jacco vermeulen said:
remind me not to try that!! 800w of heat!!! Holy dooley! lol, that's a LOT of heat! i'll stick to the KSA-50 clone 😉 (speaking of which i've ordered a stack of "samples" 😉 So it should be great!
I've found out the problem with the rouge runaway on one of the outputs!!! NO HEATSINK PASTE!!! lol, only the silicon piece of isolation stuff- nothing else! No wonder it was getting so hot and also going so high in the bias!!!!! Oh well, easily sorted anyway!! I am going to put up photo's of a previous mod i did to remove voltage drop on the power supply (through the board and also from emittor resistors to the speaker wires - the pcb could NOT handle the power and it involved 10ga wiring 🙂)....
now i am very happy as i found out WHY it was doing what it did!!!
Aaron
now i am very happy as i found out WHY it was doing what it did!!!
Aaron
Normally, they wouldn't, however only a very small portion of the actual chip was touching the silicon rubber!!! It would have need paste (or have been installed correctly in the first place) to not need paste... I'm hoping this is the problem as this chip stood way out for bias and was going into pretty much thermal runaway!
Thanks
Aaron
Thanks
Aaron
bad mouting
Aaron,
You are lucky the transistor survived. I recently fixed a car amp that had a similar problem, worked for years before the guy really turned it up one day, then poof, the loose transistor let go and shut the whole shooting match down.
It could be bad assembly or another thing to bear in mind is that steel and aluminium have different coefficients of expansion, and steel hardware can actually work itself loose over years of thermal cycling. Some manufacturers use locktite or a similar adhesive to gooberize the washers, nuts, to stop the things undoing over time, and of course nylock nuts, split washers any of these type of things would help. Of course cheap manufacturers don't do anything and use a self tapping screw directly into the aluminium heatsink, absolutely the worst possible case in a car amp, thermal cycling and vibration, can you say built in failure mode?...
Stuart
Aaron,
You are lucky the transistor survived. I recently fixed a car amp that had a similar problem, worked for years before the guy really turned it up one day, then poof, the loose transistor let go and shut the whole shooting match down.
It could be bad assembly or another thing to bear in mind is that steel and aluminium have different coefficients of expansion, and steel hardware can actually work itself loose over years of thermal cycling. Some manufacturers use locktite or a similar adhesive to gooberize the washers, nuts, to stop the things undoing over time, and of course nylock nuts, split washers any of these type of things would help. Of course cheap manufacturers don't do anything and use a self tapping screw directly into the aluminium heatsink, absolutely the worst possible case in a car amp, thermal cycling and vibration, can you say built in failure mode?...
Stuart
Re: bad mouting
Of course they put features in like that, if they lasted forever, who would they sell their units too?
😀 😀
Like i said this amp DOES have it's "cheap" parts and they are blindingly obvious!!! Once i'm done modding this and my Krell power amp is built (crosses fingers that it works!) i will move the peavey to the sub duties and sell this amp on, one day.... 🙂 I will make sure i get the enjoyment out of it in the meantime!
Aaron
Stuart Easson said:
Of course they put features in like that, if they lasted forever, who would they sell their units too?
😀 😀
Like i said this amp DOES have it's "cheap" parts and they are blindingly obvious!!! Once i'm done modding this and my Krell power amp is built (crosses fingers that it works!) i will move the peavey to the sub duties and sell this amp on, one day.... 🙂 I will make sure i get the enjoyment out of it in the meantime!
Aaron
As a sort of final update to this (nope, it's not blown 🙂) i am going to move the Vbe multiplier onto the top of the hottest chip OR right hard up against it.... To be honest, it works fine, for short full power bursts, it's great, for continuous running with, say a 28hz tone (into sub), it gets hot hot hot and bias rockets up, it does take quite a while for the vbe multiplier to get the heat of the sink, but with short bursts the hottest output will hit 70deg while the heatsink is still only 50 or so, After i rebuild the other channel i'm going to make sure it's properly on the heatsink and have the fan blowing air through the heatsink so the Vbe multiplier is at the HOTTEST end... Lets say it starts off at 1mv-2mv bias (at 25deg heatsink temp) and after a quick warm up, say, a few big bass notes) bias is up to 7-8mv, but with the fan on quickly drops... So i'm going to use the fans always on at low speed with 45 deg sensors (already mounted on the heatsinks) to kick them into full flow once the whole heatsink has reached a level that allows them to run at the right bias.... for full power bursts the chips reach 70deg or so and bias of about 40mv (!!!) but this drops back to 10mv VERY quickly (talking a matter of 5-10seconds!) so it's obvious to me the heatsink simply can't get rid of the heat quick enough (or conduct from the outputs to the heatsink)....Overall (considering i rarely run it at full power with tones for even 30 seconds) it should be ok, but more testing to do as it runs 2 ohm/channel, so gets quite warm!!
Thanks for all the help guys!
Aaron
P.S. I never thanked the person who posted up the zip file for vbe info - Thanks VERY much!
Thanks for all the help guys!
Aaron
P.S. I never thanked the person who posted up the zip file for vbe info - Thanks VERY much!
You will get better temperature compensation if you can reduce the Ic of the Vbe multiplier to about 0.5mA
This may be achieved by the addition of an extra buffer, and placing a resistor of suitable value between its base and emitter to adjust the Ic of the main transistor
I use this technique in a 180W/4ohm prototype [with a single pair of output devices] and the biasing actually decreases at higher temperatures
I do not recommend coupling the Vbe multiplier to the plastic case of any transistor. Plastic has quite high thermal resistance and is ambient cooled, so at the surface of the case you get a temperature much lower than the actual temperature of the die. Place the Vbe multiplier on the heatsink near the hottest transistor instead
This may be achieved by the addition of an extra buffer, and placing a resistor of suitable value between its base and emitter to adjust the Ic of the main transistor
I use this technique in a 180W/4ohm prototype [with a single pair of output devices] and the biasing actually decreases at higher temperatures
I do not recommend coupling the Vbe multiplier to the plastic case of any transistor. Plastic has quite high thermal resistance and is ambient cooled, so at the surface of the case you get a temperature much lower than the actual temperature of the die. Place the Vbe multiplier on the heatsink near the hottest transistor instead
Hi,
Doug Self published a series of results comparing location of Vbe multiplier. His conclusion was for fastest speed of response mount transistor on the back of one of the plastic output transistors. Medium speed was face to face on 2 sides of a common heatsink and slowest was adjacent to the output transistor. This applies to darlington output or single stage output.
If you use sziklai connected output then the Vbe must monitor the driver or predriver temperature.
regards Andrew T.
Doug Self published a series of results comparing location of Vbe multiplier. His conclusion was for fastest speed of response mount transistor on the back of one of the plastic output transistors. Medium speed was face to face on 2 sides of a common heatsink and slowest was adjacent to the output transistor. This applies to darlington output or single stage output.
If you use sziklai connected output then the Vbe must monitor the driver or predriver temperature.
regards Andrew T.
Why not find out for yourself ?
A lot of digital cameras can be set to heat mode.
My bet would be to place the transistor of the Vbe multplier at the hottest spot.
Whether plastic or alloy, that is where dQ/dT is highest.
A lot of digital cameras can be set to heat mode.
My bet would be to place the transistor of the Vbe multplier at the hottest spot.
Whether plastic or alloy, that is where dQ/dT is highest.
I was tempted to try it, i've got a K-type probe on my multimeter so makes it heaps easy to find out temps, the plastic casing doesn't get as hot as the dye, but definately gets HOT about 10-15deg off the dye...... the heatsink can be 30deg off the dye if it's a quick burst
Aaron
Aaron
my amps
hey, NUTTTR, you shoulda bought my restored yamaha p-2200, it went off its head!
as far as the soundking amp goes, i use one the same for discos and ****, what tends to happen lately, is i turn it on, it will run, but when i try to turn the volume up its protection turns on and the amp turns the sound off. Its really annoying!!! the only way to get around it is, to wait a while, or the better solution, give it a sharp technical tap.
It really annoys me because i paid $470 for it not two years ago. and its hardly had any use. and i just sold my p-2200 which had incredible bass and clear sound, i think i sold the wrong one! If at all possile, how would i go about getting more bass out of this, is it possible to just replace the caps, like instead of the **** 7000uf caps in it, put a 22,000uf or 30,000uf cap. (is it that simple?) The 30 year old yamaha had 22,000uf caps 100v!
any suggestions highly welcomed.
Daniel
hey, NUTTTR, you shoulda bought my restored yamaha p-2200, it went off its head!
as far as the soundking amp goes, i use one the same for discos and ****, what tends to happen lately, is i turn it on, it will run, but when i try to turn the volume up its protection turns on and the amp turns the sound off. Its really annoying!!! the only way to get around it is, to wait a while, or the better solution, give it a sharp technical tap.
It really annoys me because i paid $470 for it not two years ago. and its hardly had any use. and i just sold my p-2200 which had incredible bass and clear sound, i think i sold the wrong one! If at all possile, how would i go about getting more bass out of this, is it possible to just replace the caps, like instead of the **** 7000uf caps in it, put a 22,000uf or 30,000uf cap. (is it that simple?) The 30 year old yamaha had 22,000uf caps 100v!
any suggestions highly welcomed.
Daniel
She do not care about us Jacco..let's have some drink together.
And we will turn drunk...hehe.
Eva is an explendid engineer, and related audio, she know everything, unfortunatelly cannot see us Jacco!
I apreciate very much all she said.
Other guys are very clever too, but she is experienced.
Let's turn drunk Jacco, till a dog leak our face when we will tell to the dog...oh my dear!..ahahahaha!
Just a break boys... to care heart problems...let's go Vermeullen!
grumpf...ela é minha, ví primeiro! (of course kidding..... really?)
Carlos
And we will turn drunk...hehe.
Eva is an explendid engineer, and related audio, she know everything, unfortunatelly cannot see us Jacco!
I apreciate very much all she said.
Other guys are very clever too, but she is experienced.
Let's turn drunk Jacco, till a dog leak our face when we will tell to the dog...oh my dear!..ahahahaha!
Just a break boys... to care heart problems...let's go Vermeullen!
grumpf...ela é minha, ví primeiro! (of course kidding..... really?)
Carlos
- Status
- Not open for further replies.