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

Parts Where to get, and how to make the best bits. PCB's, caps, transformers, etc.

Quad or Series/Parallel Resistors as an alternative to traditional power resistors
Quad or Series/Parallel Resistors as an alternative to traditional power resistors
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 March 2018, 11:08 PM   #11
lehmanhill is offline lehmanhill  United States
diyAudio Member
 
lehmanhill's Avatar
 
Join Date: Jan 2008
Location: Chelsea, Michigan
Default Slightly More Traditional

In honor of DF96, the more traditional way of looking at this issue would be to use two resistors in series, for example wirewound each at half the value of the original resistor.

-low current noise - check
-let's assume no inductance issues
-because each resistor is at half voltage, voltage distortion =-12 dB
-each resistor is at half the power, so power distortion is -6 dB from a single resistor.
In my Post #1 example, the wirewound Vishay RWM410 would have power distortion at -37 dB.

Overall, of the traditional power resistors, this is an excellent option.

Of course, you could also use two resistors in parallel with double the value. It would have all the benefits of the series example above, except for the voltage distortion advantage.

Naturally, two resistors in either series or parallel, it will require more pcb space.
  Reply With Quote
Old 29th March 2018, 11:45 AM   #12
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
I have looked at the calculations in post 1. The arithmetic looks OK, but the conclusions may be misleading. A 'distortion' figure of -18dB sounds horrifying, but this is not non-linear distortion. It is merely a small shift in a resistor value due to the signal envelope. This may or may not affect the audio. Whether it responds to bass too depends on the thermal time constant of the resistive element, which the calculation does not address. Typically, a smaller resistor will have a bigger effect. Given the choice between a big wirewound (which may respond to signal envelope) and four ordinary resistors (which may respond to bass) I would choose the wirewound. Two ways to check: detailed measurements, detailed multiphysics simulation.
  Reply With Quote
Old 29th March 2018, 02:18 PM   #13
lehmanhill is offline lehmanhill  United States
diyAudio Member
 
lehmanhill's Avatar
 
Join Date: Jan 2008
Location: Chelsea, Michigan
Quote:
Originally Posted by DF96 View Post
I have looked at the calculations in post 1. The arithmetic looks OK, but the conclusions may be misleading. A 'distortion' figure of -18dB sounds horrifying, but this is not non-linear distortion. It is merely a small shift in a resistor value due to the signal envelope. This may or may not affect the audio. Whether it responds to bass too depends on the thermal time constant of the resistive element, which the calculation does not address. Typically, a smaller resistor will have a bigger effect. Given the choice between a big wirewound (which may respond to signal envelope) and four ordinary resistors (which may respond to bass) I would choose the wirewound. Two ways to check: detailed measurements, detailed multiphysics simulation.
Fair enough. The -18 dB has a lot to do with the small resistance value I chose, so I am definitely using a worst case. In a situation where the resistor is 1k, this isn't even worth discussing. But that small shift in resistance isn't so small relative to the 330 mOhms of the example. So it will depend on the circuit whether the resistance of the small value resistor as significant effect on the output. I guess you could say that one reason for starting this thread is that power distortion in resistors or resistor noise and distortion isn't even considered most of the time and that there will be situations where you should at least check.

As for what kind of distortion and whether it affects audio, I can't give you an objective answer at this point. I suspect that is the same as voltage distortion which Groner describes as primarily 3rd harmonic. In future, I will attempt distortion measurement in a circuit, but the results will be largely circuit dependent.

Your point about the time constant of the resistive element is also worth considering. Unfortunately, there is little information available. But let's think about this a little.

I think we can assume that the thermal mass of the wirewound resistive element is a high percentage of the thermal mass of the total resistor and that's a good thing slowing down the change in resistance with power. Therefore, the wirewound should have less modulation in response to musical transients and variation than other resistors.

With wirewound, it's easy to use a higher wattage rating resistor. I could easily have used a 10W instead of a 3W in the example and moved the power distortion calculation in post 1 from -31 to -47 dB for the wirewound. That, combined with the less modulation discussed above would be an effective choice in a circuit where this resistor position could affect sound quality.

One of the interesting things about the Quad using some of today's power SMD resistors is that those resistors have been optimized to maximize heat rejection, much of it through the pcb. The Susumu KRL for example is a face down resistor with oversize pads on the long dimension so that heat in the resistive element has the shortest, lowest thermal resistance path to the pcb. So now you have a low thermal mass or fast heating resistive element matched with fast thermal transfer to the heat sink (PCB). With my handmade pcb acting like a heatsink of about 7.5 C/W, the resistors are kept relatively low temperature. A different approach than the wirewound, but an interesting one.

As a side note, there are a lot of power SMD that are thick film and, personally, I wouldn't choose them. Thick film is relatively poor for current noise and others have measured higher distortion in specific circuits using thick film SMD resistors.

The main point of this thread is to spark the discussion of how power resistors are different than ordinary signal resistors. It is also to introduce the Quad approach which I feel is an interesting alternative. I understand that you prefer wirewound, have valid reasons for your choice, and I respect that choice.
Attached Images
File Type: jpg KRL FE R3.JPG (220.7 KB, 162 views)
  Reply With Quote
Old 29th March 2018, 04:59 PM   #14
sumotan is offline sumotan  Indonesia
diyAudio Member
 
Join Date: Oct 2013
If your curious or adventurous may I introduce you to Powertrons.
Am using it now on my ML Odyssey. All I can say is compared to wirewounds
be it Mills, Ohmite or M Resist Supreme, it is heads above them, very smooth
& transparent . I wouldn’t dare say it adds nothing to the sound but as compared
to wire wounds, it clearly adds nothing.

Cheers
  Reply With Quote
Old 29th March 2018, 07:44 PM   #15
lehmanhill is offline lehmanhill  United States
diyAudio Member
 
lehmanhill's Avatar
 
Join Date: Jan 2008
Location: Chelsea, Michigan
Interesting. I was not familiar with them. I looks like they have F and U series which are foil types and K and N series which are thick film. The foil types are likely similar to the texas components CSNG listed in the example, post 1. I am guessing that they are expensive, yes?
  Reply With Quote
Old 29th March 2018, 08:15 PM   #16
sumotan is offline sumotan  Indonesia
diyAudio Member
 
Join Date: Oct 2013
Nope they’re not, only around Usd 8.50 per pc.
Im using the FPR2-T218.
  Reply With Quote
Old 29th March 2018, 08:21 PM   #17
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
Unfortunately thermal time constants are often omitted from datasheets, yet I expect competent manufacturers to know what they are as they would have needed to know them to design the resistor.
  Reply With Quote
Old 29th March 2018, 10:13 PM   #18
scott wurcer is online now scott wurcer  United States
diyAudio Member
 
scott wurcer's Avatar
 
Join Date: Jan 2004
Location: limbo
Quote:
Originally Posted by lehmanhill View Post
-because each resistor is at half voltage, voltage distortion =-12 dB
-each resistor is at half the power, so power distortion is -6 dB from a single resistor.
You're mixing dB's here. The primary distortion is from power (temperature coefficient) a function of V squared. The signal of interest is the voltage so the -12dB is the only relevant number.
__________________
"The question of who is right and who is wrong has seemed to me always too small to be worth a moment's thought, while the question of what is right and what is wrong has seemed all-important."
  Reply With Quote
Old 30th March 2018, 12:10 AM   #19
trobbins is offline trobbins  Australia
diyAudio Member
 
trobbins's Avatar
 
Join Date: Feb 2009
Location: Melbourne, Oz
Why would you use a parallel/series configuration of 4 resistors, when based on the logic in the first post a series connection of 4 resistors is the superior "quad".
  Reply With Quote
Old 30th March 2018, 01:51 PM   #20
lehmanhill is offline lehmanhill  United States
diyAudio Member
 
lehmanhill's Avatar
 
Join Date: Jan 2008
Location: Chelsea, Michigan
Quote:
Originally Posted by scott wurcer View Post
You're mixing dB's here. The primary distortion is from power (temperature coefficient) a function of V squared. The signal of interest is the voltage so the -12dB is the only relevant number.
My mistake, at least in part. I should have used the square of the ratio in the power distortion calculations. But let's test my understanding. I am leaning on the Groner paper for theory. On page 19 of the paper, he talks about using the Quad and metal film resistors in his project. He refers to the Quad and the -12 dB as only relating to the voltage coefficient of the resistor, that is, the voltage coefficient is the change in resistance as a function of voltage at a constant temperature. He then goes on to talk about the need to include distortion from the power coefficient due to temperature change. He is saying that the voltage distortion and power distortion are separate and independent.

If that's true, then I should correct the power distortion numbers. The new calculation would be 20Log((delta R from temp @ a given power/nominal R)^2). Power distortion for the resistor itself in my post 1 example would be;

Metal Ox (Panasonic) -36 dB
Thick film (Caddock) -45 dB
Wirewound (Vishay) -62 dB
diy Quad SMD (Susumu) -106 dB
Z foil (texas components) -187 dB

More reasonable numbers. Still showing that, depending on how the change in resistance of the single resistor affects the specific circuit, there are noticeable differences in power distortion, plus voltage distortion and current noise differences remain. By the way, I also found an error in the Z foil calculation. The datasheet quoted a number that was using a heatsink. I corrected for the resistor in free air.

Any more mistakes? I am prone to them:-)

Last edited by lehmanhill; 30th March 2018 at 02:20 PM. Reason: correction of data
  Reply With Quote

Reply


Quad or Series/Parallel Resistors as an alternative to traditional power resistorsHide 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
Bass Driver & LPads, Series & Parallel Resistors lordtarquin Multi-Way 4 7th October 2017 05:58 AM
Best Way To Install Parallel Resistors Winston Smith Parts 14 16th October 2015 05:55 PM
Add (or delete) resistors in series with opamp's power supply rails ygg-it Chip Amps 8 30th November 2014 04:07 AM
Quad II: Which resistors? johnm Tubes / Valves 5 19th September 2006 01:52 PM
Opinions about Caddock MV series power resistors MEnsing Parts 6 16th October 2002 09:33 AM


New To Site? Need Help?

All times are GMT. The time now is 05:15 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki