calm and quiet.
Quote:
Originally Posted by kokoriantz View Post
You only need to adjust the series resistor to decide up to what power the IC will work in solo . The circuit bellow, the two last pairs start working above 2A for example. I simulated with the composite you used , I saw no any anomalies when the transistors activate, although I chose slowest ones . Of course c5200/a1943 are still much faster ones if you detect any wrong doings .

The heavy duty version!
  Reply With Quote

Reply


LM1875 in parallel configuration and used in a composite amplifier.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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Composite LM1875 MiniAmp and HeadphoneAmp bkercso Chip Amps 3 20th November 2019 12:49 AM
Composite LM1875 stability issues Glkjh Chip Amps 5 8th June 2019 03:32 PM
Speaker Parallel Configuration anoopkrajan Multi-Way 3 10th February 2018 02:10 PM
Need help with Triple Parallel LM1875 dynamics amplifier. danielwritesbac Chip Amps 146 9th July 2014 06:27 PM
Honeycomb composite panel enclosure, dual opposed configuration rhapsodee Subwoofers 6 1st September 2010 04:26 PM


New To Site? Need Help?

All times are GMT. The time now is 03:10 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Copyright ©1999-2020 diyAudio
Wiki