Jump to content

Recommended Posts

Posted
4 minutes ago, Pars said:

....... with no easily removed component to isolate the amp channels from them (short of cutting traces).

I gifted myself a hot air rework station for Christmas, so not a major issue. In fact I removed the negative regulator and resoldered it because I thought that was the problem along with the 10k feedback resistor.

There is in fact one removable component which is the that340 chips. Would there be any issue running the amp without those chips?

The top side was all hand soldered along with the through-hole components so I suspect the issue is the bottom because I was a little too generous with the solder paste. 

Posted

The THAT chips aren't between the regs and the power input for the amp sections. The input section of the amp obviously wouldn't work without something being in there.

From what I'm seeing in a sim, nothing bad would happen. The outputs won't be biased, etc., but only seeing low-level voltages where I looked.

Posted

I think I might have isolated the problem. The CCS LEDs appear to be mounted in reverse. For other LEDs the green stripe indicates the cathode but for LTST-C170CKT, it seems to go the other way according to the datasheet.

http://www.mouser.com/ds/2/239/liteon_LTST-C170CKT-1175291.pdf

With respect to this LED I followed the silk screen and it seems wrong. This is confirmed when I test the LED and put the positive probe on the negative side and it illuminates.

Posted

IT WORKS!!!!!

All the LEDs light up with power supply set to +18V/-18V. Right channel on the TRS is 70mV and left channel is 190mV. Temperature is a cool 140f without the heatsinks.

now we know what happens when you install CCS LEDs backwards.

  • Like 7
Posted

Installed a new negative regulator, adjusted the trimpots and soldered in the 0 ohm resistors. Finally had a listen. I have never heard the full size Dynalo SuSy so I cant compare but this little amp sounds amazing through my HD 6xx! Absolutely no noise except a tiny bit of scratchyness while turning the TKD pot. Voltage rails are +19.99/-20.03. Here are some temps I measured 

positive regulator heatsink- 50C

negative regulator heatsink - 60C

output transistors (no heatsink) - 75C

A82489C1-3962-49FC-B1CC-C414C60855CD.thumb.jpeg.c40459a22f8f2ae5fe5411b3b147be99.jpeg

 

Thank you Kerry, Spritzer, KG, Pars, mypasswordis, Sbelyo, John McLean, mwl168, RudeWolf, JoaMat and countless other people for making this possible.

Now that I have an amp I am turning my attention to casing. I'm willing to organize a group buy for enclosures but I need at least 10 to make it interesting. Ill make sure to have panels for both the TKD and Alpha pot. Anyone interested?

  • Like 9
Posted (edited)

Mini BOM part question:

For C4, C5, C6, C7, a tantalum polymer 100uf/25V cap is indicated. Not sure what is with these caps, but they seem rather expensive. All the usual suspects at Mouser run from $4-$6 each, with a couple running up to $16 ea. The ones in my BOM are all on backorder (one which was $2.91 in qty of 10 or more).

Then I run across Vishay-Polytech, which is $1.79 ea, and $1.27 in qty 10 or more. Something wrong with these compared to Kemet, AVX, etc.? Just looking for something reasonably priced and IN STOCK.

EDIT: wound up ordering the ones (Kemet T521X107M025ATE060) in Kerry's original BOM... figured he or Kevin chose them for a reason.

Edited by Pars
Posted

The reason Kemet T521X107M025ATE060 being selected might probably just be the low profile.

This cap is Al electrolytic on DIP dynalo, so I assume the ripple current rating is the major concern.

My choice is Panasonic 25SVF100M, OS-CON series.

Compare the specs:

[Name]                                 [DC Leakage]                  [ESR]                                 [Allowable ripple current]                    [DF]                              [Price]

Kemet T521                            250uA              60mA@25C 100KHZ               2000mA@45C 100KHz                 10%@25c 120Hz              $3.56 each for ten more

Panasonic 25SVF                  500uA             24mA@20C 100KHz                3200mA@<105C 100KHz             12%@20C 120Hz             $1.23 each for ten more

pros: Half more ripple current rating. Half in ESR. Almost 1/3 in price.

cons: Double in the DC leakage, but waste just 1mA  more for 4 cap. DF are both bad.  Barley fit the pad. 7mm in Height.

 

 

Posted (edited)
12 hours ago, astro said:

The reason Kemet T521X107M025ATE060 being selected might probably just be the low profile.

This cap is Al electrolytic on DIP dynalo, so I assume the ripple current rating is the major concern.

My choice is Panasonic 25SVF100M, OS-CON series.

Compare the specs:

[Name]                                 [DC Leakage]                  [ESR]                                 [Allowable ripple current]                    [DF]                              [Price]

Kemet T521                            250uA              60mA@25C 100KHZ               2000mA@45C 100KHz                 10%@25c 120Hz              $3.56 each for ten more

Panasonic 25SVF                  500uA             24mA@20C 100KHz                3200mA@<105C 100KHz             12%@20C 120Hz             $1.23 each for ten more

pros: Half more ripple current rating. Half in ESR. Almost 1/3 in price.

cons: Double in the DC leakage, but waste just 1mA  more for 4 cap. DF are both bad.  Barley fit the pad. 7mm in Height.

 

 

Where do you find said Panasonics? I can't find 100uf/25V anywhere. 20V maybe, or 68uf/25V?

EDIT: was searching tantalum. Found them; yes a bit tall, not sure how much of a problem that would be.

5 hours ago, Kerry said:

I needed a small package (ok, joke away) and tantalum usually is a good choice.

Digikey has them reasonably priced for 10 or more.

I don't find $3.80 ea in 10 or more reasonable. I've paid less for BlackGates :) Arrow wants $6.76 ea in 10 or more, and they have over 12,000 in stock. I wonder why?

Edited by Pars
Posted

Found on Mouser. Still provide the link: https://goo.gl/eKbkSC
The height is 4mm vs 6.9mm. Nearly double or just 3mm in difference!

One alternatives is slim radial two leaded Al Electrolytic cap and let it lay down on the board. 
If so, the choice would become many many a lot, but mind the diameter.
It will need more room  on the bottom side in the case which makes it look bulky.(say over 10mm)

Kemet 100uF/25V has almost the top capacitance in its own class(tantalum cap). This may explain the price.
One more thing, the endurance. 25SVF is 1000hr@105C.
Kemet T521 is 2000hr@105C and they also  claim there is no aging effect.

Kerry is right. In fact, when it comes to very limited space and relatively large capacitance to ceramic, tantalum is great.

Posted

Thanks, and yes I had found them on Mouser myself. Order was already placed at that point and I would have been concerned about the height.

The Vishay cap T55D107M025C0060 has very similar specs to the Kemet, is less than half the price ($1.27 vs $3.56), is the right footprint (2917/7343) and is short at 2.8mm. I should have ordered those instead. Not sure what the reputation of that Vishay division (Polytech) is though.

 

 

Sent from my iPhone using Tapatalk

 

 

Posted

Tried my hand at soldering a pair of the regs on, using a Hakko 936 with a 3.2mm chisel tip. Had a hard time getting some of the pads to tin on the neg reg, used plenty of flux. Switched to my smallest point tip and ran around the outsides of the chip (from the YouTube video). My hot air station hasn't arrived yet, nor have the Meanwells. I'll put the associated components on tomorrow night and probably use my bench supply and see what happens. Sorry for the crappy photo, but let me know what you think.

 

IMG_0847.jpg

  • Like 3
Posted (edited)

Thanks. Put T55D107M025C0060 in my pocket list.

No sure exactly "High Temp Load Time" on the datasheet means.
If it's endurance, then it just have half life span than Kemet.
 

By the way, do anyone have any idea  if there were any side effect swapping 20ohm output resistor with 10 ohm?

Edited by astro
ask one more question
Posted

As for cutting the output resistors in half, I guess it would double the current? Not sure if that is advisable though oftentimes more is better. I would think you would need to re-bias it however.

Group buy mini board question:
Noting the trouble I had getting some of the LDO pads to tin, is any special cleaning of these boards required? I only scrubbed the area with rubbing alcohol/toothbrush. Do I need to take some scotchbrite or something else to these boards before building?


Sent from my iPhone using Tapatalk

Posted

I used isopropyl alcohol. I had trouble soldering anything connected to the ground plane, especially since its a 3oz board. There was more then a few times that I needed to crank the temp to max to get anything to flow.

Posted

Yes, I understood what johnwmclean meant when he said soldering this board felt like it was sucking the life out of his soldering iron. I had no trouble with the pads under the regs, which are also connected to ground. Just the small pads.


Sent from my iPhone using Tapatalk

Posted

Ok stupid question time [emoji4]

I hadn’t seen this kind of symbol used on a silkscreen before for a polarized cap (the 100uf caps):

647937bfba1c8286039fd1e7fb53ffb2.jpg

 

Which end is +? I’m guessing the closed end of the symbol? I suppose I could figure out which rail is which and then I’d know for sure :eek:

 

 

 

Sent from my iPhone using Tapatalk

 

  • Thanks 1
Posted (edited)

The stripe on the Kemet is the + side and should match the solid line on the board. On my board all the capacitors have the stripe going towards the rear of the amp.

Edited by cspirou
Posted

Question regarding the mini heatsinks in the BOM, Aavid 581002B02500G. I didn't realize these were tapped for 6-32, rather than something more rational, like 4-40. The screw head for a 6-32 contacts the regulator chip, which doesn't seem like a good idea. The TO220 mounting kits and parts I have are all sized for 4-40, including the various insulators I have. I noticed in Kerry's build that he used an insulator that covers the screw head... can't tell for sure what it is exactly.

Additionally, Amb's o78/o79 are a bit taller than normal TO220s and with the Molex or 3M headers I used, sit a bit taller than the mounting hole. They also sit out from the heatsink at the bottom. In the image on Amb's site, it looks like he bent the pins slightly (but doesn't mention it).

sigma78_100_pcb_vert.png

 

I guess the screw issue is the largest one. Any thoughts/suggestions? Not enough room between the fins for a nut.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.