Hello everyone,
I own a KSA 150 with a very noisy transformer and It has to be changed. The price
Ouch! I can easily find out the voltages of the windings, since it is still functioning, however I would need the current drawn from each winding to find a suitable replacement. Anyone out there with more info?
Thank you.
I own a KSA 150 with a very noisy transformer and It has to be changed. The price
Ouch! I can easily find out the voltages of the windings, since it is still functioning, however I would need the current drawn from each winding to find a suitable replacement. Anyone out there with more info?Thank you.
The Krell KSA amps of that vintage that I know of are either Class A or very high bias designs that run hot. That means the power transformer has a more difficult task. I know the KSA 100 was 100 watts into 8 ohms. Assuming the KSA 160 is 160 watts, here's a stab at some calculations:
160 watts/8 ohms is 36 Vrms or about 51 volts peak and 51 volts/8 ohms gives about 6.5 amps peak current. A push pull Class A amp has to idle at half the required peak current. So it would need 3.25 amps of bias current per channel. Even if it's not a Class-A amp, this is still a valid estimate of the average current it would need to produce full power.
You should measure the DC rail voltage on one rail and multiply by 2 (for both rails) then 3.25 amps to get the VA requirement per channel. Multiply that number by 2 (for a stereo) amp and multiply again by at least 1.25 to allow for losses and the fact you don't want to run the transformer right up to its rating.
It's easier to look at VA rather than just current. Computing actual AC currents drawn from the transformer is daunting due to the filter capacitor charging currents, etc. So most designers just work off the VA rating of the transformer. This works well because transformer cores come in fixed sizes with each having a given VA capacity regardless of the winding voltages.
So, I'm going to guess your amp probably has around +/- 55 volt rails but they might be higher or lower. So 55*2*3.25*2*1.25 gives 893 VA. So a 1000 VA transformer would be a good choice if it's a shared power supply (you said transformer singular). If the rails or bias current is lower than my estimates, than an 800VA transformer would probably be fine.
If you want an even more accurate estimate, you can find the idle current of your amp by measuring across the emitter resistors. You also could plug the amp into a Kill-A-Watt or similar power meter and measure its draw from the AC line at idle and playing music at loud levels. The meter will directly read the approximate VA load on the transformer. If it's a true class A amp, and you have easy to drive speakers, the meter won't change much between idle and playing loud.
One problem is off the shelf transformers usually come, at best, in increments of 5 volts--i.e. 35, 40, 45, etc. To get 55 volt rails you need 38.8 Vrms. So you're faced with buying either a 35 + 35 volt transformer or a 40 + 40 volt one. I'd suggest going for the next lower voltage if you're in between values. If you end up with higher DC rails you could stress capacitors, trasnsistors, etc.
The other problem is the factory transformer may well have extra windings. Each channel may have its own windings for a semi-dual-mono design. There can also be windings for other circuitry (i.e. protection circuits, etc.) and/or higher rails for the driver stages. You won't find such a transformer off the shelf.
So, if that's the case, you need to either buy the Krell transformer, have one custom made (which would probably cost even more), use multiple transformers (if there is space), live with the noise, or perhaps (if you're lucky) find a dead KSA 160 for cheap and use its transformer.
160 watts/8 ohms is 36 Vrms or about 51 volts peak and 51 volts/8 ohms gives about 6.5 amps peak current. A push pull Class A amp has to idle at half the required peak current. So it would need 3.25 amps of bias current per channel. Even if it's not a Class-A amp, this is still a valid estimate of the average current it would need to produce full power.
You should measure the DC rail voltage on one rail and multiply by 2 (for both rails) then 3.25 amps to get the VA requirement per channel. Multiply that number by 2 (for a stereo) amp and multiply again by at least 1.25 to allow for losses and the fact you don't want to run the transformer right up to its rating.
It's easier to look at VA rather than just current. Computing actual AC currents drawn from the transformer is daunting due to the filter capacitor charging currents, etc. So most designers just work off the VA rating of the transformer. This works well because transformer cores come in fixed sizes with each having a given VA capacity regardless of the winding voltages.
So, I'm going to guess your amp probably has around +/- 55 volt rails but they might be higher or lower. So 55*2*3.25*2*1.25 gives 893 VA. So a 1000 VA transformer would be a good choice if it's a shared power supply (you said transformer singular). If the rails or bias current is lower than my estimates, than an 800VA transformer would probably be fine.
If you want an even more accurate estimate, you can find the idle current of your amp by measuring across the emitter resistors. You also could plug the amp into a Kill-A-Watt or similar power meter and measure its draw from the AC line at idle and playing music at loud levels. The meter will directly read the approximate VA load on the transformer. If it's a true class A amp, and you have easy to drive speakers, the meter won't change much between idle and playing loud.
One problem is off the shelf transformers usually come, at best, in increments of 5 volts--i.e. 35, 40, 45, etc. To get 55 volt rails you need 38.8 Vrms. So you're faced with buying either a 35 + 35 volt transformer or a 40 + 40 volt one. I'd suggest going for the next lower voltage if you're in between values. If you end up with higher DC rails you could stress capacitors, trasnsistors, etc.
The other problem is the factory transformer may well have extra windings. Each channel may have its own windings for a semi-dual-mono design. There can also be windings for other circuitry (i.e. protection circuits, etc.) and/or higher rails for the driver stages. You won't find such a transformer off the shelf.
So, if that's the case, you need to either buy the Krell transformer, have one custom made (which would probably cost even more), use multiple transformers (if there is space), live with the noise, or perhaps (if you're lucky) find a dead KSA 160 for cheap and use its transformer.
try to ask by this transformer manufacturer
http://www.tauscher.com/html/contact.html
one of the best over the world. I think, not extremly expensive.
http://www.tauscher.com/html/contact.html
one of the best over the world. I think, not extremly expensive.
Arick said:
For that you have to identify what each secondary winding is used for. Most, except the main output windings, should be very low current. Did my post above at least help with the main windings? To get amps from VA you just divide by the transformer's AC voltage.
If it turns out you can get a one-of custom transformer made for less than Krell wants for a replacement, that doesn't reflect well on Krell's customer service. I would expect a custom 800VA or 1000VA transformer to be several hundred dollars minimum.
I'm not sure I under what you're asking by "how low"? Do you mean voltage? If you want the amp to run a bit cooler it would probably be OK to go a bit lower than the factory transformer for the output stages but I wouldn't go more than 10% - 20% lower. Frankly, I don't know enough about the design of the amp to be guessing with that sort of thing. Lowering the output stage voltage will also, of course, lower the maximum power of the amp by an even greater proportion (power being related to the square of voltage).
Ah. The gauge of wire they use is part of it, but the other part is how much it loads the primary. You don't want to waste VA where it's not needed. So even if the wire can handle say 1 amp, you can still specify 0.1 amps for a winding and that leaves more VA free for the output windings. With lots of windings physical space taken up by the windings can also be an issue. You should check with the supplier and also find out what standard core/primary VA sizes they have as well. From there you can spec the low current windings and leave the rest for the output windings up to the max VA of the transformer.
Hi,
if the KSA150 is a two channel ClassA amplifier, then Rocket has seriously underestimated the transformer VA rating required to power the output stages.
3.25A per supply lead for a two channel amplifier requires a continuous DC current of 6.5Adc from the PSU.
The minimum AC rating of the transformer must be at least double this i.e. Irms >=13Aac.
However, this will run the transformer at maximum temperature and will reduce it's life/reliability considerably. The exception here is if the transformer is designed for low regulation rather than to maximum temperature. I think Krell specified some/all of their transformers for low regulation, making temperature much less of an issue.
I would recommend doubling this again for continuous duty i.e. 26Aac for a two channel ClassA amplifier biased to 3.25Adc/channel. This will be BIG, HEAVY and EXPENSIVE.
Let's hope the KSA150 is a ClassAB amplifier.
if the KSA150 is a two channel ClassA amplifier, then Rocket has seriously underestimated the transformer VA rating required to power the output stages.
3.25A per supply lead for a two channel amplifier requires a continuous DC current of 6.5Adc from the PSU.
The minimum AC rating of the transformer must be at least double this i.e. Irms >=13Aac.
However, this will run the transformer at maximum temperature and will reduce it's life/reliability considerably. The exception here is if the transformer is designed for low regulation rather than to maximum temperature. I think Krell specified some/all of their transformers for low regulation, making temperature much less of an issue.
I would recommend doubling this again for continuous duty i.e. 26Aac for a two channel ClassA amplifier biased to 3.25Adc/channel. This will be BIG, HEAVY and EXPENSIVE.
Let's hope the KSA150 is a ClassAB amplifier.
Last edited:
KSA-150 Transformer specs? Dead - but not buried
Would you be willing to share the specs and a suitable manufacturer for the replacement transformer?
I also have a dead KSA-150 in need of a replacement transformer (among other things).
| know this thread has aged. I'm hoping you're still following|
Would you be willing to share the specs and a suitable manufacturer for the replacement transformer?
I also have a dead KSA-150 in need of a replacement transformer (among other things).
| know this thread has aged. I'm hoping you're still following|
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.