I've run one channel fully loaded from the TK2050 and two channels around half load.This was with a Zalman ZM-NB32K (the orange one).
The temperature was around 65-70 degrees Celsius. Just on the edge of able to touch it. According to the datasheet, nothing to worry about.
Why does everyone is so afraid of "hot" chips?
Never looked at computer chips?
@v-bro.
Yes, I know that there is always a switching wave (frequency). But it's a switching wave with almost no current. I think it must be a bad dead-time-delay setting and mosfet choice or so.
If the STA505 is running better, I'm looking forward to upgrade it.
Removing the TP2050 is only a pain. (read: the PCB quality could be better)
Maybe we can persuade Sure Electronics to use also the STA505
The temperature was around 65-70 degrees Celsius. Just on the edge of able to touch it. According to the datasheet, nothing to worry about.
Why does everyone is so afraid of "hot" chips?
Never looked at computer chips?
@v-bro.
Yes, I know that there is always a switching wave (frequency). But it's a switching wave with almost no current. I think it must be a bad dead-time-delay setting and mosfet choice or so.
If the STA505 is running better, I'm looking forward to upgrade it.
Removing the TP2050 is only a pain. (read: the PCB quality could be better)
Maybe we can persuade Sure Electronics to use also the STA505
Because there wasn't anyone that knows suggesting it? Because a HSF change can be thought of and carried out by anyone?
And for me because I have 3 of such heatsinks lying around?
Nope, it isn't!
An externally hosted image should be here but it was not working when we last tested it.
At Conrad.nl for less than 4 eur! Just heat it until it comes 'floating' off...
Or use a somewhat more expensive method, but this ensures you can take it off in working condition:
Chip Quik: Easy removal of surface mounted devices
YouTube - Using ChipQuik to Desolder Surface Mount Components
P.S. I'm not at all worried about chips getting hot within spec as long as I know the heat is from doing actual work. If it is just from doing nothing and shoot through is the reason it's not right just to cool it because that's not curing the cause... And I strongly suspect the problem is easier solved at the source. I replace a chip like that in less than 10 minutes! Placing a gigantic and not original heatsink and especially the figuring out how takes me a LOT longer...
Last edited:
Yep, I've just measured it, 5.01v. If I push the fan blades a little, the fan spins up ok. I think the fan normally runs from 12v, so 5v is barely enough to get it spinning... It's strange how it worked when it was powered by the Sure PCB however... Maybe it provided a more stable 5v?
Today I tried powering up my SMPS without an RCD at the plug socket. I was using an RCD yesterday as I was outside when soldering things... With no RCD in the mains circuit, the fan appears to spin up every time! Most strange.
I plan to use the Sure board in the house the whole time, so I think I can use it without an RCD for now. My next step was to add some smoothing to the output of the l7805cv, as right now there is none. I think I may add some anyway, it can't hurt.
No cap?
You don't have any cap after the new regulator? Add a cap and see what happens.Yep, I've just measured it, 5.01v. If I push the fan blades a little, the fan spins up ok. I think the fan normally runs from 12v, so 5v is barely enough to get it spinning... It's strange how it worked when it was powered by the Sure PCB however... Maybe it provided a more stable 5v?
Today I tried powering up my SMPS without an RCD at the plug socket. I was using an RCD yesterday as I was outside when soldering things... With no RCD in the mains circuit, the fan appears to spin up every time! Most strange.
I plan to use the Sure board in the house the whole time, so I think I can use it without an RCD for now. My next step was to add some smoothing to the output of the l7805cv, as right now there is none. I think I may add some anyway, it can't hurt.
Tweeter pad options
I remember reading with many Tripath amps that the output stage is often a compromise so it works best with a 4ohm or 8ohm load. I remember reading that a 6ohm load is ideal for these amps.... Can anyone confirm if this is true?
Is this the case with the Sure TK2050 amp? I've finally got a chance to fire up my Sure TK2050 powered speakers today, and I'm getting a little too much hiss thru the tweeter. I think my tweeters are quite sensitive, but I'm not sure of the exact figure as I cannot find detailed specs for my drivers. I was getting the same problem when I was using a TA2020 based amp, and I found that padding down the tweeter with a resistor help as it:
1) reduced the white noise/hiss
2) made the tweeter level about the same level as the woofer
Now my tweeter is 8 ohm and I found that an 8 ohm resistor in series gave a -6db drop and matched the level of the woofer quite well, apart from the fact that the amp was now seeing a 16ohm load. It did sound different when a resistor was used, but I was using a cheap 7w ceramic resistor which I just happened to have lying around, so I'm not sure if it was the resistor or the 16ohm load to the amp that caused the change in sound.
Now my main concern here is that a 16ohm load could produce a non-linear frequency response, because this amp is intended for use with 4 or 8 ohm loads. Would I be right in saying this?
Now I must admit I've not tried to sim this, and I'm not really sure how to, but I'm using Arjen's 10uH inductors for the tweeters (along with 4.7uH XXL's for the woofers), and all other parts in the output stage are stock... If anyone could run this thru Excel or Spice or whatever they use to sim this I would be most grateful.
If there is a bad frequency response from 2KHz+ I was thinking a 2 resistor L-pad could be used here, and if I do this I have the option to make the amp see a 6 or 8 ohm load with the correct resistor values... What would be the ideal target impedance value to aim here?
Apologies for the long post!
I remember reading with many Tripath amps that the output stage is often a compromise so it works best with a 4ohm or 8ohm load. I remember reading that a 6ohm load is ideal for these amps.... Can anyone confirm if this is true?
Is this the case with the Sure TK2050 amp? I've finally got a chance to fire up my Sure TK2050 powered speakers today, and I'm getting a little too much hiss thru the tweeter. I think my tweeters are quite sensitive, but I'm not sure of the exact figure as I cannot find detailed specs for my drivers. I was getting the same problem when I was using a TA2020 based amp, and I found that padding down the tweeter with a resistor help as it:
1) reduced the white noise/hiss
2) made the tweeter level about the same level as the woofer
Now my tweeter is 8 ohm and I found that an 8 ohm resistor in series gave a -6db drop and matched the level of the woofer quite well, apart from the fact that the amp was now seeing a 16ohm load. It did sound different when a resistor was used, but I was using a cheap 7w ceramic resistor which I just happened to have lying around, so I'm not sure if it was the resistor or the 16ohm load to the amp that caused the change in sound.
Now my main concern here is that a 16ohm load could produce a non-linear frequency response, because this amp is intended for use with 4 or 8 ohm loads. Would I be right in saying this?
Now I must admit I've not tried to sim this, and I'm not really sure how to, but I'm using Arjen's 10uH inductors for the tweeters (along with 4.7uH XXL's for the woofers), and all other parts in the output stage are stock... If anyone could run this thru Excel or Spice or whatever they use to sim this I would be most grateful.
If there is a bad frequency response from 2KHz+ I was thinking a 2 resistor L-pad could be used here, and if I do this I have the option to make the amp see a 6 or 8 ohm load with the correct resistor values... What would be the ideal target impedance value to aim here?
Apologies for the long post!
Thanks, I've now added 100nF to the output leg and 330nF to the input leg of the L7805 and the fan spins up every time, RCD or no RCD.
Last edited:
47uf
I was thinking more like 47uf at least before and after the new regulator.
Thanks, I've now added 100nF to the output leg and 330nF to the input leg of the L7805 and the fan spins up every time, RCD or no RCD.
I was thinking more like 47uf at least before and after the new regulator.
The Sure datasheet or the L7805 datasheet? I only got a chance to have a quick look at the datasheet for the 7805 before deciding to for the smaller values... On page 4 the 'test circuit' shows 0.33uF on the input and 0.1uF on the output... which is why I went for these values. Here's the datasheet I was looking at:
http://focuslab.lfp.uba.ar/Extension/BDS2005/7800.pdf
Anyway, I know I've got some 47uF caps somewhere, I'll go for these instead.
EDIT: In fact after looking at the datasheet again it does seem to mention that more capacitance on the output will improve transient response, so I'd like to think that a better transient response will mean the fan is less likely to stall when being powered up. Thanks!
Last edited:
Why is the MeanWell fan running?
Why is the MeanWell fan running? What all are you powering with it? My fans have never run on either of the S-350-36 power supplies that I have been using for 6 months. Is something wrong? My supplies only measure 35c inside the fan outlet.
Why is the MeanWell fan running? What all are you powering with it? My fans have never run on either of the S-350-36 power supplies that I have been using for 6 months. Is something wrong? My supplies only measure 35c inside the fan outlet.
The Sure datasheet or the L7805 datasheet? I only got a chance to have a quick look at the datasheet for the 7805 before deciding to for the smaller values... On page 4 the 'test circuit' shows 0.33uF on the input and 0.1uF on the output... which is why I went for these values. Here's the datasheet I was looking at:
http://focuslab.lfp.uba.ar/Extension/BDS2005/7800.pdf
Anyway, I know I've got some 47uF caps somewhere, I'll go for these instead.
EDIT: In fact after looking at the datasheet again it does seem to mention that more capacitance on the output will improve transient response, so I'd like to think that a better transient response will mean the fan is less likely to stall when being powered up. Thanks!
I think he refers to the 78xx datasheet. They do need an output cap. When the reg is located very near the large filter cap an MKT input cap can be omitted but using one won't hurt. Without output cap the thing will likely oscillate. That has nothing to do with transients ! Electronics basics... Some brands are more sensitive for that. There are 78xx that are quite stable without the cap but it's common practice to use one (if you don't like trouble that is).
Try that with a 79xx 9 negative version and it will not work for sure. The positive ones are more forgiving. The negative ones also need a minimum load. That is why a "power on LED" is best connected to the negative supply. I tell this just in general as it seems you missed some basics. I do not mean this negative, I just want to share some facts.
Just don't use low ESR caps in large values after the reg. So no Pana FC/FM etc. when larger than 100 uf. The caps you choose are the required ones but you could add a somewhat larger electrolytic at the output. Let's say 10 uf 16V. In that case you can still use a Pana FC/FM. Large low ESR electrolytic caps after the regs cause similar problems as when you don't use a cap at all.
Last edited:
No problem, I'd rather be told what I've done wrong as opposed to not being told and making the same mistake twice. I admit I have got away with using 7805 regs in the past with no smoothing caps, but just because it works doesn't mean it's the correct way of doing things.
Just check out a website called thereifixedit for proof of this! 
Great info, thanks. I'm not 100% what I have in my box or bits as I'm at work right now but I'm pretty sure I've got plenty of 10uF lytics along with a few 47uF black and gold caps which I think are FM's. Once I get home in a few hours I'll get soldering, I think I'll go for the 47uF's.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.