Yes you're true, I have read that apart from giving a higher output swing (because of being bridged), it too parallel mosfets devices, increasing output current. That's bridge parallel.
Two bridged F5 just provide more voltage, not more current.
apart from that, joining their grounds in that special manner make it able to reduce distortion further.
The F5 "softstarts" as it is, I made mine with the stock PSU from Nelson's PDF and it starts up and shuts down WITHOUT ANY noise or pop-up, that's one of the best features of this amp in my opinion.
Hi all,
I have just finished a Pass F5 with Toshiba output transistors.
I know I am stupid, but I just turned on the power, without this bulb thing. Some friends will join me this evening for a HiFi meeting, so I were in a hurry, which is a bad thing.
I have borrowed another amp for this evening, but I have used a lot of time building this one, and I am very exited to have it play.
I think I had the pots maxed in the wrong direction, and R11 and 12 looks burned.
The power supply is giving out +/- 25,5 and it seems that it is not drawing any power, because nothing gets hot anymore.
Can anybody estimate what needs to be exchanged ?
Thanks in advance
Arthur.
I have just finished a Pass F5 with Toshiba output transistors.
I know I am stupid, but I just turned on the power, without this bulb thing. Some friends will join me this evening for a HiFi meeting, so I were in a hurry, which is a bad thing.
I have borrowed another amp for this evening, but I have used a lot of time building this one, and I am very exited to have it play.
I think I had the pots maxed in the wrong direction, and R11 and 12 looks burned.
The power supply is giving out +/- 25,5 and it seems that it is not drawing any power, because nothing gets hot anymore.
Can anybody estimate what needs to be exchanged ?
Thanks in advance
Arthur.
Russel...... Imagine what it's like to do a major revision to a large spacetrack radar, like we do....... that's 32 Megawatts of power. Our back-up generators are powered by two jet engines used on Boeing 727's..... I'm not so certain the old "work with one hand in your pocket" is adequate, around these power supplies....!
Output transistors would have blown...I will start there with light bulb in place and pots replaced and in the right direction.
Cheers.
Arthur....... sounds as though your power supply is still good. Your voltage levels suggest that......
Something I'd recommend, as you troubleshoot (and I'm surprised more folks don't do this.....). Build/rebuild only ONE amp PCB (double and triple check it, as you go), and then connect it to the power supply. Use the bulb tester! If you do this, you only place ONE amp PCB "at risk".
Once the first channel is working, you can compare component values, rough bias settings, etc, between the first and second PCBs..... Reduces the risk of problems in the second PCB when you fire it up.
Now connect your second amp PCB to the power supply (leaving the first one connected is OK). Then test and set bias for the second amp PCB.
This way, if you have a common problem in your boards (miswiring, wrong components, wrong bias), you'll only be sacrificing ONE PCB, and not both!
(Trust me. I have learned this the hard way, over the last 50 years....!)
Ken
It's advisable to take off your wristwatch and wedding band as well...
And possibly remove your fillings!
I once had a Mcintosh MR 67 putting 120 volts AC on the chassis. It was transferred via the interconnects ground cable to the ground on the RCA of the preamp, and to its chassis. While reaching behind the pre to where the cable attached to the RCA, with one hand, my knee came in contact with the seperate case of the tube amps power supply. Fortunately I only felt a tickling sensation as my knee started to contact the chassis....sometimes this can really sneak up on you!
Russellc
Speaking of safety -- just this past year there was an undergrad at Yale who got her hair tangled in a lathe and she died of her injuries. I worked in a die-casting plant in the summer back in the 1960's and NONE of the punch press operators had all their fingers!
Getting back to topic -- you can use other JFETs than the K170/J74 -- you may have to parallel P-Channel JFETs to get the gm you need on the gain stage and current you need to set the bias). It's a matter of necessity filling the void. You may have to add a little resistance in series with the trimpots to get the bias correct, but this is DIY after all. (You can, as well, build a "Starving Student" headphone preamp with tubes other than that specified!)
Getting back to topic -- you can use other JFETs than the K170/J74 -- you may have to parallel P-Channel JFETs to get the gm you need on the gain stage and current you need to set the bias). It's a matter of necessity filling the void. You may have to add a little resistance in series with the trimpots to get the bias correct, but this is DIY after all. (You can, as well, build a "Starving Student" headphone preamp with tubes other than that specified!)
F5 power buffer only?
Hi,
would it be possible to reduce gain of F5 to say 5db or better to unity-gain? If so, lowering bias current to say 150mA and I would have quite good class AB power buffer with NFB, right?
As I am looking for class AB buffer with some NFB or error correction. It is intended to be driven by tube preamp, but whole amp will not have global negative feedback loop.
Thanks.
Hi,
would it be possible to reduce gain of F5 to say 5db or better to unity-gain? If so, lowering bias current to say 150mA and I would have quite good class AB power buffer with NFB, right?
As I am looking for class AB buffer with some NFB or error correction. It is intended to be driven by tube preamp, but whole amp will not have global negative feedback loop.
Thanks.
Very true. Here is a distortion vs bias curve for the F4 type circuit. In
http://www.passlabs.com/pdfs/articles/leaving_class_a.pdf
You can see some distortion spectra and waveforms as well.
Attachments
So I want to have lower bias because of lower heat - thus class AB operation. I do not need more power (I am fine with 30W-50W into 8 ohms), but I need to lower the distortion in class AB mode(bias). How can I do that if I do not want global feedback from output back to preamp? Thats why I was thinking adding additional gainstage between preamp and buffer and making close loop of NFB within buffer and added middle gain stage - F5. Can it be done to lower closed loop gain of F5 - thus increasing NFB to say 5db or better to unity-gain without stability problems and lowering the overall distortion in lower bias mode?
I am not an electrical engineer, just had this idea.
Thanks.
I am not an electrical engineer, just had this idea.
Thanks.
Last edited:
Very true. Here is a distortion vs bias curve for the F4 type circuit. In
http://www.passlabs.com/pdfs/articles/leaving_class_a.pdf
You can see some distortion spectra and waveforms as well.
Interesting, if man do not need more power, than cca 1.3A bias seems to be best compromise with F4.
F5 Monoblocks.
This is my first post so here goes. I have just completed these amp's and they have been running on and off for 5 days. They definately give my 845 SET amp's a run for their money. The bass drum on track 4 of Doug Macleod "Come to Find" startled me with its realism. The drum has the sound of the drum skin and a real "whack", not just a soft thump. Guitar is utterly believable with string sounds that could cut paper. On the EMI Classics recording of Scheherazade the dynamic range and clarity almost allows you to imagine you are there at the recording session.
Build: JIM's Audio F5 and power supply PCB's, Conrad heatsinks, Panasonic caps, 160VA Jaycar transformers. Power rails bypassed with 0.25uf polystyrene caps. Front handle from Bunnings. Hard wired mains lead with 10A cable and mains plug.
Equipment: Osborn speakers, Rega Saturn CD, passive ladder attenuator, DIY interconnect and speaker leads.
Blown away by this amp. SET's on shaky ground.
This is my first post so here goes. I have just completed these amp's and they have been running on and off for 5 days. They definately give my 845 SET amp's a run for their money. The bass drum on track 4 of Doug Macleod "Come to Find" startled me with its realism. The drum has the sound of the drum skin and a real "whack", not just a soft thump. Guitar is utterly believable with string sounds that could cut paper. On the EMI Classics recording of Scheherazade the dynamic range and clarity almost allows you to imagine you are there at the recording session.
Build: JIM's Audio F5 and power supply PCB's, Conrad heatsinks, Panasonic caps, 160VA Jaycar transformers. Power rails bypassed with 0.25uf polystyrene caps. Front handle from Bunnings. Hard wired mains lead with 10A cable and mains plug.
Equipment: Osborn speakers, Rega Saturn CD, passive ladder attenuator, DIY interconnect and speaker leads.
Blown away by this amp. SET's on shaky ground.
Attachments
& for fun 
> would it be possible to reduce gain of F5 to say 5db or better to unity-gain?
I had this idea around July last year, and discussed with Nelson.
The worry was (in)stability.
We even did some simulations which shows stability should be OK.
But one has to build to find out.
And since my original protos are with MarkLai, I cannot try it on them any more.
Here is my schematics (unproven).
Patrick
I had this idea around July last year, and discussed with Nelson.
The worry was (in)stability.
We even did some simulations which shows stability should be OK.
But one has to build to find out.
And since my original protos are with MarkLai, I cannot try it on them any more.
Here is my schematics (unproven).
Patrick
