diyAudio (
-   Car Audio (
-   -   how much capacitance can i add safely? (

Clipped 24th September 2007 01:02 PM

how much capacitance can i add safely?
going to change the caps out in the output section of a 2150sx,

whats the general rule on the limit to safely add capacitance?
theres currently 8000uf (8 x 1000uf) in capacitors.

if theres too much capacitance will the current rush fry the transistors? :hot:

or how about using cyclon cells instead of caps? or even using those caps they use in that monster rockford amp?

closing the bottom cover is not a concern, i'll make a custom bottom cover with feet turned on a lathe.


my newest victims:

Perry Babin 24th September 2007 05:47 PM

Why are you adding more capacitance?

Clipped 25th September 2007 04:25 AM

i have a feeling this is a trick question :D

the old caps need to be changed...a few are swollen on top.

also i have a 225 hcca which i replaced with 1800uf caps and the bass hit harder afterwards.

it wasnt just do to new caps, i tried 1000, 1200, 1500 and 1800 all new caps, and the higher i went the tighter/harder the bass was consistently...but at a loss of smoothness when playing fullrange.

the old mmats mod also adds capacitance to the rails...somewhere i think between 200,000-400,000 uf.

Perry Babin 25th September 2007 04:47 AM

I've tried adding capacitance to amps before and there was essentially no difference in the output. Look at the rail voltage (with an oscilloscope) while driving the amp hard. If you see high frequency noise on the rails, you need more capacitance. If you see nothing other than ripple directly related to the audio signal peaks, more capacitance isn't likely to be an improvement. If you look at the waveform on the power supply FETs, it's likely going to have the same audio-related ripple. This is generally due to ripple at the B+ input. Using large power wire and an adequate charging system can eliminate most of this ripple.

I've worked on some of the MMATS amps and the extra capacitance was needed because the amp wasn't stable without it. This generally applied to the class D amps that used the HIP4080 IC. I'm not sure which amps you're referring to.

Clipped 25th September 2007 05:12 AM

eventhough no difference in actual measurements, did it sound different to the ears?

i had that amp on a pair of 6x9's...maybe the difference wouldnt have been as prominent on subs.

Perry Babin 25th September 2007 05:37 AM

The 'stability' I refer to is not the 2-ohm stable type of stability. The amps would not operate properly without the added capacitance. The newer ones have much more internal capacitance than the old ones.

I never rely on my ears to determine the results of tests. I can look at the output on a scope and measure the output and see that there is essentially no difference. Even if it helped add a fraction of a volt to the output, that would not be audible at 300+ watts of power. I love dealing with audio equipment but some the the subjective test results (no measurements) that some people believe they hear amaze me. Changing op-amps is the worst. People change from a good op-amp to one with a bit better specs and claim to hear a world of difference. It's simply not possible unless the old op-amps were defective. I guess if it gets more people interested in the hobby, it's a good thing but junk science really annoys me.

To make your amplifier usable into 2 ohms mono, you could try reducing the rail voltage. This would reduce the power dissipation across the output transistors so they wouldn't run too hot and fail.

If you do this, the B+ filter capacitors will be seeing more ripple and may run hotter than normal. You would need to monitor their temperature to be sure they don't overheat.

You would also need to monitor the temperature of the power supply transistors. At reduced rail voltage, the drive voltage will be a bit lower (on average) and the power supply FETs may heat up more than normal. If the power supply end of the amp heats up faster than the audio end of the amp, this may not be a practical modification for this amp.

1moreamp 25th September 2007 07:04 AM

Years ago I did the big C change and most people did say it sounded tighter and cleaner on the bass line area. Any mid range and highs change I attributed to overall power supply noise reduction. I have heard others here describe certain high end sound quality factors being attributed to high frequency oscillations and power supply noise issues.

I found the amps I tested ran cooler, and exhibited a more robust character then before the change. But I attributed this to a much more stable power supply.
Plus the original caps were probably damaged by heat and use.

Newer cap designs and chemistry have brought about a world of change. Todays caps typically have 3 to 4 times the capacitance in the same size package as the caps of the early 90's.

Perry's testing is impeccable as usual, and I myself did find rail drops of up to 2 volts not uncommon when loading the power supply with excessive cap value. So on your old school Orion, and PPI, and Autotek yes a noticeable aural difference might be had by changing the original engineering, IMHO you just might be replacing defective components also..

Caps back then were not cheap as today so it was common practice to build a power supply that needed less storage. It saved money and space inside the amp case.

My trial and error taught me not to exceed a factor of two point five or three when adding C value. There were no sonic gains to be had above that limit and Thermal issues also reached a zenith at that amount of change.

Nowadays older and hopefully more wiser I tend not to alter original engineering when possible. It just opens too many doors to too many issues. Especially with todays higher power levels, you can be opening a real can of worms.
Amps back then had +&-25 to 35 volts. Nowadays I see +&- 60 to 80 volts common place. Performing such a upgrade to a high voltage amp might not really achieve any real word gains. On lower rail amps I believe the sonic effects you heard are more pronounced. Possibly due to simple energy storage and delivery at lower voltage rail values.

I think this topic is also related to the different sound qualities of the two different power supply schemes used over the years.

PPI used a regulated power supply that would deliver rated power from 11 to 15 volts regardless of your cars varying power output. While others ran unregulated power supplies so their amp delivered more power at 14 volts then at 12.

Adding large C values adds regulation or stiffness to a power supply, It also allows the cars supply to take softer hits, so the overall effects can be interesting to say the least.....:)

I am not disagreeing with Perry, but I am sharing my early learned knowledge about this highly subjective issue. I hope its taken with a grain of salt also.....been there, and done that, Hindsight says if you can't quantify the subjective difference then its hard to prove any real world value from a engineering point of view..........

Clipped 25th September 2007 08:23 AM

thanks guys,

Perry I went back and edited my post after remembering what you told me about stability and oscillation, i didnt want people to misunderstand so i deleted that phrase.

the deleted part read - 'can i make a 2150sx more stable if i add more capacitance'

jnb 28th September 2007 11:40 PM

Since you have to change them out anyway, you may find that the new caps have different characteristics regardless of total capacitance, the effects of which may not be known yet (but which may also be nil). I doubt that adding capacitance on its own will be a major issue, but making it only a slight increase may be better for the preceding power supply vis-a-vis inrush. I think there is a sonic benefit to be had, considering you are going to be watching for problems anyway, and helped by the fact that electro's are smaller now.

All times are GMT. The time now is 12:04 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