The Multi Amp aka Dynalo Mk2

[GrindingThud]

I'm trying to source a 3RP/2142GSA1A10K pot for the build.  Any good sources online for this?

[Kerry]

Looks about right.

[luvdunhill]

I determined that the JFET combos were mismatched, and got that corrected

I thought my numbering scheme was infallible! Well, in my head at least it is.

I bought a bunch of LEDs to test the voltages.  I liked the LTST-C170CKT since it had a high Vf.  Changing the LED will have a big effect on the output current if it's not matched to the resistor. 

After all these years I just don't understand why people don't build LED-based CCSes on a breadboard, use a pot to dial in the current and buy the resistor to match the LED. If I had to count the posts of people looking for specific LEDs...... /rant off (not really, still on. Don't be that guy!)

[Kerry]

I've been using the diode tester on my multi-meter (BK 2709B) to check the forward voltage on the LED and transistor.  It get's you fairly close to actual currents:

current = ( Diode(Vf) - Transistor(be) ) / resistance

In fact, when I'm matching transistors for simple current sources I pay more attention to the voltage across the transistor than the hfe (though I match both anyway ).

 

[Pars]

I thought my numbering scheme was infallible!

These weren't the devices that I got from you, but the ones that I had in there. Yours are matched very well. I just haven't brought myself to use them yet



Sent from my iPhone using Tapatalk

[mtoc]

@Pars Hi Pars, as for the bridges on GR LV, what kinds of bridges are needed? Do some of them measure better? (on the 6.5 bits stuff or even 8.5 bits)

[Pars]

The ones in the BOM are what is needed. I use Schottkys because I believe that they induce less switching noise. The GRLV is very well protected against diode switching noise, so I don't know whether they actually help in this case. I couldn't see anything on my scope when looking at them, but I might not know what to look for and I certainly don't have access to a screen room.
I don't know what you are talking about in your last sentence.


Sent from my iPhone using Tapatalk

[luvdunhill]

He wants your digits man.

[Pars]

I'm not sure how to feel about that


Sent from my iPhone using Tapatalk

[mtoc]

oh, sorry. should be digits...

where to find to the BOM? I've only got the gerb file...

Edited December 27, 2016 by mtoc

[Pars]

What do you want the BOM for? The Dynalo Mk2 (this thread) or the GRLV (a different thread)? The respective BOMs are posted in their thread...


Sent from my iPhone using Tapatalk

[mtoc]

8 hours ago, Pars said:

What do you want the BOM for? The Dynalo Mk2 (this thread) or the GRLV (a different thread)? The respective BOMs are posted in their thread...

The BOM of GRLV. You've mentioned its BOM (the Excel BOM) in the GRLV thread for a few times, but I still haven't find it.

I got the gerb file (zip) and the schematics (pdf), unfortunately both of them haven't have the standard/stock bridge model that we should look into.

[Pars]

I thought it was in there. See attached.

GR LV BOM_Sep26.xls

[mtoc]

46 minutes ago, Pars said:

I thought it was in there. See attached.

GR LV BOM_Sep26.xls

I have two of these in my drawer, is that a bit crazy? RBV5006, max RMS voltage 600V, max average forward current 50A (but these are old, datasheet says production begins in 1998). The bridge in the BOM is 100V/4A

http://www.allcomponents.ru/synsemi/rbv5006.htm
http://www.datasheetcatalog.com/datasheets_pdf/R/B/V/5/RBV5006.shtml

[Pars]

^ this belongs in the golden reference thread, not here.

Question: do individual devices run hotter at specific bias than others at the same bias/current?
All of my devices (MPSW06/56) are matched by hFE to the best of my ability. Vbe was consistent as well as Ib when matching (monitoring Ib and Ic). Since getting an IR thermometer and looking at temps, I have a couple of devices running at 55C (Q13 on both boards, interesting). Both are running at similar current to neighboring devices and the R value is within 0.02 ohms of others. hFE is also very close. I thought by measuring all of the R values and drop across them that I would see something out of the ordinary for the devices running warmer, but I'm not seeing anything.
I'll probably just drop the bias resistors down to 255R for now and revisit when I case the amp to ensure that temps aren't over 50C or so. Any advice on when temp would be a problem appreciated as well. Current is verging on 15mA; these two are running at 14.8 or so (don't have access to data right now). These are thru hole boards also.


Sent from my iPhone using Tapatalk

[luvdunhill]

Something nearby Q13 is generating heat? If the voltage drop across the resistors is the same, they should share the same current.

[Pars]

Nothing that would really generate any heat around that, and I was checking temps around the board anyhow. There are 3-4 others at 50C or more, some in the 30s. This was all done at approximately the same time with the amp having run for an hour or so.

 

Board 1
NPN	device ID	hFE	temp °C	Vr	Ie		ohms		PNP	device ID	hFE	temp °C	Vr	Ie		ohms
Q22	10	293.3	39	0.2926	14.64	R44	19.991		Q24	45	293.3	52	0.2916	14.59	R47	19.989
Q21	32	293.1	39	0.2996	15.00	R45	19.978		Q23	18	293.1	46	0.3004	15.03	R46	19.985
Q16	44	93.0	45	0.2998	14.99	R40	19.999		Q17	64	293.0	44	0.2995	14.98	R43	19.992
Q12	8	292.8	40	0.2936	14.69	R32	19.982		Q14	30	292.7	42	0.2949	14.75	R37	19.996
NPN	device ID	hFE	temp °C	Vr	Ie		ohms		PNP	device ID	hFE	temp °C	Vr	Ie		ohms
Q11	74	278.9	35	0.2933	14.68	R35	19.983		Q13	11	278.8	55	0.2897	14.50	R36	19.980
Q15	40	279.1	35	0.2929	14.67	R41	19.970		Q18	31	279.4	47	0.2989	14.96	R42	19.982
Q19	5	279.4	45	0.2896	14.50	R51	19.973		Q25	39	279.5	49	0.2944	14.74	R48	19.969
Q20	86	279.6	47	0.2865	14.34	R50	19.982		Q26	90	279.6	38	0.2893	14.49	R49	19.968
bias			VR39	1.7420								VR39	1.7420
Board 2
NPN	device ID	hFE	temp °C	Vr	Ie		ohms		PNP	device ID	hFE	temp °C	Vr	Ie		ohms
Q22	20	298.6	31	0.2879	14.42	R44	19.965		Q24	93	298.6	50	0.2813	14.08	R47	19.983
Q21	53	298.4	36	0.2883	14.43	R45	19.981		Q23	60	298.4	50	0.2807	14.06	R46	19.969
Q16	55	298.1	42	0.2875	14.39	R40	19.979		Q17	25	298.4	44	0.2905	14.54	R43	19.983
Q12	62	297.2	45	0.2839	14.21	R32	19.973		Q14	76	297.2	42	0.2867	14.36	R37	19.963
NPN	device ID	hFE	temp °C	Vr	Ie		ohms		PNP	device ID	hFE	temp °C	Vr	Ie		ohms
Q11	88	282.2	30	0.2834	14.19	R35	19.968		Q13	55	282.3	55	0.2819	14.12	R36	19.963
Q15	7	282.6	36	0.2869	14.36	R41	19.975		Q18	99	282.5	48	0.2862	14.32	R42	19.980
Q19	85	282.8	43	0.2862	14.31	R51	20.004		Q25	56	282.6	40	0.2860	14.33	R48	19.965
Q20	64	283.3	46	0.2814	14.09	R50	19.975		Q26	49	283.3	37	0.2831	14.16	R49	19.986
bias			VR38	1.7344								VR38	1.7380

 

Edited December 28, 2016 by Pars

[Pars]

I know, tl;dr

Question on using an IR thermometer on these.
When I aim at the to92 flats, since these are in rows of 4, it seems like the far end device measures hotter than the others, and temp increases as I go down the row. Shooting them from the top results in more even but somewhat lower temps. I think the thermo is picking up more than the one device when going for the flats down the row.
Thoughts?


Sent from my iPhone using Tapatalk

[MLA]

On a tangent; the HD800 sound brilliant out of the Dynalo, but I find myself boosting the bass a couple of dB most of the time for it to sound right to my ears.

So far I've used digital EQ, but I wonder if someone has a recommendation for how to incorporate a small add-on circuit/daughterboard that could fill the same function inside an existing Dynalo build? I know there is a ton of circuit suggestions out there, but I'm not good enough at this to figure out what would work well with the Dynalo. 

[mypasswordis]

Do the super dupont mod or similar on your hd800, then there's no eq necessary 

[MLA]

I did, and that works perfectly when I've got the place to myself and can turn up the volume.

However, my social setting requires low volume listening more often than not. At those levels, I need a small bass boost to preserve balance.

It's just for convenience; with a switch on the amp I would not have to walk over to the computer to change settings . 

 

[Pars]

You might be able to implement something like this: http://www.diyaudio.com/forums/headphone-systems/199530-o2-amp-crc-diode-cap-heatsink-mods-6.html

Pay no attention to it being for the O2 amp

You would need a 4P switch to pull this off (simple bass boost on/off).

[MLA]

Many thanks, that looks exactly like it!

Did some more reading and that O2 mod seems quite similar to how the M3 bass boost is implemented. 

I have a spare Dynalo in the basement to experiment with; I will give this approach a try and see what comes out of it.

[rumina]

Not a Mini Dynalo but a small one. Finished my Dynalo today, thanks to this forum and the members the build was easy and free of trouble, thanks everyone. Was fun to put it in a small case and solve this puzzle. Also a big thanks to Kevin Gilmore for this design and sharing it with us.

http://www.diyaudio.com/forums/gallery/showgallery.php?cat=2232

Thanks to sbelyo for the group buy of the pcbs.

 

[Pars]

2 hours ago, rumina said:

Not a Mini Dynalo but a small one. Finished my Dynalo today, thanks to this forum and the members the build was easy and free of trouble, thanks everyone. Was fun to put it in a small case and solve this puzzle. Also a big thanks to Kevin Gilmore for this design and sharing it with us.

http://www.diyaudio.com/forums/gallery/showgallery.php/cat/2231

Thanks to sbelyo for the group buy of the pcbs.

 

Looks good.

It appears that you populated all 3 bridge rectifiers in the Power Supply? You should remove the center one.  It is there for people who want to use a center-tapped transformer. Since you are using a dual secondary transformer, only the 2 outer bridges are used (BR1 and BR2 in the schematic). With a center-tapped transformer (say 20-0-20), only the center bridge (BR3 in schematic) is used.