Low Forward Drop Diode for Baker Clamp - Suggestions?

[PB2]

Any favorite diodes for this application, say 100 mA, 100 V?
1N4148 does not sim well since I need .4-.5V forward drop rather than .6 to .8V.

Are there reliable Germanium diodes available besides old stock?

 

[Mooly]

Some of the fast recovery types in the LT library have lowish simulated forward volt drops.

Try RFUH5TF6S and RFNL10TJ6S and others like them. At 100ma they are around 630mv compared to 850 for a 1N4148 type.

 

[indianajo]

all the silicon diodes I salvage from PCAT switcher supplies and the like have about .450 v voltage drop. 1n4148 are still .7 v. Low loss is great for switcher supplies. I don't know how you buy one. 6A6 from diodes inc is reading .423 v on the DVM
Low current 40 v schottky diodes BAT48 read .195 on the meter but 100 v HTE5L100 (multicomp) reads .175 v. Perhaps 2 of the latter in series for .350 v?
100 ma voltage drops will obviously be higher.

 

[Mark Tillotson]

If you want a low forward voltage pick a high current diode. forward voltage depends on current density, high current diodes have a lower current density for a fixed current than signal diodes.


Also the 4148 can't handle 100V reliably.


Schottky diodes and high voltage aren't a great combination, the reverse leakage rockets with temperature and reverse thermal runaway is a distinct possibility unless its heatsinked.


As for germanium, even worse.

 

[Vovk Z]

I'm not shure about which diodes type is best for Baker clamp, but there is enough amount of 100V Shottky diodes, SR1100 for example.
SS120, S120 - 200V shottky.
BAV103, STPS1150, SS115.

 

[PB2]

Low capacitance is required which rules out most power diodes.

Interesting that 1N277 Germanium looks like a good fit:
https://www.jameco.com/Jameco/Products/ProdDS/35941CSC.pdf

Also interesting that 1N914B is rated for lower forward drop:
https://www.mouser.com/datasheet/2/258/1N914AB(DO-35)-1626016.pdf

1N914B would have to be selected for above average breakdown voltage.

 

[PB2]

Looks like BAT46 might work.
100V, 150mA, 6pF capacitance.

 

[Vovk Z]

1N914 is almost the same as 1N4148.

 

[Vovk Z]

If power Shottky are not good, then there is "Small signal shottky diode" type. For example "BATxx" and "BATxxx" series (large variety for different situations).

 

[PB2]

I'm not shure about which diodes type is best for Baker clamp, but there is enough amount of 100V Shottky diodes, SR1100 for example.
SS120, S120 - 200V shottky.
BAV103, STPS1150, SS115.

It needs to be a small signal diode for low capacitance, the SS120 family is 50pF and
much higher.
BAV103 is 6 pF and might be a good choice.

 

[wrenchone]

Schottky diodes may be problematic for Baker clamp duty because of higher leakage - same is true for germanium diodes. Larger silicon diodes, though they will have lower drop, will also have higher capacitance, so you'd be adding extra nonlinear Miller capacitance to your VAS stage. BAV19 or BAV21 might be suitable.

 

[PB2]

Found suggestions for:
BAS21J (50ns) (<2pF) This is a Silicon switching diode 300V, 250 mA, but does not have a low forward voltage.
FDH300 6pF 125V, 200mA, this is also a Silicon switching diode and also does not have a low forward voltage.

in this thread: Audio Power Amplifier by Douglas Self

 

[PB2]

Article on the subject: hifisonix.com: Audio Amplifier Design anti-saturation-diodes

 

[PB2]

Bonsai states that he likes a current limiting resistor in the collector lead of the
beta multiplier in front of the VAS to avoid saturation which I first suggested in
this post from 2007:
https://www.diyaudio.com/forums/sol...chwork-product-simulation-10.html#post1312476

 

[wrenchone]

If you are Baker clamping to begin with you are destroying the saturation capability of the VAS transistor, guaranteeing a high voltage drop. Schottky or germanium diode clamp diodes risk spuriously turning on the VAS transistor simply due to their leakage component at elevated temperatures. The only place where one routinely got away with using Schottky Baker clamps was in the S and LS series of TTL logic, limited to a Vcc of 5V.

My suggestion is to use the low capacitance silicon diodes and eat the losses, elevating the supply voltage somewhat if necessary to make up for the loss in voltage drive.