LME49811 + Sanken STD03N/P

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Dear Andrew,

We did measure the system with an audio precision ATS1, and there is no audible crossover distortion and THD and IMD are all below 0,008%

The bias for the STD03 N/P's is advised lower then average bij the Sanken application sheet

Official Sanken advice 40mA through both Re's.

So let's assume we stick to Sanken's datasheet and we use Re's of 0,22ohm and we want a current of 20mA through each Re we still have a voltage of 4,4mV across each Re (8,8mV total)
which is still lower then average. of course we can make it higher but then the Thermal tracking will work less precise.

In our first listening samples we sticked exact to the Sanken datasheet, means 40mA across both Re's

In that case for Re 0,1 we need 4mV. total across both Re's and for Re 0,22 we need 8,8mV


Just listening wise the 0,1 Re version with 4mV. sounds better then the 0,22 Re Version with 8,8mV. Both version where thermal stable.

For those who are interested, I attach the application sheet from sanken.

With best regards,
Bas
 

Attachments

  • sap16 app.pdf
    63.8 KB · Views: 677
Dear,

Here a image of our prototype.

It is a 3 channel amplifier, with variable gain preamp section. 2x LME49810 + STD03 for mid and high, and BPA300 LM3886 for the subwoofer section.

In the final model I we consider to make the subwoofer amp also semi discrete with a bridged/parallel LME49810/STD03 combination.

Like Andrew point out many times on other sections on the Forum the heat dissipation of the LM3886 is a limitation, hence the fan on top of the heatsink :D

With best regards,
Bas
 

Attachments

  • img_01292.jpg
    img_01292.jpg
    54.2 KB · Views: 1,307
Dear Jack,

I have two pairs of SAP15 and some free samples of LME49811 in my drawer.. And I was planning to make a good amplifier with these devices for my living room...
So, do you think I can adaptate your inverted design to my SAPs? And have the same results as yours?
What about the input buffer? Do you use a preamp or buffer stage for this?
 
Dxvideo said:
Dear Jack,

I have two pairs of SAP15 and some free samples of LME49811 in my drawer.. And I was planning to make a good amplifier with these devices for my living room...
So, do you think I can adaptate your inverted design to my SAPs? And have the same results as yours?
What about the input buffer? Do you use a preamp or buffer stage for this?

I am trying to get it to remain thermally stable at a higher bias and much higher rail voltage, and would say that you're OK with a supply around 30V, but it gets more difficult with a 45V supply which is what I had planned to install them in.

FWIW, when I run the simple test circuit on the PDF, the THD% is down to the limits of my equipment (an AP SYS2022).

You might want to look at PansonHK's boards which are modular -- if you don't like the Sankens you can switch to TIP142/7's.
 
Dont you like its crossover distortion performance since youre looking for higher bias levels? In fact I was also not sure about to make this amp with SAPs, discrete darlingtons or lateral MOSFETs with gate drivers.. However without SAPs (or STDs in your application) Vbe multiplier will be needed.. Means a bit more complex layout.. In the other hand, SAPs (or STDs) provide a better thermal tracking as I heard..
Anyway, I planning to use this amp with 30+30 AC, not more than 42v DC per rail... So I can get 75W into 8R (and may be 150 per 4R) I think my SAP s will be enough for this power level (I will not use the on board resistors may be 0.1R external instead like yours)
I have tried TIP142/147 pair with my LM4702 amp (originally working with 2SK1058/2SJ162 pair) However MOSFETs sounded me (and still sounding) more musical.. TIPs sounded more bright I guess... But may be this situation is valid only with TIP series.. And may be SAP gives more natural sound..
By the way, have you compared inverted and non-inverted design by the meaning of sound quality?
 
In my prototype, I'm running about 7 mA through the diode stack and 75 mA through the output devices. I'm not seeing any bias stability issues. The supply voltage is +/-45 V.

This is with a tiny heat sink and my target supply voltage is +/-60 V. But still... Looks darn good.

I'm currently using 1.5 kOhm between the SINK and SOURCE pins.

I've measured the THD at output powers from 100 uW to 1 W as this is where the amp is likely to operate the majority of the time for casual listening. I don't see any harmonics peek up above the noise floor (85 dB dynamic range on the test gear) until I hit 1 W output. The THD graph posted a few posts back looks like my results.

For those interested, I have more details in this thread.

~Tom
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.