My experience with the US Postal service has been very poor. Anyone in Australia would be wise to choose the next level up of postage. The equivalent of registered.
Out of three packages I've had sent over the years from the US as standard airmail exactly ZERO have arrived. The last one came back to the sender some six months after being sent. Sending using the more expensive postal option (I think it added about $10) it arrived safely in a reasonable time.
Tony.
I'm in Canada and Canada Post is generally quite good, but not cheap. Where I worry is when the package leaves the country and is handed into someone else's care for final delivery. In any event, higher levels of service were offered at the buyer's expense. In my case all opted for standard air mail.
Jason,
Thank you for your explanations. I agree with you that we have a working PCB and it is already sounding good (from your report).
I will be building this version a bit later as I am busy completing other ones. But I will report any changes I notice after modifications. Thank you for shipping the boards and thanks to LC for this masterpiece.
Regards
Thank you for your explanations. I agree with you that we have a working PCB and it is already sounding good (from your report).
I will be building this version a bit later as I am busy completing other ones. But I will report any changes I notice after modifications. Thank you for shipping the boards and thanks to LC for this masterpiece.
Regards
Last edited:
Let's not derail the thread with a bunch of postal service banter; we know there's risk to sending anything that isn't registered or trackable. If something goes awry we find ways to deal with it individually on a case by case basis. In my humble opinion, we have said enough about that, including myself.
Let us hang tight and hopefully we see the first of my packages land at their destinations late this week or sometime next week and get onto some building!
Let us hang tight and hopefully we see the first of my packages land at their destinations late this week or sometime next week and get onto some building!
No, it never really was about a GB.
For me it was simply finding loving homes for my excess children
. If I end up re-ordering (and I imagine I will - demand was higher than I expected) I will move such an activity to a new thread in the appropriate area. I managed to obfuscate this one much more than I thought I would for just a handful of PCBs. To PMI, and everyone else, I apologize for that.That said, this thread is about building the VSSA and we have PCBs on their way around the world. Hopefully they all find their mark without delay and we get to hear about some happy builders before too long
.
What would the technical reasoning be? The 2Q CCS should be better than 1Q+LED even though that arrangement in itself isn't bad either.
As a thought, it would not be difficult to sub in an LED in place of one of the transistors and adjust the resistance to get the current at the correct level should one desire it.
Last edited:
Packages arriving...
Just a reminder that it would be really nice to get feedback when your packages arrive. So far:
still4given - 2 - Delivered
ivanlukic - 2 - Shipped
quan - 2 - Shipped
triplej - 2 - Shipped
baswamin - 2 - Shipped
potepuh - 2 - Shipped
spind - 2 - Delivered
brianco - 2 - Shipped
Sheldon - 2 - Shipped
Just a reminder that it would be really nice to get feedback when your packages arrive. So far:
still4given - 2 - Delivered
ivanlukic - 2 - Shipped
quan - 2 - Shipped
triplej - 2 - Shipped
baswamin - 2 - Shipped
potepuh - 2 - Shipped
spind - 2 - Delivered
brianco - 2 - Shipped
Sheldon - 2 - Shipped
Module set-up
Hi guys,
For those who have received or will be receiving my boards here's some information on calibration / setup. Please note that it is possible set the adjusters in such a way as to cause problems (as in minimum resistance for any one, combination of two, or all three); this layout was not intended to be fool-proof. Perhaps a future revision may be more forgiving in this regard. Be certain of your starting points with the adjustments and there should be no issues. It is helpful to have at least two multimeters available if possible, one to monitor VAS current and another to monitor DC offset.
The two trimmers for the CCSs VR1 and VR3 will have two effects, first is to control VAS current and second to null out DC offset. They are interactive, you will be going between the adjusters to bring the modules into adjustment. The trimmer VR2 is just for setting quiescent bias in the OPS.
1) Pre-set the CCS trimmers VR1 and VR3 to 270Ω as a starting point.
2) Set the bias trimmer VR2 to maximum resistance so as to yield minimum bias current.
3) Insert an ammeter into the +V supply to measure total module current draw. As an alternative you can measure voltage drop across resistors placed in both of the +/-V supply lines. I'd suggest perhaps 10Ω in each supply line and measure across one of them with a volt meter.
4) Power up with a limited supply that outputs your target supply voltage and verify there is no fault; some DC offset is likely to be present since we haven't yet adjusted anything but it should be no more than a few tenths of a volt.
5) Turn the bias trimmer to set up some bias current though you won't need full bias just yet. Perhaps 30mA to 50mA total module draw to start. Having some bias seems to make DC offset adjustment easier.
6) Measure across either R13 or R14 (10Ω in the VAS emitter circuit) to determine VAS current, every 10mV will represent 1mA. The target range will be anywhere from 12mA to 16mA, so we want to achieve a voltage reading from 120mV to 160mV.
7) Monitor DC offset at the output.
8) Adjust VR1 and VR3 such that you achieve VAS current in the desired range and the lowest possible DC offset at the same time. Clockwise should reduce current, counter-clockwise should increase it.
9) Place module on a normal power supply. Adjust bias current to the desired level with VR2, I suggest a total module current anywhere from 120mA to 160mA would be appropriate. Recheck VAS current and DC offset, adjust if necessary.
10) Allow module to reach equilibrium and as a final check ensure DC offset , VAS current and total module current are within tolerances.
Then enjoy the music. Hopefully this helps give builders a trouble-free set-up.
Hi guys,
For those who have received or will be receiving my boards here's some information on calibration / setup. Please note that it is possible set the adjusters in such a way as to cause problems (as in minimum resistance for any one, combination of two, or all three); this layout was not intended to be fool-proof. Perhaps a future revision may be more forgiving in this regard. Be certain of your starting points with the adjustments and there should be no issues. It is helpful to have at least two multimeters available if possible, one to monitor VAS current and another to monitor DC offset.
The two trimmers for the CCSs VR1 and VR3 will have two effects, first is to control VAS current and second to null out DC offset. They are interactive, you will be going between the adjusters to bring the modules into adjustment. The trimmer VR2 is just for setting quiescent bias in the OPS.
1) Pre-set the CCS trimmers VR1 and VR3 to 270Ω as a starting point.
2) Set the bias trimmer VR2 to maximum resistance so as to yield minimum bias current.
3) Insert an ammeter into the +V supply to measure total module current draw. As an alternative you can measure voltage drop across resistors placed in both of the +/-V supply lines. I'd suggest perhaps 10Ω in each supply line and measure across one of them with a volt meter.
4) Power up with a limited supply that outputs your target supply voltage and verify there is no fault; some DC offset is likely to be present since we haven't yet adjusted anything but it should be no more than a few tenths of a volt.
5) Turn the bias trimmer to set up some bias current though you won't need full bias just yet. Perhaps 30mA to 50mA total module draw to start. Having some bias seems to make DC offset adjustment easier.
6) Measure across either R13 or R14 (10Ω in the VAS emitter circuit) to determine VAS current, every 10mV will represent 1mA. The target range will be anywhere from 12mA to 16mA, so we want to achieve a voltage reading from 120mV to 160mV.
7) Monitor DC offset at the output.
8) Adjust VR1 and VR3 such that you achieve VAS current in the desired range and the lowest possible DC offset at the same time. Clockwise should reduce current, counter-clockwise should increase it.
9) Place module on a normal power supply. Adjust bias current to the desired level with VR2, I suggest a total module current anywhere from 120mA to 160mA would be appropriate. Recheck VAS current and DC offset, adjust if necessary.
10) Allow module to reach equilibrium and as a final check ensure DC offset , VAS current and total module current are within tolerances.
Then enjoy the music. Hopefully this helps give builders a trouble-free set-up.
Any chance of more info? One possibility is thermal stability. As you mention simulations I'm excluding sonics. Also over on the CFA / VFA thread there is suggestions that a CCS and a mirror is better for PSRR and thermal drift. All things that can be experimented with on this board IMHO.
As a fact seems LC use for his high power VSSA (FET+LED). First picture (http://www.diyaudio.com/forums/vend...-one-mosfet-amplifier-module.html#post3772084).
I think two questions arise about the ccs:
1. If we assume a really good PSU, does VSSA really need any kind of ccs to the optimal sonic performance? Shaan's answer is clear no. He used ccs in an experimental version of PeeCeeBee (not mentioned what type used) and the result described as "something strange" or "clinical" His opinion is that vssa sings perfect without ccs, and the music appears to be free of any limitation. Lazy Cat has totally opposite experience: the difference is like "day and night" benefit of ccs.
2. If yes it is not so clear to me, what is the correlation between the applied ccs and the sonic quality. Theoretically 2 Q's ccs is better than any other in respect of thermal stability, accuracy of FB, lower THD, etc.(see Rod Elliott's article e.g.) and very close the ideal constant current source though some builders found that bjt+led work better with this topology as ivanlukic mentioned above. How these collerate to the listening impressions. Has anyone experience with other type ccs, like TL431?
Thanks
egra
1. If we assume a really good PSU, does VSSA really need any kind of ccs to the optimal sonic performance? Shaan's answer is clear no. He used ccs in an experimental version of PeeCeeBee (not mentioned what type used) and the result described as "something strange" or "clinical" His opinion is that vssa sings perfect without ccs, and the music appears to be free of any limitation. Lazy Cat has totally opposite experience: the difference is like "day and night" benefit of ccs.
2. If yes it is not so clear to me, what is the correlation between the applied ccs and the sonic quality. Theoretically 2 Q's ccs is better than any other in respect of thermal stability, accuracy of FB, lower THD, etc.(see Rod Elliott's article e.g.) and very close the ideal constant current source though some builders found that bjt+led work better with this topology as ivanlukic mentioned above. How these collerate to the listening impressions. Has anyone experience with other type ccs, like TL431?
Thanks
egra
Last edited:
I had done the basic 'Shaan' style PeeCeeBee and liked it very much. I was looking to explore different avenues with the VSSA 'core' and this represents a start towards my evolution of the concept.
As far as the CCS goes, the simulations on my end don't seem to show much, if any, difference between a 2Q CCS and 1Q+LED, but I'm no expert either.
I'm also following ostripper's work as well and see some good things going on there that might be worth looking at incorporating into a future revision.
As far as the CCS goes, the simulations on my end don't seem to show much, if any, difference between a 2Q CCS and 1Q+LED, but I'm no expert either.
I'm also following ostripper's work as well and see some good things going on there that might be worth looking at incorporating into a future revision.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.